Rural–Urban Dynamics

It has increasingly been recognised that rural and urban areas are inextricably
interlinked. This book adopts a fresh approach to the issue of rural–urban
dynamics through a study of the changing nature of livelihoods, mobility and
markets in ten study sites across four countries of Africa and Asia. Building
on detailed fieldwork conducted in Ghana, Tanzania, Vietnam and Thailand,
the authors explore how settlements and livelihoods are being transformed as
long-term inhabitants and recent migrants embrace new economic activities,
many of which are linked to global markets.
The book is structured around the concept of ‘frontier’ which is concep-
tualised as being a dynamic space where the forces of economic, demographic
and social change are brought to bear. The study sites include agricultural
frontiers (coffee, cocoa, pineapples and fresh fruit), handicraft and manu-
facturing frontiers, and mining frontiers (gold and diamonds). In all of the
cases, global value chain dynamics have played a pivotal role in shaping local
livelihoods. Some settlements are developing into new urban centres whilst
others are suffering from a boom and bust experience due to the unreliability
of export markets. The similarities and differences between the frontier
settlements are drawn out by comparing frontiers of similar types and by
highlighting the theoretical and policy implications of the findings from all of
the frontiers.
The originality of the book lies in its combination of conceptual clarity,
methodological coherence and empirical richness. By combining detailed
empirical findings with theoretical insight from debates on livelihoods, global
value chains, mobility patterns, settlement dynamics and rural–urban rela-
tions, the book sheds new light on these issues within an overall framework of
development trajectories in Africa and Asia. Given scholars’ and inter-
national agencies’ current interest in the spatial dimensions of economic
development, this contribution is particularly timely with its fresh geograph-
ical approach to development issues. This book is a pertinent and authorita-
tive read for anyone researching or learning in the field of development.

Jytte Agergaard is an associate professor in the Department of Geography

and Geology at the University of Copenhagen. Her main research focus is on
mobility, migration and livelihoods. She has in particular focused on devel-
opment issues in Nepal, researching frontier development, migration, gender
inequalities and community-based schooling. Recently, she has been engaged
in development in Vietnam, and lived in Hanoi and Dak Lak Province for ten
months in 2005–2006. She has also been concerned with recent trends
in human geography, especially current trends in social geography and
geographies of globalisation. She is a member of a number of networks on
development research in Denmark and abroad.

Niels Fold is professor in the Department of Geography and Geology at the

University of Copenhagen. His current research interests revolve around the
dynamics of cross-continental food commodity chains and food trade in
the Asia-Pacific region. He has worked on globalisation and economic
restructuring in Africa, in particular on global agro-industrial chains
with relevance for perennial crops in Ghana (cocoa, oil palm, shea nuts).
Previously Niels was engaged in research on industrial development in
Malaysia, both on upgrading processes in the palm oil industry and on local
industrialisation trajectories (Penang).

Katherine V. Gough is a British-trained geographer who is employed as

associate professor in the Department of Geography and Geology at the
University of Copenhagen. Her main research focus is urbanisation in the
Global South with a particular interest in urban land and housing markets,
urban governance and civil society, home-based enterprises, urban youth and
urban mobility. Her principal research areas are Latin America and West
Africa though more recently she has started working in the Asia-Pacific
region. She has a particular interest in conducting comparative research and
longitudinal studies. Katherine works in an advisory capacity to the Danish
International Development Agency (DANIDA) on urban issues.
Routledge Studies in Human Geography

This series provides a forum for innovative, vibrant, and critical debate within
Human Geography. Titles will reflect the wealth of research which is taking
place in this diverse and ever-expanding field.
Contributions will be drawn from the main sub-disciplines and from
innovative areas of work which have no particular sub-disciplinary allegiances.

Published:
1. A Geography of Islands
Small island insularity
Stephen A. Royle

2. Citizenships, Contingency and the Countryside

Rights, culture, land and the environment
Gavin Parker

3. The Differentiated Countryside

Jonathan Murdoch, Philip Lowe, Neil Ward and Terry Marsden

4. The Human Geography of East Central Europe

David Turnock

5. Imagined Regional Communities

Integration and sovereignty in the global south
James D. Sidaway

6. Mapping Modernities
Geographies of Central and Eastern Europe 1920–2000
Alan Dingsdale

7. Rural Poverty
Marginalisation and exclusion in Britain and the United States
Paul Milbourne
 8. Poverty and the Third Way
Colin C. Williams and Jan Windebank

9. Ageing and Place

Edited by Gavin J. Andrews and David R. Phillips

10. Geographies of Commodity Chains

Edited by Alex Hughes and Suzanne Reimer

11. Queering Tourism

Paradoxical performances at Gay Pride parades
Lynda T. Johnston

12. Cross-Continental Food Chains

Edited by Niels Fold and Bill Pritchard

13. Private Cities

Edited by Georg Glasze, Chris Webster and Klaus Frantz

14. Global Geographies of Post Socialist Transition

Tassilo Herrschel

15. Urban Development in Post-Reform China

Fulong Wu, Jiang Xu and Anthony Gar-On Yeh

16. Rural Governance

International perspectives
Edited by Lynda Cheshire, Vaughan Higgins and Geoffrey Lawrence

17. Global Perspectives on Rural Childhood and Youth

Young rural lives
Edited by Ruth Panelli, Samantha Punch and Elsbeth Robson

18. World City Syndrome

Neoliberalism and inequality in Cape Town
David A. McDonald

19. Exploring Post Development

Aram Ziai

20. Family Farms

Harold Brookfield and Helen Parsons

21. China on the Move

Migration, the state, and the household
C. Cindy Fan
22. Participatory Action Research Approaches and Methods
Connecting people, participation and place
Sara Kindon, Rachel Pain and Mike Kesby

23. Time-Space Compression

Historical geographies
Barney Warf

24. Sensing Cities

Monica Degen

25. International Migration and Knowledge

Allan Williams and Vladimir Baláž

26. The Spatial Turn

Interdisciplinary perspectives
Barney Warf and Santa Arias

27. Whose Urban Renaissance?

An international comparison of urban regeneration policies
Libby Porter and Kate Shaw

28. Rethinking Maps

Martin Dodge, Rob Kitchin and Chris Perkins

29. Rural–Urban Dynamics

Livelihoods, mobility and markets in African and Asian frontiers
Edited by Jytte Agergaard, Niels Fold and Katherine V. Gough

Not yet published:

30. Design Economies and the Changing World Economy
Innovation, production and competitiveness
John Bryson and Grete Rustin

31. Critical Reflections on Regional Competitiveness

Gillian Bristow
Rural–Urban Dynamics
Livelihoods, mobility and markets in
African and Asian frontiers

Edited by Jytte Agergaard, Niels Fold and

Katherine V. Gough
First published 2010
by Routledge
2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN
Simultaneously published in the USA and Canada
by Routledge
270 Madison Ave, New York, NY 10016
Routledge is an imprint of the Taylor & Francis Group,
an informa business
This edition published in the Taylor & Francis e-Library, 2009.
To purchase your own copy of this or any of Taylor & Francis or Routledge’s
collection of thousands of eBooks please go to www.eBookstore.tandf.co.uk.

© 2010 Selection and editorial matter: Jytte Agergaard, Niels Fold and
Katherine V. Gough; individual chapters the contributors
All rights reserved. No part of this book may be reprinted or
reproduced or utilised in any form or by any electronic,
mechanical, or other means, now known or hereafter
invented, including photocopying and recording, or in any
information storage or retrieval system, without permission in
writing from the publishers.
British Library Cataloguing in Publication Data
A catalogue record for this book is available from the British Library
Library of Congress Cataloging-in-Publication Data
Rural–urban dynamics: livelihoods, mobility and markets in African
and Asian frontiers / [edited by] Jytte Agergaard, Niels Fold and
Katherine V. Gough.
p. cm. — (Routledge studies in human geography)
Includes bibliographical references and index.
1. Rural–urban relations—Africa. 2. Rural–urban relations—Asia.
I. Agergaard, Jytte, 1960– II. Fold, Niels, 1955– III. Gough,
Katherine V., 1963–
HT384.A35R87 2009
307.74—dc22 2009003521

ISBN 0-203-87394-7 Master e-book ISBN

ISBN10: 0–415–47562–7 (hbk)

ISBN10: 0–203–87394–7 (ebk)

ISBN13: 978–0–415–47562–4 (hbk)

ISBN13: 978–0–203–87394–6 (ebk)
Contents

List of illustrations xi
List of figures xii
List of tables xiv
Notes on contributors xvi
Acknowledgements xix

1 Introduction 1
JYTTE AGERGAARD, NIELS FOLD AND KATHERINE V. GOUGH

2 Conceptualising and performing comparison of

rural–urban dynamics 9
KATHERINE V. GOUGH, JYTTE AGERGAARD, NIELS FOLD AND
LASSE M ø LLER-JENSEN

3 Small town development in the Ghanaian cocoa frontier 23

MICHAEL HELT KNUDSEN

4 Living from coffee in Vietnam’s Central Highlands:

Susceptible livelihoods and diverse
settlement transformations 41
JYTTE AGERGAARD

5 Rise and fall of smallholder pineapple production in Ghana:

Changing global markets, livelihoods and
settlement growth 59
KATHERINE V. GOUGH AND NIELS FOLD

6 Transforming livelihoods and settlements: Fruit

production and urbanisation in Vietnam’s Mekong Delta 74
HOANG XUAN THANH AND CECILIA TACOLI
x Contents
7 Agricultural frontier settlements: Markets,
livelihood diversification and small town development 90
NIELS FOLD AND CECILIA TACOLI

8 Handicrafts in Vietnam: Rural urbanisation in the

Red River Delta 99
KATHERINE V. GOUGH AND DANG NGUYEN ANH

9 A revolution in the Thai handicraft industry?

Handicrafts, integration and rural development in
Northern Thailand 116
JONATHAN RIGG, SURIYA VEERAVONGS,
PIYAWADEE ROHITARACHOON AND LALIDA VEERAVONGS

10 From farm to factory: Village change in a

rice-growing region 134
JONATHAN RIGG, SURIYA VEERAVONGS,
PIYAWADEE ROHITARACHOON AND LALIDA VEERAVONGS

11 Handicraft and manufacturing frontier:

Transforming livelihoods, re-shaping settlements 151
KATHERINE V. GOUGH AND JONATHAN RIGG

12 Rural–urban transitions in Tanzania’s northwest

mining frontier 158
DEBORAH FAHY BRYCESON AND ROSEMARIE MWAIPOPO

13 The dynamics of the gold mining industry and its

effects on settlements and livelihoods in Wassa
West District, Ghana 175
PAUL W.K. YANKSON

14 Frontier mining settlements: Livelihood promises

and predicaments 189
DEBORAH FAHY BRYCESON AND PAUL W.K. YANKSON

15 Conclusion 198
NIELS FOLD, KATHERINE V. GOUGH AND JYTTE AGERGAARD

Index 206
Illustrations

3.1 Labourers opening cocoa pods in Amoaya 31

3.2 The main street running through Bonsu Nkwanta where
settlement growth has taken place since the late 1990s 35
4.1 Typical coffee settler’s house, located close to the
smallholder’s coffee-growing land 45
4.2 The northern entrance to the provincial capital Buon Ma
Thout, which is gradually expanding to include rural coffee
settlements on the outskirts 49
5.1 Freshly harvested pineapples lie by the roadside ready to be
packed into containers and loaded into a waiting truck 63
5.2 The street running through the centre of Obom where most
of the stores are located 67
6.1 The Mekong Delta’s extensive waterways system links
orchards to market nodes, from where produce is shipped to
domestic and international markets 77
6.2 Agricultural markets drive the transformation of market
nodes in the Mekong Delta from large villages to centres with
more urban features 87
8.1 An elderly woman engaged in embroidery in her home in
Quat Dong 104
8.2 Workers engaged in lacquerware production in a workshop
in Ha Thai 107
8.3 Wealth generated by lacquerware production has enabled
many households to build new homes in Ha Thai but the ponds
and surrounding agricultural land have been polluted 112
9.1 Surrounding agricultural landscape of Ban Mor 120
9.2 Sanding mango wood in a workshop in Ban Mor 123
9.3 Saa paper ‘factory’ 126
10.1 Dormitory block in Ban Khokmayom 141
10.2 Ban Khokmayom and Rojana Industrial Park 142
12.1 Gold production in Nyarugusu 161
12.2 Diamond production in Manganzo 163
Figures

3.1 Map showing location of study settlements in

Juaboso District 27
4.1 Map showing location of study communes in
Dak Lak Province 46
4.2 Head of household’s year of arrival in Dak Lak Province
in cumulative per centage of total number of households 48
5.1 Map showing location of study settlements in the
pineapple frontier 65
6.1 Map showing location of study settlements in the
Mekong River Delta 79
6.2 Typical market chains of Lo Ren milk fruit, Hoa Loc
mango and Nam Roi pomelo 86
8.1 Map showing location of study settlements in the
Red River Delta 102
9.1 Map showing location of study settlements in
Northern Thailand 118
9.2 Sum of primary, secondary and tertiary activities,
by selected occupation, Ban Mor, 1995 and 2005 121
9.3 Sum of primary, secondary and tertiary activities,
by selected occupation, Ban Pasakluang, 1995 and 2005 122
10.1 Map showing location of study settlements in
Central Thailand 136
10.2 Age and occupational status of adult household
members: Ban Khan Haam and Ban Khokmayom, 2005 143
10.3 Formal/skilled and casual/unskilled occupations among
working-age household members: Ban Khan Haam
and Ban Khokmayom, 2005 143
10.4 Daily mobility and purpose on the day prior to the survey:
Ban Khan Haam and Ban Khokmayom, 2005 146
12.1 Map showing location of study settlements in
northwestern Tanzania 162
12.2 Migrant household heads’ year of arrival 164
12.3 Household heads’ livelihood activities 167
 Figures xiii
12.4 Size of household land holdings 168
12.5 Percentage of total population aged 15–34 in 1978,
1988 and 2002 169
13.1 Location of Teberebie gold mine in Wassa West District 178
Tables

1.1 Overview of the nine frontier cases in relation to the

frontier typology 5
2.1 Summary of fieldwork stages 18
2.2 Data collection in the frontier settlements 19
3.1 Population data on the four study settlements 28
3.2 Livelihood characteristics of the four study settlements 29
4.1 Background statistics for the four survey communes,
2004 47
4.2 Background statistics for the survey villages 47
4.3 Diversity in housing standards in the four survey
communes 49
4.4 The significance of households’ engagement in coffee
farming 52
4.5 Patterns in income diversity 1996 and 2006: number
of households acquiring their income from various
income activities 53
4.6 Percentage of households possessing selected means of
transportation 55
5.1 Ethnic groups of the inhabitants of Pokrom and Obom 66
5.2 Employment sector for workers in Pokrom and Obom 69
6.1 The two study provinces: essential information 78
6.2 Structure of respondents’ average household incomes in
2005 82
6.3 Long-term mobility 84
8.1 Participation in income-generating activities in Quat
Dong and Ha Thai by gender 108
8.2 Attractiveness of handicraft to young people in Quat
Dong and Ha Thai 109
8.3 Size of houses in Quat Dong and Ha Thai 110
8.4 Possession of selected household goods in Quat
Dong and Ha Thai 113
9.1 Summary statistics: Ban Mor and Ban Pasakluang 122
9.2 Age and occupational status, by proportion of respondents
in each age category, 2005 129
 Tables xv
10.1 Ban Khan Haam and Ban Khokmayom, background
statistics 135
10.2 Mobility in Ban Khokmayom and Ban Khan Haam 145
12.1 Comparative household demographic statistics by
settlement 164
12.2 Stated primary and secondary occupations of household
heads 166
13.1 Major economic activities of households in Tarkwa in
1995 and 2005 181
13.2 Proportion of income from household economic activities
generating most income in Tarkwa in 1995 and 2005 181
13.3 Proportion of income from household economic activities
generating most income in Teberebie and Adieye in 1995
and 2005 182
Contributors

Deborah Fahy Bryceson is a reader in urban studies at the Geographical and

Earth Sciences Department at the University of Glasgow. She has been
engaged in academic as well as policy-oriented research in Africa for many
years. One of her main research areas is livelihood, mobility and settle-
ment in processes of economic transition. This has included studies on
de-agrarianisation, occupational change, rural transport, urbanisation in
developing countries, urban economies, urban growth and mobility pat-
terns of the poor, and urban food supply constraints. She was involved in
coordinating a large collaborative programme on de-agrarianisation and
rural livelihoods with research teams in various African countries funded
by the Dutch Ministry of Foreign Affairs. More recently she has
researched the social history of urban violence in East Africa’s major
capital cities.
Dang Nguyen Anh is senior researcher and programme manager for Popula-
tion and Development at the Asia-Pacific Economic Center, Hanoi. He
has over 20 years’ experience in internal and international migration issues
in Vietnam and Southeast Asia, and has advised international agencies
(ILO, UNDP) and bilateral agencies (DFID) on these topics. Since 2000
he has been collaborating with IIED in an on-going project on rural–
urban linkages in Vietnam, supported by Danida, Sida and SDC.
Michael Helt Knudsen has an MSc and PhD in geography from the
Department of Geography and Geology, University of Copenhagen. His
main research focus is the livelihoods, mobility and settlement dynamics
taking place in the Global South. His principal research area is Ghana,
West Africa. He is currently employed by a consulting company dealing
with global climate change issues and their impact on livelihoods and
ecosystems around the globe.
Lasse Møller-Jensen is an associate professor at the Department of
Geography and Geology, University of Copenhagen. He is an expert in
remote sensing and Geographical Information Systems who has specialised
in urban applications of remote sensing techniques with special reference
to developing countries. Lasse implements dynamic spatial models in
 Contributors xvii
raster- and vector-based Geographical Information Systems, and has
developed a location-allocation model for service planning with particular
reference to Ghana. He also has a special interest in the web-based
distribution of geo-data.
Rosemarie Mwaipopo (Sociology Department, University of Dar es Salaam)
has done extensive research on small-scale mining communities in Tanza-
nia, documenting the growth and changing nature of gold- and diamond-
digging communities. She is interested in the conflicts arising at the
interface between large- and small-scale mining. She did her Masters at
the University of Dar es Salaam and her PhD in anthropology at the
University of Cape Town.
Jonathan Rigg is a professor in the Department of Geography at Durham
University and has been working on development issues in Southeast Asia
for a quarter of a century. His research focuses on the ways in which
trajectories of change in the region have reworked livelihoods, reoriented
spatialities of living, and redrawn the boundaries between rural and
urban. Jonathan is interested in the trade-offs and interactions
between these changes, encompassing such issues as human mobility, de-
agrarianisation, and the delocalisation of work. The social and cultural
roots and outcomes of economic processes, and their spatial manifest-
ations, are also central to his research.
Piyawadee Rohitarachoon is a lecturer at Naresuan University, Phayao,
Thailand and was formerly a researcher at Chulalongkorn University’s
Social Research Institute. Her interests lie in how social and economic
transformation are affecting quality of life, and the politics of development.
Cecilia Tacoli is a senior researcher at the International Institute for
Environment and Development, based in London. She coordinates a long-
term programme of collaborative work with partners in Asia and Africa
that documents processes of urbanisation and the related transformations
in the livelihoods of the rural and urban poor, with a focus on policy
implications.
Hoang Xuan Thanh is the director of Ageless, a consultancy firm based in
Hanoi which specialises in rural and urban poverty reduction policy
analysis and research. Since 1992, he has collaborated with international
NGOs, international and bilateral donor agencies and the government of
Vietnam in developing participatory approaches and building capacity in
both public and private sectors. He is co-author of a 2004 research project
on rural–urban linkages and livelihood transformations in the Red River
Delta, in collaboration with IIED, and is currently working with Oxfam
and ActionAid Vietnam on developing a methodology for the monitoring
of poverty in several Vietnamese provinces and cities. He is a member of
the Vietnam Economics Association.
xviii Contributors
Lalida Veeravongs is a researcher at the Chulalongkorn University Social
Research Institute (CUSRI), Bangkok, Thailand. Her work has
encompassed projects investigating livelihood and political change in the
North, Northeastern and Central regions of the country.
Suriya Veeravongs is a senior researcher at the Chulalongkorn University
Social Research Institute (CUSRI), Bangkok, Thailand. She has been
researching the rural and urban dimensions of social and cultural change
in Thailand since the 1970s, and has worked with numerous international
organisations and national agencies.
Paul W.K. Yankson is professor and Director of the Remote Sensing
Applications Laboratory at the Department of Geography and Resource
Development, University of Ghana. His research has focused on urban
and regional development issues, particularly on urbanisation in Ghana,
the urban informal sector, urban and peri-urban land management and
governance, local government reforms and regional development, resource
management, remote sensing and GIS application in urban studies. He has
undertaken consultancy services for both national and international
institutions such as the UNDP, World Bank, and ILO/JASPA (Jobs and
Skills Programme for Africa). He has served on a number of national
committees including the National Inter-sectoral Working Group on
Proposed Urban Development Programme.
Acknowledgements

In the kind of empirically based research presented in this book, greatest

thanks are due to the very many respondents in the frontiers studied. Without
their patience in answering our numerous questions the book would not exist.
The research programme was funded by the Danish Ministry of Foreign
Affairs’ Consultative Research Committee for Development Research (grant
no. 91206). Along the way, the programme benefited from constructive
discussions held with an advisory board consisting of Erik Bryld, Poul Ove
Pedersen, Jonathan Baker, Jens Lorentzen, and Franz-Michael Rundquist.
In Ghana, the assistance of Mr Leo Laryea with the fieldwork in the
pineapple frontier is gratefully acknowledged; as always he competently led a
small team of research assistants and made even the most hesitant informant
feel at ease. Dr Agnes Budu is thanked for stimulating discussions and logis-
tical assistance. In the cocoa frontier, Nana Bawa proved to be an invaluable
contact in Bonsu Nkwanta, and the assistance of Solomon Dodoee and
Sandow Mumuni in the field is gratefully acknowledged. In Juaboso District,
the District Planning Officer Mr Issac B. Ankomah and the District Chief
Executive Mr Asiedu M. Abrampah were both extremely supportive of the
study. In the mining frontier, the assistance of the staff of the Wassa West
District Assembly, in particular, Mr Timothy Nankara, the young Planning
Officer who unfortunately passed away before the end of the project, is deeply
appreciated. The role played by the team of field assistants led by Florence
Koomson and including Franklin Owusu, Emmanuel Amissah and Isaac
Aikens, Gilbert Bart-Plange, Angelo Turkson and Hagar Korankye-Appiah,
in ensuring the success of the various phases of the research is graciously
acknowledged.
Turning to Vietnam, in relation to the Mekong Delta frontier, we would
like to thank Dinh Thi Thu Phuong and Nguyen Thu Huong for their help in
the field and in collecting and analysing secondary data. Dr Nguyen Minh
Chau (director) and staff at the Southern Fruit Research Institute (SOFRI)
in My Tho also provided practical help and guidance and generously shared
their in-depth knowledge of fruit production and markets. In the coffee
frontier, access to the field was organised through the Dak Lak Department
of Agriculture and Rural Development under the most appreciated guidance
xx Acknowledgements
of the Director of the Agro-forestry Department, Dr Tran Ngoc Thanh. The
assistance of Mr Linh and Mr Hung in the field is also gratefully acknow-
ledged. In the handicraft frontier, we are particularly grateful for the support
of VAPEC Hanoi and especially the director Professor Dr Le Van Sang. In
the field, the competent research team was led by Ms Nghiem Thi Thuy and
Dr Le Kim Sa. The assistance of Mr Pham Viet Dinh in Quat Dong and Mrs
Nguyen Thi Hoi in Ha Thai is gratefully acknowledged.
The work in Thailand has benefited from, and been partially supported by,
a Canadian Social Sciences and Humanities Research Council (SSHRC)
grant on ‘The Challenges of the Agrarian Transition in Southeast Asia’ (see:
http://www.caac.umontreal.ca/en/chatsea_intro.html). We would also like to
acknowledge the assistance of Burghard Rauschelbach, Director of the
Programme for Enterprise Competitiveness/Eco-Efficiency, Deutsche
Gesellschaft fuer Technische Zusammenarbeit (GTZ) GmbH.
Chapter 12 on the mining frontier in Tanzania is dedicated the memory of
Professor Faustin Ndalawa Madulu of the Institute of Resource Assessment
at the University of Dar es Salaam who passed away on 2 July 2007. Profes-
sor Madulu coordinated the household survey with the assistance of another
demographer S. Ngallaba, formerly of the Tanzanian Central Statistical
Bureau, and a team of four MA student enumerators from the University of
Dar es Salaam: Dina Enock, Musa Magufuli, Rose Mbwete, and Floren
Silangwa. In addition we are grateful to local teachers who helped familiarise
the team with the various neighbourhoods within the settlement. Professor
Madulu’s early recognition of the importance of mining to the local rural
economy during the 1990s was the catalyst for this study.
At Routledge, Michael P. Jones has patiently guided us through the
publishing process. Britta Munch and Kent Pørksen of the Department and
Geography and Geology, University of Copenhagen kindly assisted with the
figures and maps.
1 Introduction
Jytte Agergaard, Niels Fold and
Katherine V. Gough

For decades, scholars have been interested in rural–urban relations, rural–

urban interactions, and rural–urban dynamics. Work on themes such as urban
peasants, the role of intermediate urban centres, and urban bias in develop-
ment, are all ‘old’ concerns. These themes, however, need to be re-examined in
the light of new and intensifying globalisation processes. Lives and liveli-
hoods are being profoundly re-worked as processes of market integration,
modernisation and industrialisation are brought to bear in new and more
intense ways. The functional integration of economic activities across regions,
together with new trading regimes, is manifested in new international divisions
of labour that link people and places in the Global South to global markets in
new ways. Peasants are becoming post-peasants; households are being divided
by generation and gender; livelihoods are becoming increasingly delocalised;
interlocking livelihoods and occupational multiplicity are displacing more
singular ways of making a living; and de-agrarianisation is replacing the
more familiar process of agrarian transition.
This book is concerned with how changes in local livelihoods in relation
to economic globalisation are articulated as spatial transformations; in par-
ticular how the rural and the urban are taking new forms. Through a study
of nine ‘frontier’ sites in Africa and Asia we explore the multi-faceted and
changing nature of the links between livelihoods, mobility and markets. We
conceptualise a ‘frontier’ as being a particularly dynamic space where the
forces of economic, demographic and social change are brought to bear.
These include agricultural frontiers, handicraft and manufacturing frontiers,
and mining frontiers where global market dynamics have played a pivotal
role in shaping development trajectories. By combining detailed empirical
findings with theoretical insights from debates on rural–urban linkages,
global value chains, livelihoods and mobility patterns, we throw new light on
these issues within an overall framework of development trajectories in
Africa and Asia.
This book is the product of the efforts of a multi-national team of scholars,
mainly geographers, who have worked together over four years to elucidate
these issues. The research programme, and the book, have been guided by a
number of overall questions:
2 Jytte Agergaard, Niels Fold and Katherine V. Gough
• Focusing on rural–urban dynamics, how do particular trajectories of
market integration impact on livelihood, mobility and settlement trans-
formation, and what similarities and differences can be drawn by
comparing similar frontiers?
• Focusing on the national level, what is the significance of frontier dynam-
ics, how do they impact on national development trajectories, and in this
respect, what are the methodological challenges involved in scaling up
the findings of the frontier settlement studies?
• What lessons can be drawn from cross-continental comparisons of rural–
urban dynamics in African and Asian countries, and what factors are
behind differences in settlement transformation?
• What analytical advances concerning interactions between global market
dynamics, livelihoods and mobility have been uncovered, and in this
respect, what intrinsic methodological limitations of the frontier approach
should be considered?
• What are the policy implications of the findings, in particular regarding
how to design policy mechanisms that both capture the complexity of
rural–urban dynamics and are directed towards improving the livelihoods
of marginalised groups and promoting regional economic development?

Frontier development
A straightforward definition of a frontier is provided by Mikesell (1960: 62)
who characterises a frontier as ‘the outer edge of settlement within a given
area’. The frontier, however, is usually represented by an image of resource
abundance. Although invasion and settlement of uncultivated areas have
taken place throughout human history, the role of frontiers in development
stems from the frontier thesis on American development put forward by
Fredrick Jackson Turner in 1893 (Barbier, 2005). Settlers’ ability to tame the
virgin environment, he argued, fostered economic equality, created strong
individualists who could cooperate, and encouraged democracy and liberty
as guiding principles – all characteristics of the American ‘national’ char-
acter. Turner portrayed pioneers as self-made men who tamed and civilised
the ‘wild west’ independent of any guidance or support from the Old World
hence creating a New World. Frontier development is generally characterised
as being the inclusion of ‘unexplored regions’ or ‘undeveloped space’ into the
national economy (Cleary, 1993).
Turner’s ideas have been hotly debated. One fundamental criticism is that
the thesis builds more on romanticism and nationalism than objective facts
and rests on a naïve expression of environmental determinism (Mikesell,
1960; DeKoninck, 2000). Turner also overlooked how different population
groups interact while creating frontier settlements and how their different
social and cultural ideals influence the nature of the frontier. Furthermore,
free land is just one manifestation of a frontier. The role of petty capitalism
and urbanisation may be as important for understanding the political and
 Introduction 3
economic dynamics of frontiers. It has been argued that the following issues
are central to understanding frontier dynamics: (1) the relationship between
immigrants and indigenous people; (2) whether colonisation is, or is not,
checked by environmental, technological, or human barriers; (3) the social
and economic conditions prevalent on the frontier; (4) the nature of mobility
and urbanisation within the region (Mikesell, 1960; Cleary, 1993).
Modern frontier development tends to involve a state strategy (Barbier,
2005). In order to deal with immediate economic and political problems,
states have eagerly supported the opening up of frontiers and thus secured
land and incomes for smallholders. This strategy is often called a ‘demo-
graphic safety valve’ because pressures on land and resources in densely
populated and resource-scarce regions are removed by providing new oppor-
tunities at the frontier (Shrestha, 1990; Cleary, 1993). While often accom-
panied by planned settlement, and thus ‘support’ for pioneers, the long-term
strategy of frontier settlement is also to secure the state and dominating
classes access to surplus resource extraction (Shrestha, 1990; De Koninck,
2000). However, continued frontier development cannot be assumed as the
engine behind it may lose momentum, investments may stall, and settlers may
have to revise their ideas and plans. Building on this observation, Barbier
(2005) has argued that, while frontier expansion will inevitably open up the
local economy and lead to a boom, the boom will be followed by a bust even
if commodity prices and trade remain unchanged. Recently, this simplified
boom-and-bust approach has been modified by Ha and Shively (2008),
who argue that socio-economic groups of smallholders may respond to price
fluctuations differently according to their ethnicity and access to resources. In
the same vein, it has been argued that socio-spatial variation within the fron-
tier region may also impact on smallholders’ responses to price fluctuations
on the world market (Agergaard et al., 2009).

Frontier regions, settlement and cases

Drawing on the traditional frontier concept, and the related critique pre-
sented above, we expand the conceptualisation of frontier regions in order to
construct an analytical space in which intense global–local interactions take
place. The frontiers selected are characterised by being heavily linked to world
market dynamics through intensified exports of a particular commodity
(coffee, cocoa, gold) or product category (handicraft, fresh fruit). These fron-
tier regions are characterised by rapid changes in demographic structure,
economic basis, occupational possibilities, and land use. Consequently they
experience high rates of immigration and changing livelihood opportunities,
such as through the establishment of new commercial activities. Previously
stagnating economic activities are being replaced by new activities, pre-
dominantly the production of export-oriented commodities or processed
goods which are experiencing increasing demand on the global market. Thus,
the term ‘frontier’ is not only used to denote areas that are advancing in a
4 Jytte Agergaard, Niels Fold and Katherine V. Gough
spatial sense, but also includes areas that are experiencing social and eco-
nomic fluidity due to new opportunities.
Associated with the influx of people into these frontier regions are new
settlement processes and patterns. Existing settlements are consolidated and
thus become increasingly densely populated, as well as expanding on their
fringes. In some cases new settlements are established to house the migrants.
Some of these settlement patterns defy mainstream views of the rural–urban
dichotomy as they are difficult to characterise as either rural or urban. Others
consist of overgrown villages or small towns that have experienced rapid and
spontaneous expansion without developing the physical and functional infra-
structure or organisational capacity linked to urbanisation. Some settlements
may in time acquire the economic strength to transform themselves into
regional growth centres and thus reduce the migration pressure on major cities.
The term ‘settlement’ thus connotes a diversity of forms of human habitation
along the rural–urban continuum. Accordingly, settlements located within
these frontier regions provide an ideal location in which to explore the role of
rural–urban links within livelihood strategies, from the perspective of both the
‘winners’ and ‘losers’ in society. Central to such an analysis is gaining a histor-
ical understanding of the origins of rural–urban linkages and how they have
been shaped by individuals and institutions over time.
The book adopts a comparative approach to rural–urban dynamics and
explores the similarities and differences in how they impact on livelihoods,
mobility and settlement in two continents where the changes taking place
are strikingly different. In Africa, as the manufacturing sector generally offers
limited income-generating opportunities, most diversification of income is
confined to the service sector or to extractive enterprises like small-scale min-
ing. In Southeast Asia, the expansion of the industrial and manufacturing sec-
tor offers a wider range of possibilities for productive employment, although
this tends to vary depending on national and regional characteristics. The
selection of four countries, namely Ghana, Vietnam, Tanzania and Thailand,
reflects the authors’ long-standing research experience in these countries, and
each frontier project was selected in order to generate new knowledge regard-
ing rural–urban dynamics in their respective contexts (see Table 1.1 below for
an overview of the frontiers).
Ghana has a long history of integration in the world market through gold,
timber and cocoa, and the structural composition of exports has changed
very little over the years, even though some industrial development has
occurred in the major cities. Three cases were selected: (1) settlements in
Ghana’s ‘last cocoa frontier’ in the Western Region; (2) settlements in the
gold mining areas of Wassa West District, part of Ghana’s rich Ashanti gold
ore area; and (3) settlements dominated by pineapple production for the
world market located in the original cocoa frontier region north of Ghana’s
capital, Accra.
Vietnam is experiencing remarkable economic development, with booming
exports of both labour-intensive manufactured goods and agricultural
 Introduction 5

Table 1.1 Overview of the nine frontier cases in relation to the frontier typology

Frontier typology Country

Ghana Vietnam Tanzania Thailand

Agricultural Cocoa Coffee

frontiers Pineapple Fresh fruit
Handicraft and Embroidery Mango wood-crafts
manufacturing and lacquer- and Saa paper
frontiers ware ‘Rural’ based
industrial estate
Mining frontiers Gold Gold and
diamond

commodities, including rice, the country’s main food crop. The economy has
been undergoing a comprehensive process of liberalisation since the mid
1980s, with the dismantling of state-owned enterprises and the steady reduc-
tion of central planning. Three cases were selected: (1) settlements dominated
by coffee production in Vietnam’s main coffee-producing province, namely
Dak Lak Province in the Central Highlands; (2) settlements characterised by
a transformation from food-crop production (predominantly rice) to fresh
fruit production for export in the Mekong River Delta; and (3) settlements
traditionally producing home-based handicrafts located in the Red River
Delta close to the capital of Hanoi that are incorporated into a global market
for handicrafts.
Tanzania is going through a comprehensive restructuring process of its
commercial base as conditions for peasant agricultural production are
seriously deteriorating. Smallholder households have increasingly diversified
into mining and trade in order to secure the necessary cash income. One case
was selected: settlements characterised by artisanal miners and small-scale
miners (gold and diamond) to some extent related to large-scale foreign com-
panies located in the northern regions of Tanzania.
Thailand has for many years enjoyed a prominent position in the global
market as an exporter of labour-intensive manufactured goods, in addition to
having an impressive portfolio of agricultural export commodities. However,
the concentration of industrial production in the Bangkok metropolitan area
has resulted in a trend towards the relocation of manufacturing production in
the rural areas. Two cases were selected: (1) settlements located in Chiang Mai
Province in northern Thailand producing home-based and semi-industrial
handicrafts of mango wood and Saa paper for a global market; and (2) rural
settlements located in proximity to an industrial estate in Ayuthaya Province
in central Thailand, which is also attracting workers from other rural
settlements.
6 Jytte Agergaard, Niels Fold and Katherine V. Gough
Structure and organisation of the book
The dynamic livelihood and settlement transformations in the nine frontier
regions are presented and analysed in separate chapters grouped according
to frontier type. Following each frontier type, a further chapter explores the
similarities and differences between the findings from the frontiers drawing
out generalisations where possible. These are then pulled together in the
concluding chapter.
In Chapter 2, following this introduction, the key concepts that run
throughout the book are presented, namely rural–urban dynamics, markets,
livelihoods and mobility. The chapter also introduces the methodological
approaches adopted in the collection and analysis of the data.
Chapter 3 by Michael Helt Knudsen, which is the first of the agricultural
frontiers, is concerned with why some cocoa frontier settlements grow into
small towns, while others stagnate and hardly change in size and function. The
focus of Chapter 4, by Jytte Agergaard, is how settlement transformations
relate to the changes in the coffee sector and how these transformations, and
the dramatic collapse of coffee prices in the late 1990s and early 2000s, have
impacted on livelihoods in the coffee settlements. In Chapter 5, Katherine V.
Gough and Niels Fold explore changes in livelihoods and settlements located
on the Ghanaian pineapple frontier. Here the pineapple sector is undergoing
a rapid transformation due to changes in the global demand for pineapples in
favour of a new variety, MD2. The last agricultural frontier chapter, Chapter
6 by Hoang Xuan Thanh and Cecilia Tacoli, portrays the multiple impacts of
rapid urbanisation and industrialisation on rural settlements that, in recent
years, have changed their main source of income from rice to fresh fruit. The
authors are particularly concerned to show how the expansion of activities
related to the fruit trade has opened up new opportunities for all groups while
at the same time widening income inequalities.
Chapter 7 by Niels Fold and Cecilia Tacoli explores how markets, liveli-
hood diversification and small-town development interact in the agricultural
frontiers studied. The authors illustrate how the dynamics of agricultural
value chains and the urban markets linked to them impact on livelihood
opportunities and affect the resilience of rural settlements. Markets vary
with the product and associated global value chains and have a differential
impact on the economic dynamics of the settlements. Forward and backward
linkages with the production of the major crops relate to processes of local
economic diversification that cover both farm and non-farm activities. Chan-
ging livelihood opportunities in the settlements increasingly involve some
form of mobility and in many cases result in growing social differentiation
and inequality.
Chapter 8 by Katherine V. Gough and Dang Nguyen Anh is the first of the
handicraft and manufacturing frontier chapters. Focusing on how handicraft
production in the Red River Delta is increasingly becoming directed towards
the global market and leading to rural urbanisation, the authors show how
 Introduction 7
the importance of handicrafts to livelihoods varies according to the nature of
the craft and is leading to both increasing wealth and greater inequalities. In
Chapter 9, Jonathan Rigg et al. are concerned with the impact of handicraft
production on rural development in northern Thailand. They illustrate the
changing nature of handicraft work in the study villages, the greater ease of
mobility and the changing priorities of villagers. The change from farm to
factory in Ayuthaya Province, Thailand is the focus of Chapter 10, also by
Jonathan Rigg et al. The chapter is particularly concerned with the remaking
of Thai rural economies and with how rural life is becoming dissociated from
farming.
Chapter 11 by Katherine V. Gough and Jonathan Rigg reflects on the wider
issues of handicraft and manufacturing production located in rural areas.
They explore the degree to which the notions of ‘rural’ and ‘urban’, and the
activities, lives and lifestyles that characterise these spaces, are being unsettled
by the transformations under way in the study sites. This leads the authors to
question whether farm–nonfarm relations generate a virtuous cycle and
indicate the importance of how development interventions are structured and
targeted.
In Chapter 12, Deborah Fahy Bryceson and Rosemarie Mwaipopo trace
mining-led urbanisation in a long-established diamond-mining settlement
and a younger gold-mining settlement in northwestern Tanzania. They show
how, even though mining may be declining as a livelihood possibility, it is a
highly dynamic vehicle of in situ urbanisation. The effects of the revival of
the gold-mining industry in Wassa West District in Ghana are the focus of
Chapter 13 by Paul W.K. Yankson. He shows how despite mining being a
dynamic economic sector, it generates little employment for local inhabitants
whose livelihoods are characterised by unemployment and poverty, partly
due to loss of land.
In Chapter 14, Deborah Fahy Bryceson and Paul W.K. Yankson reflect on
livelihood potentials and dilemmas in frontier mining settlements. They review
the position of mining in the national economies of Ghana and Tanzania,
outlining some of the salient differences between small-scale and large-scale
gold-mining, and contrasting small-scale gold-mining to diamond-mining.
Weighing the differences and conflicting tendencies of mining development in
frontier regions, the authors suggest that small-scale mining can nonetheless
ultimately lead towards sustainable urban settlement.
The book is rounded off with a concluding chapter that seeks to pull
together the findings from the different frontiers in order to answer the overall
questions that guided the research.

References
Agergaard, J., Fold, N. and Gough, K.V. (2009) ‘Global–local interactions: Socio-
economic and spatial dynamics in Vietnam’s coffee frontier’, Geographical Journal,
172(2): 133–145
8 Jytte Agergaard, Niels Fold and Katherine V. Gough
Barbier, E.B. (2005) ‘Frontier expansion and economic development’, Contemporary
Economic Policy, 23(2): 286–303.
Cleary, D. (1993) ‘After the frontier: Problems with political economy in the modern
Brazilian Amazon’, Journal of Latin American Studies, 25(2): 331–49.
De Koninck, R. (2000) ‘The theory and practice of frontier development: Vietnam’s
contribution’, Asia Pacific Viewpoint, 41(1): 7–22.
Ha, D.T. and Shively, G. (2008) ‘Coffee boom, coffee bust and smallholder response in
Vietnam’s Central Highlands’, Review of Development Economics, 12(2) 312–26.
Mikesell, M.W. (1960) ‘Comparative studies in frontier history’, Annals of the Associ-
ation of American Geographers, 50(1): 62–74.
Shrestha, N.R. (1990) Landlessness and migration in Nepal, Boulder, San Francisco
and Oxford: Westview Press.
2 Conceptualising and
performing comparison of
rural–urban dynamics
Katherine V. Gough, Jytte Agergaard,
Niels Fold and Lasse Møller-Jensen

The demographic, social and economic transformations occurring in most

parts of Africa and Asia are based on a complementary relationship between
rural and urban areas. Rather than clear-cut linkages and transitions, there is
a blurring of the rural–urban divide (Tacoli, 2008). In this chapter we explore
the ways in which rural–urban linkages have been studied, arguing for a more
fluid approach that focuses on the dynamics between rural and urban areas.
We then provide a background to the key concepts that we consider are
central to an analysis of rural–urban dynamics, and which run throughout
the book, namely markets, livelihoods and mobility. Subsequently we intro-
duce the reader to the methodological approaches adopted in the collection
and analysis of the data in this multi-sited comparative project.

Conceptualising rural–urban dynamics

Although development theory and practice have traditionally been based on
a dichotomy between rural and urban areas, populations and activities, in
recent years increasing attention has been paid to the extensive linkages
between them. Cecilia Tacoli (1998) was one of the first scholars to provide
an analytical framework for analysing rural–urban linkages. Her framework
consisted of two types of linkages: linkages across space (including flows of
commodities, people, and information) and linkages between sectors (such as
between agriculture, services and manufacturing). These linkages were seen
as being central to achieving balanced economic growth and reducing
vulnerability due to the critical role they play in the livelihood strategies of
the poor.
In relation to economic growth, urban and rural economies often enjoy a
symbiotic relationship. Cities benefit when agricultural productivity increases
as rural areas provide important markets for urban services and manu-
factured goods. A boom in commercial agriculture boosts demand for
marketing, transportation, construction and finance, which urban centres
often provide. Rural areas can also benefit from the growth of cities as they
provide markets for agricultural and dairy products, and for rural non-farm
output. Rural industries often supply parts and components to nearby urban
10 K. V. Gough, J. Agergaard, N. Fold, L. Møller-Jensen
manufacturers. Urbanisation can also help raise productivity through
technology transfers, educational services and training (Tacoli, 1998).
Increasing competition over scarce resources in rural areas has resulted in
many individuals and households migrating to urban areas in search of better
economic opportunities. However, individual and household mobility depend
on economic, intellectual and familial resources, and social differences such
as gender, generation and ethnicity further impact migration decision making
and selectivity. Flows of people, though, are not always from rural to urban
areas. Many urban dwellers maintain strong links with their rural roots,
returning to their home villages for short visits and even investing in housing.
Return migration to rural areas also occurs in periods of stagnation and
depression in urban manufacturing and service activities. Many of the elderly
have traditionally returned to their home villages in their old age though there
may be an increasing trend for them to stay in urban areas. Consequently,
households cannot always be defined as either urban or rural. In order to
minimise risk, and in an attempt to maximise income, many households are
multi-spatial with some members working in urban areas while others remain
in rural areas (Tacoli, 2002). These multi-spatial households are not just
found at the local scale but include households with members who have
migrated internationally.
Migrants often send remittances to relatives and communities in sending
areas which constitute an important aspect of financial flows between rural
and urban areas along with transfers such as pensions to migrants returning
to their home settlements, investments and credit from urban-based institu-
tions. Information also flows freely between urban and rural areas as a wide
range of types of information is sought by both rural and urban dwellers
such as access to land and housing, education and employment opportun-
ities, and the prices of raw materials and commodities. The improved avail-
ability of old and new information and communication technologies provides
better opportunities for creating and sustaining linkages between places.
These spatial flows of commodities, products, people, money and ideas
overlap with inter-linkages between sectors both at the household level and
at the wider level of local economies. They include backward and forward
linkages between agriculture, manufacturing and services, such as production
inputs and the processing of agricultural raw materials. Urban centres, espe-
cially small and intermediate ones, rely on broad-based demand for basic
goods and services to develop their secondary and tertiary sectors. Overall,
synergy between agricultural production and urban-based enterprises has
been argued to be the key to the development of vibrant local economies and,
on a wider level, to less unequal and more ‘pro-poor’ regional economic
growth (Tacoli and Satterthwaite, 2003). In this virtuous cycle of rural–urban
relations, rising agricultural incomes create a demand for more consumer
goods and services, spurring the development of non-farm activities, absorb-
ing surplus farm labour, further boosting demand for agricultural output,
and thus feeding back into increased agricultural production and rising
 Comparison of rural–urban dynamics 11
incomes (Evans and Ngau, 1991; Evans, 1992; Grabowski, 1995). There is
also reason, though, to think that some aspects of rural–urban relations
undermine traditional production systems and may even marginalise more
vulnerable groups in rural society as they are left out of the modernisation
‘loop’ (Hart, 1996).
The last decade has seen a renewed interest in the role of rural–urban
linkages in economic development at the global, national and local levels
(Lynch, 2005). At the global level, the liberalisation of trade and production
plays a critical role in shaping rural–urban linkages. The increased availability
of imported manufactured and processed goods affects consumption pat-
terns in both rural and urban settlements, but since these are often cheaper
than locally produced goods, local manufacturers and processors can be
negatively affected. In the agricultural sector, trade in export crops is largely
controlled by international companies who tend to by-pass local urban
centres for processing and marketing, and who also retain much of the
value-added and do not necessarily invest it in the producing region or even
nation (Lerise et al., 2000).
At the national level, macro-economic policies linked to reform and
structural adjustment impact on rural–urban linkages. The sharp reduction
in subsidies to agricultural inputs has affected the incomes of small-scale,
under-capitalised farmers, while the retrenchment of workers in the formal
sector has deepened financial insecurity in the urban centres. At the same
time, the increase in the cost of food and the introduction of user fees
for education and health services has forced many households to seek
cash incomes through employment diversification – including non-farm
occupations for rural residents, often located in urban centres – migration
and urban agriculture.
At the local level, the nature and scope of rural–urban interactions is
influenced by the nature of agricultural land, population density and distri-
bution patterns, farming systems and the availability of roads and transport
networks. Local governments can play an important role in supporting
positive rural–urban linkages. The challenge is to identify the themes and
processes at the local level that may have broader resonance and relevance,
and those that are linked to the specificities of place and the particularities of
households and individuals (Tacoli, 2006). The contribution of rural–urban
linkages to livelihoods varies depending on households’ and individuals’
wealth and status and, therefore, their gender, age, ethnicity and in many
cases their religious and/or political affiliation.
Not only are rural–urban linkages multi-faceted, but the rural and urban
can be conceptualised and categorised in differing ways: in administrative
terms, in terms of livelihoods and economy, and in social and cultural terms.
The administrative approach is the space of planners and policy makers
where, based on identified objective criteria, places become designated and
classified as rural or urban. Often using a definition of the urban according to
population size thresholds (sometimes combined with other criteria), areas
12 K. V. Gough, J. Agergaard, N. Fold, L. Møller-Jensen
are delimited and mapped accordingly. The livelihood and economy
approach links rural and urban spaces with certain types of production and,
by association, with particular occupations and livelihoods (Rigg, 2007). The
rural is dominated by primary production and agriculture and, thereby, with
occupations and livelihoods associated with farming and primary produc-
tion. The urban, on the other hand, is associated with manufacturing and
industry and occupations and livelihoods are, likewise, shaped by this fact.
Together, these two approaches appeal to many scholars and policy makers
because they permit neat mapping and categorisation, at least from afar.
The social and cultural approach does not treat the rural and urban as an
administrative space or a space of production, but is an idea of rural and
urban in terms of their social and cultural meanings. This is more nebulous
and harder to pin down as it exists independently of administrative desig-
nation and economic activity and is constantly in flux as societies transform.
It is growing in importance as the ‘objective’ reality of the rural and urban is
being eroded. Thus Thompson (2004, 2007) proposes that rural villages have
become de facto urban spaces in terms of production, consumption and
social interaction.
Given the complex nature of the terms ‘rural’ and ‘urban’, and their
intricate inter-linkages, we prefer in this book to talk about ‘rural–urban
dynamics’. Central to the conceptual approach is the belief that people and
activities are multi-scalar and embody and inhabit different scales of thought
and action simultaneously. Thus the rural and urban are both dynamic spaces
and places where there are constant movements of commodities to and from
markets and of people engaged in diverse livelihood strategies. Settlements
can be difficult to define as being either urban or rural: rather they are better
seen as forming a rural–urban continuum. Thus markets, livelihoods and
mobility are all central to our conceptualisation of rural–urban dynamics
and will be explored in more detail here.

Markets
With the global trend towards market-led strategies, market interactions are
an increasingly crucial factor in the development of rural–urban dynamics.
As many countries in the Global South are considered to have a comparative
advantage in the extraction of natural resources, such as gold and diamond
mining, and export-oriented agricultural production, it is essential that there
are efficient economic linkages connecting producers with both domestic and
external markets. Having access to markets is thus central to ensuring sus-
tainable livelihoods. As markets have become increasingly global in orienta-
tion, one of the key ways in which they have been analysed is through global
value chain (GVC) analysis, the broad aim of which is to trace commodities
and the value added from source to the final consumer.
The GVC approach emerged in the mid 1990s, building on concepts from
world-systems theory and focusing on North–South inequalities. Three core
 Comparison of rural–urban dynamics 13
components of the approach were identified: an input–output structure, a
territoriality (or ‘geography’) and a governance structure (Gereffi, 1994). The
first component mapped GVCs in terms of material flows and technical
transformation processes while the second identified major locations for
production and consumption. The third encapsulated the dual processes of
increased division and intensified coordination of production on a global
scale. The merit of the initial approach was the identification of two arche-
typal types of chain governance, namely ‘buyer-driven’ and ‘producer-driven’
GVCs, each characterised by specific sets of lead firms, entry barriers and
organisational structure. Producer-driven chains are primarily coordinated
by transnational companies in industries characterised by high capital
intensity in the manufacturing process, as well as by high technological and
organisational barriers to entry. Upstream and downstream activities are
organised and controlled by these dominant lead firms, who themselves
command substantial productive capacity. Buyer-driven chains are primarily
coordinated by large retailers, brand-name merchandisers or trading com-
panies in industries that are characterised by high labour intensity and high
barriers to entry concerning product design, financing and marketing. Down-
stream activities are controlled through complex, tiered networks of (over-
seas) contractors and the lead firms do not engage in processing activities.
This simple dichotomisation of corporate industrial organisation has been
heavily critiqued (Henderson et al., 2002; Whitley, 1996) and more recently,
five forms of governance have been suggested based on the characteristics of
the transactions between suppliers and lead firms, and the capacity of the
supplier (Gereffi et al., 2005). These governance forms determine the condi-
tions for upgrading suppliers involved in GVCs. Upgrading denotes the
processes whereby suppliers enter into new and higher-value added manu-
facturing processes that require greater levels of technical or organisational
competency. In chains characterised by some governance forms, lead firms
may actively try to upgrade their supplier base by assisting local firms to
climb different sorts of learning curves. A somewhat parallel line of enquiry
is preoccupied with global production networks (GPNs) and their impact on
different spatial configurations (Coe et al., 2004). Instead of upgrading,
the key concepts here are value creation, value enhancement and value cap-
ture within a spatial, in particular regional, context. The aim is to examine
the conditions for starting regional production activities, the degree to which
value is added to production activities over time, and the scope to which
added value is retained and siphoned out to other regional activities, whether
they are somehow linked or used to initiate completely new and previously
unseen activities. GPN proponents have also stressed the need to incorporate
multilateral organisations, state authorities, trade associations, trade unions
and NGOs into the analytical framework. These actors can influence organ-
isational structures and dynamics in the GVCs (or production networks)
as much as ‘internal’ company actors directly involved in the flow of goods,
money and information.
14 K. V. Gough, J. Agergaard, N. Fold, L. Møller-Jensen
Even though the analytical framework of GVC analysis has been consider-
ably improved by adding spatial and institutional dimensions, various
dilemmas remain. First is the issue of how to translate successful upgrading
processes at the firm level upwards to wider regional or national economy
levels. The concept of value capture has not yet been operationalised beyond
the analysis of regional spin-off effects and it is not clear whether upgrading
is positive or negative (Bair, 2005). Second, both the GVC approach and the
spatial impacts of GPN situate their attention on upgrading at the company
level. The outcome for workers is not an explicit concern but is implicitly
deduced from the results of upgrading processes, i.e. if local companies
upgrade, workers will benefit in terms of skill upgrading and (presumably)
higher wages. However, this is not an evident outcome as there are no
mechanisms to ensure that the upgrading of firms works its way downwards
to the factory floor. Third, when it comes to the question of how chain
dynamics and upgrading processes directly affect small-scale producers, the
GVC literature is conspicuously silent.
It is this latter concern that has been of particular interest in our studies
of rural–urban dynamics in frontier regions. In particular, how are changes
in demand at the global level worked down through the GVC to affect produ-
cers such as small-scale farmers or artisans in the frontier settlements? How
are these producers linked into the GVC and how does their ability, or in-
ability, to meet certain quality criteria (standards) affect their position as a
supplier for the global market? In order to answer these sorts of questions,
it is necessary to link GVC analysis, with its top-down approach, with liveli-
hood analysis which adopts a bottom-up approach. It is to the livelihoods
approach that we now turn.

Livelihoods
Livelihoods have been a focus of both research and policy since the 1990s.
The initial livelihoods approach is generally traced back to Robert Chambers
and Gordon Conway who published an IDS discussion paper in 1992 on
‘Sustainable rural livelihoods: Practical concepts for the 21st century’. They
saw livelihoods as comprising people, their capabilities and their means of
making a living including food, income, tangible assets (such as resources and
stores) and intangible assets (such as claims and access). Their approach
was, of course, not developed in a vacuum. It built upon a trend for a more
actor-oriented approach to development which was adopted in development
studies as a reaction to the structural perspective of dependency theories and
neo-Marxist approaches (De Haan and Zoomers, 2005). The actor-oriented
approach, especially encapsulated in the work of Norman Long (1984),
focused on the micro-world of family, networks and communities and drew
attention to related issues such as poverty, vulnerability and marginalisation.
During the 1980s, a range of household studies were undertaken both under
the framework of ‘new household economics’ and more socially inspired
 Comparison of rural–urban dynamics 15
‘survival strategies’. The latter showed how household decisions are often
made within ‘the confines of limiting structural constraints, although families
nevertheless operate with a degree of relative autonomy’ (Humphries, 1982,
cited in de Haan and Zoomers, 2005: 29). Many of these household studies
showed how poor households were being increasingly marginalised from the
benefits of economic growth.
One of the reasons Chambers and Conway put forward for developing
the livelihood approach was to counter the pessimistic approach of many
household studies which they also felt tended to undervalue future liveli-
hoods. The livelihood approach was soon adopted by major international
and aid agencies including UNDP, Oxfam and CARE (De Haan and Zoom-
ers, 2005). The concept was further developed in the late 1990s primarily by
British academics such as Scoones (1998) on a sustainable rural livelihoods
approach, Carney (1998) on natural resources in livelihoods, Ellis (1998) on
the diversification of livelihood activities and Bebbington (1999) on capitals
and capabilities. One of the most widely quoted definitions of livelihoods is
still from Carney:

A livelihood system comprises the capabilities, assets (including both

material and social resources) and activities required for a means of
living. A livelihood is sustainable when it can cope with and recover from
stresses and shocks and maintain or enhance its capabilities and assets
both now and in the future, while not undermining the natural resource
base.
(1998: 2)

Another major impetus to the livelihoods approach followed its adaptation

by the Department for International Development (DFID) in the UK’s
development policy and the subsequent research projects and policy debates
which emerged from this, including a new focus on urban livelihoods (see
Rakodi with Lloyd-Jones, 2002).
Recently, de Haan and Zoomers (2005), in a paper ironically called
‘Exploring the frontier of livelihoods research’, have called for the livelihood
approach to become more powerful analytically by improving its theoretical
depth through incorporating the issues of sovereignty and power. They claim
that, ‘Although transforming structures, mediating processes, institutions
and organisations appear in all livelihood frameworks, there is a tendency
within livelihood studies to downplay these structural features and to focus
on capitals and activities’ (de Haan and Zoomers, 2005: 33). As they argue,
livelihood activities are not neutral but encompass processes of inclusion and
exclusion. Access thus depends on social relations which are not necessarily
harmonious. The livelihood approach, they argue, needs to incorporate
power relations in order to improve its conceptualisation of access. In order
to do this, they suggest using the methodological tool of livelihood trajector-
ies which makes use of life histories but tries to penetrate into a deeper layer
16 K. V. Gough, J. Agergaard, N. Fold, L. Møller-Jensen
of beliefs, needs, aspirations and limitations that is contextualised in relation
to power and institutions (de Haan and Zoomers, 2005).
In our analysis of rural–urban dynamics, we have elected to explore the
underlying processes through linking the livelihood approach with GVC
analysis. In this way, a bottom-up (livelihood) analysis can be combined
with a top-down (GVC) analysis thus furthering our understanding of the
dynamics of livelihood transformations. This moves livelihood analysis away
from a methodological focus exclusively oriented around individuals and
households.

Mobility
Mobility is a ubiquitous term having both a spatial and temporal dimension
(van Dijk et al., 2001) and can also refer to movement in society. A wide
range of forms of mobility have been documented including rural–rural,
urban–rural, rural–urban, nomadism and refugeeism operating at scales
stemming from the global to the local. Mobility is so widespread that it
has been argued that mobility should not be seen as a rupture in society
but as a normal way of life, with immobility being the anomaly (de Bruijn
et al., 2001).
Recently it has been claimed that ‘a “mobility turn” is spreading into and
transforming the social sciences, not only placing new issues on the table
but also transcending disciplinary boundaries and putting into question the
fundamental “territorial” and “sedentary” precepts of twentieth-century
social science. It seems that a new paradigm is being formed within the social
sciences, the “new mobilities” paradigm’ (Hannam et al., 2006: 1–2). Aspects
that these authors highlight include: the relation between human mobilities/
immobilities and unequal power relations; the relation between mobility
systems and economic regulation and governance; and the complexity of
mobility and the dynamics of physical, informational, virtual and imagina-
tive forms of mobility. They claim that the mobilities paradigm illustrates
how all places are tied into at least thin networks of connections that stretch
beyond each place resulting in nowhere being an island. While moving
between places can be a source of status and power, not everyone has an
equal relationship with mobility (Sheller and Urry, 2006). As Adey (2006: 83)
claims, mobility ‘means different things, to different people, in differing social
circumstances’. The mobility paradigm is thus part of a broader theoretical
project that aims to go beyond the imagery of terrains as spatially fixed
geographical containers for social processes, and calls into question scalar
logics such as local/global as descriptors of regional extent.
There has been a particular focus within the mobility turn on advances in
new technology and new ways of living and interacting, such as the airport
encounter, the internet, virtual travel, and ipods. As Jonathan Rigg (2007)
has claimed, this has inevitably meant that the Global North has been taken
as the primary site of interest and research and the space where new
 Comparison of rural–urban dynamics 17
mobilities are emerging. As he asks, ‘Is this yet another case where the Global
South has to patiently wait before the ripples of change wash on its shores?’
(Rigg, 2007: 118). In an attempt to pre-empt this he sketches out an alterna-
tive new mobilities paradigm focusing on the Global South which highlights
the linkages between mobility and livelihoods. It is also important not to
overlook other geographical research which continues to illuminate the
demographic, cultural and economic dynamics of migration (Blunt, 2007).
Recent research has highlighted how circular and temporary migrations are
key elements of household livelihoods in both rural and urban settlements
in the Global South (Tacoli, 2008). This mobility is linked to the diversifica-
tion of income sources and together with the movement of commodities is
an important aspect of the links between rural and urban areas. Mobility is
thus central to both livelihoods and markets and is key to understanding
rural–urban dynamics.

Performing comparison of rural–urban dynamics

Markets, livelihoods and mobility have all been shown to be central to an
understanding of rural–urban dynamics. In order to explore the ways in which
they interact in frontier settlements a multi-stage comparative method-
ological approach was adopted.

Comparative approach
Comparative empirical work has always been present in the social sciences
but has wavered in relation to its popularity. Recently, both Nijman (2007)
and Ward (2008) have called for a comparative (re)turn in urban studies. With
the comparative approach to rural–urban dynamics adopted in this book, we
partially contribute to this turn. In conducting comparative research it is vital
to be aware of what is being compared: ‘Because similarities or differences
are not given in the things themselves but in the ways they are contextualised’
we must ‘compare meanings, ways of constructing relationships between
objects, persons, situations, events’ (Melhuus, 2002: 82).
Comparisons are relevant in a number of differing ways within the research
presented here. One of the most obvious, and the one around which this
book is structured, is to compare frontier types: i.e. within agricultural, handi-
craft/manufacturing, and mining frontiers. Comparison was uppermost in
the selection process of the frontiers within the respective countries: hence
each of the frontier types are represented in more than one country. Thus, in
Ghana and Tanzania a mining frontier was studied, and in Vietnam and
Thailand a handicraft frontier was studied, thus facilitating comparisons
within Africa and Asia. In both Ghana and Vietnam agricultural frontiers
were studied, facilitating comparisons between Africa and Asia. Another
potential comparison is to look across frontier experiences within particular
countries to analyse how national conditions impact rural–urban dynamics.
18 K. V. Gough, J. Agergaard, N. Fold, L. Møller-Jensen
By comparing the similarities and differences between the transformations in
markets, livelihoods and mobility in the frontier settlements, we are searching
for possible generalisations as well as highlighting diversity.
A common methodology was adopted in order to facilitate comparison
across all of the frontiers studied. This methodology was developed through
workshops held on a yearly basis for all programme partners. These consti-
tuted a crucial element in the programme as they acted as a locus for the
integration of the programme partners, the cross-fertilisation of methods and
ideas, and the exchange of results. The first workshop, held in Copenhagen,
focused on drawing up the common methodology in detail. At the second
workshop, where the group congregated in Hanoi, the preliminary findings of
the first stages of the fieldwork were presented and the setting up of a com-
mon database was discussed. The final workshop, held in Ghana, was more
of a writing workshop where the researchers of the different frontier types
had the chance to discuss their findings in relation to each other and the
outline for this book was drawn up. Holding the second and third workshops
in the two main programme countries facilitated contact with local academics
and visits were made to two of the frontiers.

Common methodology
In order to explore the nature of rural–urban dynamics in the frontier settle-
ments, and the processes underlying the transformations, a methodology was
constructed which combined analysis of livelihoods, mobility and markets. A
range of research methods were employed including analysis of secondary
data and the collection of both qualitative and quantitative data. As outlined
above, to facilitate comparisons between the frontiers, as far as possible a
common methodology was followed. The fieldwork was conducted in three
stages (Table 2.1).
The first stage, conducted in 2004/2005, consisted of initial settlement
selection within the frontier regions. The number of settlements studied within

Table 2.1 Summary of fieldwork stages

Timing of Fieldwork methods

fieldwork

Stage 1 Selection of Semi-structured Focus group Interviews with

study settlements interviews with discussions in key informants in
local leaders settlements commodity chains
Stage 2 Questionnaire survey with randomly selected households in all settlements
Stage 3 In-depth interviews Focus group discussions Interviews with key
in each settlement with actors engaged in informants in
production of the key commodity chain
commodity
 Comparison of rural–urban dynamics 19
each frontier differed (Table 2.2) depending on variations in the commodities
being studied and the settlement types in the frontier. Semi-structured
interviews were held with key people in the settlements, including local civic,
education, health and religious leaders, in order to draw up a picture of
the changing settlement characteristics. Secondary data were also collected to
add to our knowledge of settlement transformation. Focus group discussions
were held in the settlements with a number of different groups including
chiefs and elders/local leaders, women’s groups and youth groups. Topics
covered included the history of the settlement, changing livelihoods, mobility,
and the production of the key commodity in the frontier. Most of these
interviews were conducted by the authors working together with research
assistants, and were taped and subsequently transcribed. Interviews were
made with key players in the global value chain of the commodity (primarily
traders and exporters) and initial interviews were held with relevant business
associations, government institutions and academics. This first stage of
fieldwork enabled us to obtain an overview of the characteristics of the
settlements, the changing livelihoods of the inhabitants and relations between
the key actors in the value chain.
A questionnaire survey formed the second stage of the fieldwork which
took place during 2005/2006. The aim was to discover more detail at the
household level regarding changing livelihoods, mobility and engagement in
the key commodity. Households in each settlement were selected to answer
the questionnaire survey which was conducted by local assistants accom-
panied at times by the authors. The number of households interviewed
differed in each frontier according to the size of the settlements and vari-
ations in the populations. An attempt was made to select the households as
randomly as possible; in some frontiers, such as in Vietnam where household
census data exist, it was possible to make a random selection whereas in
others, such as in Ghana where no such census was available, houses were
selected at random. The questionnaire had a core that was common to all of
the frontier settlements where the same questions were asked but, due to

Table 2.2 Data collection in the frontier settlements

Frontier Number of settlements Number of questionnaires

studied conducted

Cocoa (Ghana) 4 360

Coffee (Vietnam) 4 566
Pineapple (Ghana) 2 150
Fruit (Vietnam) 3 300
Handicraft (Vietnam) 2 200
Handicraft (Thailand) 2 228
Manufacturing (Thailand) 2 137
Mining (Ghana) 4 484
Mining (Tanzania) 3 216
20 K. V. Gough, J. Agergaard, N. Fold, L. Møller-Jensen
cultural differences, the coding of the answers inevitably varied. This was
followed by a section which was particular to the frontier being studied where
questions in relation to the livelihood activity in focus were asked. In relation
to global value chains, emphasis was put on the identification of barriers to
entry and processes of differentiation, particularly the possibilities for local
companies to upgrade their functional roles within the chains and the
impacts on small-scale producers.
In the third and final stage of fieldwork, in 2006/2007, a return visit was
made to a number of households in each settlement to conduct in-depth
interviews. The households were selected from the questionnaire survey to
cover a range of household types varying by age, gender, ethnicity, length of
residence in the settlement, degree of engagement in the key commodity, etc.
The in-depth interviews traced the respondents’ life histories, focusing on
their changing livelihoods, mobility patterns and settlement experiences.
Again, the number of in-depth interviews conducted varied by frontier
according to the complexity of the settlement and the key commodity in
question. As part of the commodity chain analysis, return visits were made to
relevant actors to find out the latest developments in the sectors in question.

Data analysis and dissemination

The questionnaire data from each frontier survey were digitally coded using
a two-level database model. On the first level, household data such as sources
of income, land and livestock ownership, etc. were recorded into a dataset
(household-level). On the second level, data related to individual household
members were recorded into a separate dataset (person-level). This level
included, among other things, data on individual mobility. The selected strat-
egy maintained data integrity and reduced data repetition while providing
flexibility for various types of data analysis. The majority of the question-
naire and coding scheme was identical for all frontiers to provide the basis
for cross-cutting analysis. It was necessary, however, to produce modifications
to the scope of possible values for some of the variables in the dataset, due
to the cultural and economic differences between the countries. The mani-
festation of these differences necessitated country-specific additions to the list
of answer categories that could be understood only in the local context and
proved difficult to resolve into a global set of meaningful categories. For this
reason, data analysis was performed separately on each frontier rather than
creating a large common database. Data analysis based on value listings,
frequency counts and bi-variate cross-tabulations was mainly carried out in
SPSS. The results obtained from the basic analysis of the survey data of each
frontier were subsequently combined with the qualitative data and analysed
within the framework of the themes that this book sets out to investigate.
An important aspect of dissemination of the programme findings was
the holding of policy seminars in each of the respective countries. These
were attended by central and local politicians, aid agencies, key actors in the
 Comparison of rural–urban dynamics 21
commodity chain, and settlement leaders. Feeding back into policy making
was a key aim of the programme and the animated discussions at the work-
shops indicated the importance of this aspect of the methodology. The
feedback from policy makers in turn influenced our thinking regarding the
transformations of the various frontiers and has contributed to the shaping
of the arguments put forward in this book.

References
Adey, P. (2006) ‘If mobility is everything then it is nothing: towards a relational
politics of (im)mobilities’, Mobilities, 1(1): 75–94.
Bair, J. (2005) ‘Global capitalism and commodity chains: Looking back, going
forward’, Competition and Change, 9(2): 153–80.
Bebbington, A. (1999) ‘Capitals and capabilities: A framework for analyzing peasant
viability, rural livelihoods and poverty’, World Development, 27(12): 2012–44.
Blunt, A. (2007) ‘Cultural geographies of migration: Mobility, transnationality and
diaspora’, Progress in Human Geography, 31(5): 684–94.
Carney, D. (ed.) (1998) Sustainable rural livelihoods: What contribution can we make?
Nottingham: Russell Press Ltd.
Chambers, R. and Conway, G.R. (1992) Sustainable rural livelihoods: Practical
concepts for the 21st century, Institute of Development Studies Discussion Paper
296. Brighton, Institute of Development Studies.
Coe, N., Hess, M., Yeung, H.W., Dicken, P. and Henderson, J. (2004) ‘ “Globalizing”
regional development: A global production networks perspective’, Transactions of
the Institute of British Geographers, 29: 468–84.
De Bruijn, M., van Dijk, R. and Foeken, D. (2001) ‘Mobile Africa: An introduction’,
in de Bruijn, M., van Dijk, R. and Foeken, D. (eds), Mobile Africa: Changing
patterns of movement in Africa and beyond. Leiden: Brill, pp. 1–26.
De Haan, L. and Zoomers, A. (2005) ‘Exploring the frontier of livelihoods research’,
Development and Change, 36(1): 27–47.
Ellis, F. (1998) ‘Survey article: Households strategies and rural livelihood diversifica-
tion’, Journal of Development Studies, 35(1): 1–38.
Evans, H. (1992) ‘A virtuous circle model of rural–urban development: Evidence
from a Kenyan small town and its hinterland’, Journal of Development Studies,
28(4): 640–67.
Evans, H. and Ngau, P. (1991) ‘Rural–urban relations, household income diversifica-
tion, and agricultural productivity’, Development and Change, 22: 519–45.
Gereffi, G. (1994) ‘The organization of buyer-driven global commodity chains:
How U.S. retailers shape overseas production networks’, in G. Gereffi and
M. Korzeniewicz (eds), Commodity chains and global capitalism. London: Praeger,
pp. 95–122.
Gereffi, G., Humphrey, J. and Sturgeon, T. (2005) ‘The governance of global value
chains’, Review of International Political Economy, 48(1): 78–104.
Grabowski, R. (1995) ‘Commercialization, nonagricultural production, agricultural
innovation, and economic development’, The Journal of Developing Areas, 30:
41–62.
Hannam, K., Sheller, M. and Urry, J. (2006) ‘Mobilities, immobilities and moorings’,
Mobilities, 1: 1–22.
22 K. V. Gough, J. Agergaard, N. Fold, L. Møller-Jensen
Hart, G. (1996) ‘The agrarian question and industrial dispersal in South Africa:
Agro-industrial linkages through Asian lenses’, Journal of Peasant Studies,
23(2–3): 245–77.
Henderson, J., Dicken, P., Hess, M., Coe, N. and Yeung, H.W. (2002) ‘Global produc-
tion networks and the analysis of economic development’, Review of International
Political Economy, 9(3): 436–64.
Humphries, J. (1982) ‘The family and economy: Notes towards a relative autonomy
approach’, Paper presented at the conference on Demographic Research in Latin
America, Mexico City.
Lerise, F., Kibadu, A., Mbutolwe, E. and Moshi, N. (2000) The case of Lindi and its
region, Southern Tanzania, IIED Working Paper 2, London: IIED.
Long, N. (1984) Family and work in rural societies, London: Tavistock Publications.
Lynch, K (2005) Rural–urban interaction in the developing world, London and New
York: Routledge.
Melhuus, M. (2002) ‘Issues of relevance: Anthropology and the challenges of
cross-cultural comparison’, in Gingrich, A. and Fox, R (eds), Anthropology by
comparison, London and New York: Routledge, pp. 70–91.
Nijman, J. (2007) ‘Comparative urbanism’, Urban Geography, 28: 1–6.
Rakodi, C. with Lloyd Jones, T. (2002) Urban livelihoods: A people-centred approach to
reducing poverty, London: Earthscan.
Rigg, J. (2007) An everyday geography of the Global South, London and New York:
Routledge.
Scoones, I. (1998) Sustainable rural livelihoods: A framework for analysis, IDS working
paper, Brighton: Institute of Development Studies.
Sheller, M. and Urry, J. (2006) ‘The new mobilities paradigm’, Environment and
Planning A, 38: 207–26.
Tacoli, C. (1998) ‘Rural–urban interactions: A guide to the literature’, Environment
and Urbanization, 10 (1): 147–66.
Tacoli, C. (2002) Changing rural–urban interactions in sub-Saharan Africa and their
impacts on livelihoods: A summary, Rural–Urban Working Paper 7, London: IIED.
Tacoli, C. (2006) ‘Editor’s introduction’, in Earthscan reader in rural–urban linkages,
London: Earthscan, pp. 3–14.
Tacoli, C. (2008) ‘Links between rural and urban development in Africa and Asia’,
New York: United Nations Secretariat, UN/POP/EGM-URB/2008/09.
Tacoli, C. and Satterthwaite, D. (2003) The urban part of rural development: The role
of small and intermediate urban centres in rural and regional development and
poverty reduction, Rural–Urban Working Paper 9, London: IIED.
Thompson, E.C. (2004) ‘Rural villages as socially urban spaces in Malaysia’, Urban
Studies, 41(12): 2358–76.
Thompson, E.C. (2007) Unsettling absences: Urbanism in rural Malaysia, Singapore:
Singapore University Press.
Van Dijk, H., Foeken, D. and van Til, K. (2001) ‘Population mobility in Africa:
An overview’, in Bruijn, M., van Dijk, R. and Foeken, D. (eds), Mobile Africa:
Changing patterns of movement in Africa and beyond, Leiden: Brill, pp. 9–26.
Ward, K. (2008) ‘Editorial – toward a comparative (re)turn in urban studies? Some
reflections’, Urban Geography, 29(5): 405–10.
Whitley, R. (1996) ‘Business systems and global commodity chains: Competing or
complementary forms of economic organisation?’, Competition and Change, 1:
411–25.
3 Small town development in the
Ghanaian cocoa frontier
Michael Helt Knudsen

Introduction
This chapter deals with urbanisation processes in a predominantly rural area
of Ghana, far from the major urban centres. For more than a century, cocoa
production has been one of the most important factors determining migra-
tion patterns and settlement development in rural Ghana. As cocoa produc-
tion expanded through the forest areas of southern Ghana, migrant farmers
and labourers followed cocoa to its new sites. In the new cocoa frontiers, old
settlements expanded and new settlements were established to house the
migrants. The same process has also occurred in Ghana’s current cocoa fron-
tier in the Western Region. The Western Region emerged as the new Ghanaian
cocoa frontier in the 1960s and 1970s, and since the 1990s it has accounted for
around half of Ghana’s total cocoa output (Fold, 2004; Gockowski, 2007).
Over time, some of the frontier settlements have reached a population size
qualifying them to be regarded as urban areas or small towns. Other settle-
ments, however, have stagnated, hardly changing in size or function.
The aim of this chapter is to explore the nature of settlement growth in the
current cocoa frontier. In particular, the question of why some settlements
grow while others stagnate is explored. It shows that settlement growth can be
difficult to predict, but also that the location and nature of the population
(indigenous vs. migrants), which directly affect livelihood opportunities, are
both central factors. To set the scene, the next section briefly introduces the
role of cocoa production in Ghana. Subsequently, two small towns and two
rural settlements in the Ghanaian cocoa frontier are presented. Against this
background, the following section discusses the processes behind settlement
growth in the frontier by examining the diverse development trajectories of
the two small towns and their differing functions for nearby rural settlements.
The final section provides some concluding remarks.

A brief introduction to cocoa in Ghana

The history of cocoa in Ghana is the story of how, for more than a century, a
single crop has had implications for almost every facet of the economic,
24 Michael Helt Knudsen
social and political life of the country (Mikell, 1989/1992). The commercial
growing of cocoa was initiated in 1879 on what was then the Gold Coast
(becoming Ghana after independence) (Amoah, 1995). The crop achieved
immediate success among farmers, and by as early as 1910–1911 the Gold
Coast had become the leading cocoa producer in the world (Leitner and
Harding, 2004). The territory’s importance as a producer peaked in the early
1930s, when it accounted for about 40 per cent of global production (Fold,
2002). Ghana retained its position as the world’s largest producer of cocoa
until 1976–1977 (Amoah, 1995). However, an unstable political environment
and low producer prices in the late 1970s and the early 1980s resulted in
production declining to around 10 per cent of world production in the early
1980s (Bulíř, 2003). Since 1983, a series of structural adjustment programmes
have been implemented, one of the consequences of which has been a
substantial increase in cocoa producer prices and the recovery of cocoa
production.
Today Ghana is the second largest producer of cocoa on a global scale,
only surpassed by neighbouring Ivory Coast (Vigneri and Santos, 2007).
Cocoa production is at present still an important contributor to the country’s
GDP as one of the top export commodities in revenue terms (US Depart-
ment of State, 2008). The cocoa sector is important for rural employment,
both as a smallholder activity for around 800,000 cocoa farmers, and for
temporary (seasonal) employment by a high number of seasonal wage
labourers (Quartey, 2007). The sector also contributes to employment in
urban areas, which function as administrative centres, as well as centres for
the storage, grading and transport of cocoa.
The reform of the cocoa sector resulted in a partial dismantling of the
state-owned marketing board Ghana Cocoa Board (COCOBOD) and the
removal of subsidies on farm inputs with effect from 1992 (Hutchful, 2002).
This means that the purchasing of cocoa beans up-country and the transport
of beans to the ports have been liberalised. These liberalised activities are
handled by local and international Licensed Buying Companies (LBCs),1
including the former monopsony Produce Buying Company (PBC) of
COCOBOD.2 One result of the multiple-buyer system has been that there is
now a far greater number of cocoa-buying sheds spread over the cocoa-
producing areas, often reaching far into the rural areas. This has provided
farmers with easier access to sell their produce, as well as making it possible
for them to change buyers if they feel cheated.
Since the reforms, COCOBOD has retained its monopoly of quality con-
trol and export marketing through the Quality Control Division (QCD) and
the Cocoa Marketing Company (CMC) respectively. This means that all
cocoa purchased by the LBCs must be resold at a fixed price to CMC, which
acts as the sole seller of Ghana’s cocoa on the world market. Even though a
stated intention of the liberalisation of the cocoa sector was to liberalise the
export of cocoa, none of the LBCs have so far obtained a licence to export
(www.cocobod.gh). Furthermore, COCOBOD supports farmers by carrying
 Small town development in the Ghanaian cocoa frontier 25
out mass spraying of cocoa farms and providing credit opportunities for the
purchase of fertiliser. These initiatives are mainly conducted through the
agricultural extension service of the Ministry of Food and Agriculture
(MoFA) and the LBCs.
Another important heritage from the pre-SAP period is that COCOBOD
continues to fix the floor price for all domestic purchases of cocoa. This is
done by announcing a pan-seasonal and pan-territorial price at the beginning
of the major crop period (October–March) which usually lasts for the whole
cocoa year, i.e. also covering the harvest during the minor crop period (May–
July), although revised prices are occasionally announced in the early phase
of the latter. This means that the COCOBOD pays a uniform producer price
per kilo to all cocoa farmers throughout the season, irrespective of their
location. The price is determined on the basis of the estimated revenue for
the particular cocoa year, made possible due to the considerable volume of
forward selling by the CMC, itself made possible by the consistently good
quality of cocoa beans produced in Ghana. The forward sales are used as
collateral on the international financial markets, and a substantial loan is
provided each year by a consortium of international banks (Fold, 2004). Due
to the continued high quality of Ghanaian beans, they are sold at a premium
price on the world market.
For the past century, the production of cocoa has resulted in the establish-
ment of frontiers and thus sparked migration and settlement development in
Ghana. The expansion of cocoa production has followed a characteristic
geographical pattern whereby virgin forest land has been cleared and culti-
vated as a consequence of existing cocoa-producing regions gradually col-
lapsing due to ageing cocoa farms, low soil fertility, pests and/or shifting land
use. In general, cocoa production has moved westwards from its cradle on
the Akwapim Ridge, from where it spread to the Ashanti Region and later,
around the 1940s, to the forest areas of Brong Ahafo and Central regions,
which emerged as new cocoa frontiers. Around the 1960s, the frontier
expanded into the Western Region. At present the Western Region is the
main producer of cocoa in the country and is considered the last Ghanaian
cocoa frontier, as no more virgin forest is available in the country.

Small towns and rural settlements in the Ghanaian

cocoa frontier
In Ghana, an urban area is defined as a settlement with a population of
5000 or more, though there is no stated requirement for degree of closeness
(GSS, 2002). No official definition, however, exists for small towns. Owusu
(2005) suggests placing Ghana’s present-day settlement hierarchy within a
four-tier system. At the top of the system he places the large towns and cities
of Accra, Tema Municipality, Kumasi and Sekondi-Takoradi, all with popu-
lations of 250,000 or more. Next are the medium-sized urban centres, with
populations between 50,000 and 250,000, while small towns have populations
26 Michael Helt Knudsen
between 5000 and 50,000. This definition will be adopted for the discussion
on small-town development and functions in this chapter.
In the last inter-censual period, i.e. from 1984 to 2000, the Ghanaian popu-
lation grew from 12.2 million to 18.9 million, which corresponds to an annual
growth rate of 2.7 per cent (GSS, 2005). Alongside the population growth,
there has been an increasing concentration of the population in urban areas,
from 32.2 per cent in 1984 to 43.8 per cent in 2000. The average annual growth
rate between 1984 and 2000 of the 364 localities in Ghana defined as urban is
4.6 per cent, which is significantly above the overall national growth rate
(GSS, 2002; GSS, 2005). Apart from the population growth taking place near
the two largest cities of Ghana (Accra and Kumasi),3 areas of high popula-
tion growth are also found in some rural areas that are remote in relation to
the large urban centres. An important part of the explanation is the existence
of agricultural frontier regions, i.e. regions experiencing a high degree of
immigration by people wishing to undertake specific farming activities.
This chapter builds on research in four settlements4 in Juaboso District,
Western Region (see Figure 3.1), which is located at the heart of the Ghanaian
cocoa frontier and has been one of the highest cocoa-producing districts in
the country for more than a decade. Ninety-two per cent of the population
in the district lives in settlements of fewer than 50005 (Juaboso District
Assembly, 2006). In 2006, the population of Juaboso District was 156,428,
with an annual growth rate between the last two inter-censual periods (1984
and 2000) of 3.5 per cent, higher than both the regional and national averages
of 3.2 per cent and 2.7 per cent respectively. Forty-eight per cent of the popula-
tion is reported to be migrants, almost entirely from other regions of Ghana
(Juaboso District Assembly, 2006). Immigration into the Juaboso area for
agricultural purposes was significant as early as the 1940s. However, it was
not until the 1960s and 1970s that the inflow of migrants gained momentum
along with the start of the boom in the production of cocoa (Boni, 2005).
The influx of migrants has continued into the new millennium. However,
while the first five decades were dominated by migrants to the cocoa sector,
the last decade has witnessed a large influx of migrants with the sole purpose
of engaging in the non-farm sector6 (Knudsen, 2007). This high influx of
migrants, combined with natural growth, resulted in the expansion of old
settlements, as well as the establishment of new ones. Some of these settle-
ments have grown into what can be defined as urban areas. The district today
is characterised by settlements clearly being dominated by either indigenous
or migrant households.7
Research was conducted in two small towns (Bodi and Bonsu Nkwanta)
and two nearby rural settlements (Amoaya and Kefass) (see Figure 3.1). These
settlements were selected to capture the diversity of settlements in the district
in terms of size, location, age and population composition (indigenous versus
migrants). All four settlements are located far from any major urban areas,8
and are not declared district capitals. Bodi and Bonsu Nkwanta are two of the
largest settlements in Juaboso District. Both have experienced remarkable
 Small town development in the Ghanaian cocoa frontier 27

Figure 3.1 Map showing location of study settlements in Juaboso District.

population growth, settlement expansion, and economic and occupational

reorientation in the last two decades. Furthermore, according to population
estimates from 2006, both have passed the official urban population threshold
of 5000 within the last five to ten years and should now officially be defined as
urban areas, or in Owusu’s (2005) definition, small towns (Table 3.1). The two
rural settlements, Amoaya and Kefass, have grown continuously in recent
decades, but they have not witnessed anything like the population and settle-
ment growth experienced by the two small towns (Table 3.1).
The population figures for 2006 that are included in this study (Table 3.1)
28 Michael Helt Knudsen

Table 3.1 Population data on the four study settlements

Bodi Amoaya Bonsu Kefass

Nkwanta

Established 1894 c.1850 1970 1969

Population size 1984 3379 1382 69 122
2000 6929 3155 2308 1374
2006 10,000 4000 8000–9000 2000–2500
(estimated)
Indigenous households (%) 97 73 10 6
Migrant households (%) 3 27 90 94
Sample size 130 60 115 55

Source: Author’s household survey, 2005/2006 and the ‘2000 Population and Housing Census’
(GSS, 2002).

have been estimated by the author, based on the 2000 Population and Hous-
ing Census (GSS, 2002), interviews with key informants and observations
made in the four settlements. The population estimates include only house-
holds living permanently in the settlements. Such population estimates are, of
course, problematic, as neither key informants’ estimates nor my own obser-
vations can be precise. Furthermore, the migrant settlements especially are
characterised by populations that are seasonally very mobile, which means
that the size of the population residing there fluctuates significantly accord-
ing to the cocoa season. This question relates to the problem of drawing the
line between who can actually be defined as settlers. Here, ‘settlers’ refers to
migrant households who have permanent residence in the settlement and
who are living the majority of the year in it. Despite these problems in
establishing exact population figures, there were clear indications of trends in
population development in each settlement. As this is a study about the
processes leading to small-town growth, these trends were the most crucial
ones to identify.

Bodi and Amoaya: old indigenous settlements

Bodi is one of the oldest and currently also the largest settlement in Juaboso
District,9 being characterised by continued population growth (Table 3.1).
Bodi is the capital of the Bodi stool land,10 with the chief of Bodi controlling
large stretches of land, which secures privileged access to land for the local
indigenous households. According to the fieldwork survey indigenous
households make up 97 per cent of the total population (Table 3.1). Hence,
population growth stems primarily from natural growth and intra-stool land
migration.11 Restricted access to land explains, to a large extent, the small
number of migrant households in the settlement. Contrary to many other
indigenous stools, the Bodi stool has been reluctant to sell land to migrants,
 Small town development in the Ghanaian cocoa frontier 29

Table 3.2 Livelihood characteristics of the four study settlements (all figures in
per centages)

Bodi Amoaya Bonsu Kefass

Nkwanta

Households with access to land 100 95 51 67

Average land size (hectares) 9 6 5 6
Households involved in cocoa farming 99 93 50 69
Households involved Landowner 98 86 79 81
in cocoa farming as:
Tenant 2 12 18 13
Labourer 0 2 3 5
Primary source of Cocoa farming 98 88 46 62
income 2006
Trading activities 1 0 23 9
Skilled work 1 3 17 9
Other activities:
• Farm 0 5 3 7
• Non-farm 0 3 11 13
Sample size 130 60 115 55

Source: Author’s household survey, 2005/2006.

and has kept the land within the local matrilineages instead. The majority of
migrants entering Bodi today are seasonal labour migrants, few of whom
decide to settle there.
Until the middle of the twentieth century, the settlement’s economy was
based on food production, mainly as a subsistence activity. Around the 1950s,
cocoa and coffee became important cash crops and primary sources of
income for most households in Bodi. When the production of coffee ended in
the 1980s, cocoa had become by far the most important source of income for
the vast majority of households in the settlement. Nearly all households in
2006 earned their primary source of income from farming cocoa on their own
land (Table 3.2), income that on average constituted more than four-fifths of
total household income.
Bodi has one of the largest weekly periodic markets in the district, attract-
ing traders from the whole district. Increasingly local farmers have also
become involved in trading activities as a secondary source of income and as
a means of reducing the adverse effects of the seasonal cocoa economy.
Nevertheless, business activities are relatively low outside the weekly market
day. The market in Bodi does not function as a market for locally produced
foodstuffs, but rather for non-local foodstuffs imported from outside the
district. This is primarily due to the fact that the majority of land that falls
under the Bodi stool has been cleared for cocoa cultivation, resulting in a
minimum of food production and that largely for subsistence. In fact, the
market is primarily based on trade in a variety of non-farm products. There is
30 Michael Helt Knudsen
some involvement in service activities, such as people running bars, hairdress-
ers and communication centre operators, while production activities are
limited to the processing of palm oil and some minor blacksmithing and
carpentry businesses. A small health clinic was set up by the local Anglican
Church in 1972, but today it is ill-equipped and only served by a local midwife
subsidised by state funds. Bodi has been designated by the Ghana Cocoa
Board (COCOBOD) as the location of storage and administrative facilities
for the Licensed Buying Companies (LBCs) operating in the area.12 This has
resulted in some economic spin-off activities in the settlement. In addition, a
few public schools, a police station and the provision of electricity were
mentioned by key informants as the only public services in the settlement.13
The settlement is located at the intersection of several of the larger rough
roads running through the district, making the settlement relatively easy to
access.
Amoaya is one of the nearby rural settlements that depends on the limited
services, shops and market provided in Bodi. The settlement is located
around 30 minutes’14 rough drive west of Bodi, access to it being restricted by
a poor road network. Amoaya is an even older indigenous settlement than
Bodi. The settlement’s stool chief, also called ‘Amoayahene’, provides land
to the indigenous population, and during the 1980s–1990s he also did this for
a number of migrant farmers. The latter circumstance may explain why,
according to the fieldwork survey, more than 25 per cent of the households in
Amoaya consist of migrants (Table 3.1). Since the early 1990s, the stool has
been short of land, which means that hardly any land is distributed. Today,
the majority of households in Amoaya are involved in, and highly dependent
on, cocoa farming (Illustration 3.1) (Table 3.2). Only a few households are
engaged in non-farm activities as a primary source of income, though they
are acquiring some importance as a secondary source of income. The small
market in Amoaya has only limited infrastructural services, and trading activ-
ities provide few sources of alternative employment. The lack of access to
land and limited opportunities within the non-farm sector have resulted in an
increasing number of young people leaving the settlement to search for jobs
in other places. As nearby Bodi only offers limited opportunities for employ-
ment, the young people most often choose to migrate to the district capital,
Juaboso, or further afield, to larger urban areas outside the district. This out-
migration is probably one of the main reasons for the low population growth
in Amoaya as compared to Bodi (Table 3.1).

Bonsu Nkwanta and Kefass: young migrant settlements

Bonsu Nkwanta was established in 1970 by migrants who obtained land from
the Boisan stool. For the last two decades, the settlement has developed from
being an agglomeration of a few hamlets into what can be defined as a small
town and the most thriving market town in the district. The first migrants in
the settlement acquired large areas of farmland, and today they are among
 Small town development in the Ghanaian cocoa frontier 31

Illustration 3.1 Labourers opening cocoa pods in Amoaya.

Photo by Michael Helt Knudsen

the largest and wealthiest landowners in the district. The most significant
factor in the development of Bonsu Nkwanta into a centre for trade was the
establishment of a market in 1985 based on foodstuffs cultivated on infant
cocoa farms to provide shade for the cocoa seedlings. When cocoa production
increased, the market simultaneously expanded and encouraged settlers to
engage in trade, including in non-farm products. A decade earlier, a timber
company constructed a road to the area, providing relatively good access to
the newly established market. As the settlement has since grown in size and
importance more roads have been constructed, so that today it is located at
the intersection of several larger rough roads running through the district.
The growing market started to attract migrants coming with the sole pur-
pose of engaging in trading and service activities. For the last decade or so,
these migrants have dominated migration to Bonsu Nkwanta. The high influx
of migrants to engage in non-farm activities explains the fact that more than
half of the households in the fieldwork survey earned their primary source of
income from non-farm activities in 2006 (Table 3.2). The importance of non-
farm activities is visible when business life during the cocoa season is at its
highest. The settlement then becomes crowded with traders and small shops,
serving the farmers when they are paid for their harvests. The non-farm
sector is primarily driven by all sorts of trading activities, mainly trading
32 Michael Helt Knudsen
in non-farm products, that are often bought in the major urban areas of
Kumasi, Accra and Sekondi-Takoradi. The high influx of traders and customers
to take advantage of the more lively economy in Bonsu Nkwanta further
spurs trading activities. Likewise, commercial service activities benefit, such
as hairdressers, seamstresses, bar-owners, guest houses and prostitution.
However, as soon as the cocoa season reaches its end, business life slowly
winds down, migrant settlers and seasonal labourers leave for their home
towns, and traders relocate to more lucrative areas.
The level of public services is similar to that in Bodi, although in 2003
Bonsu Nkwanta was provided with a health clinic sponsored by the
Ghanaian state. Like Bodi, Bonsu Nkwanta is also a designated centre for
LBC administration and the storage of cocoa for a large rural hinterland.
Due to its many non-farm activities and level of services, Bonsu Nkwanta
is generally perceived to be more ‘urban’ than Bodi by both visitors and the
people living there. Indeed, while 58 per cent of the population in Bonsu
Nkwanta stated that they lived in an urban area, only 3 per cent of residents
in Bodi felt that they lived in an urban area.
Kefass was established around the same time as Bonsu Nkwanta, also
driven by the influx of migrants, who were able to obtain land from the
Boisan and Amoaya stools. In the mid 1970s the inhabitants established
a market that quickly grew in size and became significant in the area.
Together with the increasing production of cocoa, the market attracted more
settlers, and the settlement started to grow. Until the end of the 1980s Kefass
was reported to be larger than nearby Bonsu Nkwanta, but soon afterwards
it started losing its market share to Bonsu Nkwanta, whose market started to
grow. Today the market in Kefass is very small and only active one day a
week, while most inhabitants do their shopping and trading in Bonsu
Nkwanta. This leaves the settlement with a low level of non-farm activities
throughout most of the year. However, during the peak of the cocoa harvest
season, bars are lively and trading activities intensify. Since the high influx of
migrants to Kefass in the late 1980s and early 1990s, the immigration rate has
decreased dramatically, though it is still contributing to population growth in
the settlement. The growth of Kefass has been significantly restricted by poor
infrastructural development. Indeed, there is only one road leading to Kefass,
which ends there. The population of Kefass is dependent on a range of
services in Bonsu Nkwanta, like the health clinic, bank and police station, as
these are not present in the settlement. Furthermore, there is no electricity,
communication centre or mobile phone coverage in Kefass. In addition,
young people often choose to leave the settlement, as they cannot obtain land
for cocoa farming, refuse to work as farm labourers or cannot find work in
the non-farm sector. Many of them decide to commute to Bonsu Nkwanta
for work and/or leisure, or even to move there permanently.
 Small town development in the Ghanaian cocoa frontier 33
Diverse trajectories of frontier settlement development
The primary driver for settlement growth in Juaboso District since the 1960s
has been the presence of vacant forest land and the booming production of
cocoa. The indigenous settlements of Bodi and Amoaya were already well
established before cocoa came along, but the introduction of the crop stimu-
lated economic life in the settlements and attracted many households that
previously lived on their individual farms around these settlements. On the
other hand, the establishment and growth of the two migrant settlements of
Bonsu Nkwanta and Kefass can be seen as a direct result of migrants settling
to acquire new land and engage in cocoa farming. Later, the migrant settle-
ments grew due to the high number of labour migrants wishing to take up
employment on cocoa farms or, since the late 1990s, to engage in the growing
non-farm sector. As most purchases from, and investments in, the non-farm
sector are derived from the production of cocoa, the cocoa sector can be said
to have continuously contributed to settlement development in the district.
However, the descriptions of the four settlements above clearly indicate that
settlements on the Ghanaian cocoa frontier have grown up rather differently.
In the case of the two small towns, this has resulted in them having different
functions for their rural hinterland. This section will focus on analysing the
settlements in relation to a range of interrelated factors that affect settlement
growth and small-town functions. These factors include location, access to
land and mobility patterns among households.
Location is one of the central reasons why some settlements on the frontier
grow more than others. The fact that both Bodi and Bonsu Nkwanta are
relatively easy to access in terms of location, road infrastructure and trans-
port has been a major factor in the growth of these settlements into small
towns. Indeed, their location has resulted in Bodi and Bonsu Nkwanta
becoming ‘natural’ centres and markets for trade and services. In contrast,
the ‘isolated’ location and lack of sufficient road infrastructure in Amoaya
and Kefass have resulted in limited transport activity to these settlements and
only a few activities taking place outside agriculture. Furthermore, transport
fares to these settlements are exorbitant by comparison with those to the two
nearby small towns.
It is also due to Bodi and Bonsu Nkwanta’s relatively favourable location
that they have been designated by COCOBOD for the location of storage and
administrative facilities for the Licensed Buying Companies (LBCs) operat-
ing in the area. The presence of the LBCs has improved employment possi-
bilities and made the two small towns nodal points for the transport of cocoa
beans, thus bringing both people and goods to and from the small towns.
Furthermore, it has spurred the local economy, as LBC employees and vis-
itors require lodgings, daily provisions of food and drinks as well as other
goods. As the limited processing of the harvested cocoa pods (opening,
fermenting and drying) is done on the farms or by individual households and
not in a central process facility located in the small towns, other economic
34 Michael Helt Knudsen
spin-off activities directly related to the crop are relatively limited. In
Amoaya and Kefass, the cocoa sector is represented only by purchasing
clerks buying cocoa from small sheds. The economic spin-off activities
related to this and benefiting the rural settlements are minimal.
Households’ access to land also plays a central role in how settlements on
the frontier develop and the function they acquire for their rural hinterland,
as this directly affects livelihood opportunities. Two main developments
can be identified on the Ghanaian cocoa frontier. First, the livelihoods of
the indigenous households in Bodi and Amoaya are strongly dependent on
income from cocoa and therefore on access to land, while involvement in
non-farm activities almost exclusively takes place as a secondary source of
income. The main reason for this livelihood pattern is that the indigenous
population are the traditional rulers of land under the Bodi and Amoaya
stool lands and have continually been provided with good access to land for
cocoa farming. Even though land resources have been largely exhausted
under the two stools, it seems that the upcoming generation of farmers are
ensured land through inheritance. Hence, the low involvement in non-farm
activities in Bodi and Amoaya can be explained by the fact that it is still
possible to make a living from the land (cocoa). This livelihood pattern is
supported by the organisation of the Ghanaian cocoa sector, i.e. the provision
of nearby, reliable markets and fixed prices for cocoa produce. This has
enabled farmers to carry on with the production of cocoa while adding a
second source of income to their portfolio (Knudsen, 2007).
The relatively low involvement in non-farm activities in Bodi has had
consequences for the functions of this small town for its nearby rural settle-
ments, including Kefass. The weekly market day in Bodi is important for both
the population of Bodi and the nearby rural settlements as a market for
imported foodstuffs and all sorts of non-farm products. However, the pres-
ence of the periodic market has neither spurred non-farm activities outside
the weekly market day, nor boosted employment possibilities or development
considerably. Bodi has important functions in relation to the provision of
health services, legal/police matters and education. Nevertheless, as Bodi
only offers basic services within these fields, inhabitants in Bodi and in the
surrounding rural area are forced to travel to the district capital, and often
outside the district altogether if their needs are greater.
The difficulties for migrants in accessing land in Bodi, combined with the
limited employment opportunities within the non-farm sector, as well as the
low level of private and public services, have hampered immigration to Bodi
both from outside the district and from nearby rural settlements like Amoaya.
Indeed, apart from the weekly market day in Bodi, daily mobility to Bodi is
low. Longer-term mobility and actual migration from Amoaya to Bodi are
also low, as individuals and households leaving Amoaya most often choose to
travel to the district capital or further afield to look for employment and
business opportunities, as these are not present in Bodi.
Secondly, looking at the migrant settlement of Bonsu Nkwanta, there is a
 Small town development in the Ghanaian cocoa frontier 35
high occurrence of livelihoods fully or partly based on non-farm activities.
This can be seen as a consequence of two main factors. First, apart from
those migrant households that arrived early where land was in abundance
and was easy and cheap to acquire, access to land has been restricted for
migrant households that arrived later. Therefore, these later households have
been ‘pushed’ to establish livelihoods partly or fully based on non-farm activ-
ities. Secondly, the development of Bonsu Nkwanta into a regional trading
centre has made it attractive for settlers there to engage in a variety of non-
farm activities and employment (Illustration 3.2). Furthermore, this devel-
opment has ‘pulled’ migrants to Bonsu Nkwanta with the sole purpose of
engaging in non-farm activities. Both of these ‘push’ and ‘pull’ factors have
intensified in the last two decades due to the exhaustion of land resources, the
immense growth in population and the level of non-farm activities in the
small town. Involvement in non-farm activities among households in Kefass
is rather low compared to Bonsu Nkwanta, but higher than in the indigenous
settlements. It is especially Kefass’s isolated location that has hampered the
development of the local market and service activities there.
While the level of public services is more or less the same in Bodi and
Bonsu Nkwanta, the presence of commercial services shows a more diverse
picture. During the cocoa season, the provision of commercial services

Illustration 3.2 The main street running through Bonsu Nkwanta where settlement
growth has taken place since the late 1990s.
Photo by Niels Fold
36 Michael Helt Knudsen
increases drastically in both Bodi and Bonsu Nkwanta as the purchasing
power among cocoa farmers is relatively high, and due to the large influx of
seasonal migrants. Despite the two small towns having more or less the same
number of settlers, Bonsu Nkwanta attracts a significantly higher number of
seasonal migrants, customers, traders and commuters than Bodi and thus
has a much greater demand and (seasonal) population basis for commercial
services.
Households’ mobility patterns are closely related to their involvement in
the non-farm sector and thus significant for settlement development and
small-town functions. The low involvement in, and presence of, non-farm
activities in Bodi and Amoaya are both the result of, and result in, the low
mobility of the indigenous population. Indeed, most mobility among
indigenous households in Bodi takes place either within the district or to
neighbouring districts and is mainly related to income-generating activities
(daily travel to cocoa farms, walking or cycling) and social visits. Trips to
larger urban areas are rare and most often related to health or legal matters.
While households in Bodi have more mobility on a daily and weekly basis
due to the town’s location at the intersection of several larger roads and
access to transport, households in Amoaya are more mobile in relation to
longer-term mobility and actual migration.
The migrant households in Bonsu Nkwanta and Kefass, on the other
hand, are more mobile. Looking at Bonsu Nkwanta, the flourishing non-
farm sector and the high involvement in these activities of households are
indeed a result of the high mobility of members of the migrant households.
In contrast to the indigenous households in Bodi and Amoaya, the migrant
households in Bonsu Nkwanta show a remarkably high degree of mobility on
both a seasonal and monthly/weekly basis. Outside the cocoa season, most
migrants travel to ‘home towns’ to help out on family land and to visit family
and friends. On a weekly and monthly basis, migrants travel to larger urban
areas to visit children at school there, for other social visits (funerals, mar-
riages, etc.) or to attend services that cannot be found in Juaboso District.
This mobility pattern has enabled many migrants to engage in the non-farm
sector as they purchase goods on their travels, mainly to the larger urban
areas,15 to resell in Bonsu Nkwanta. As the migrant households increase their
involvement in non-farm activities, travels to larger urban areas for buying
goods, materials and tools correspondingly intensify. Hence, migrant house-
holds’ mobility patterns are both a reason for, and a consequence of, their
involvement in the non-farm sector.
For the migrant households in Kefass, however, this mobility pattern has
not resulted in a strong non-farm sector. The main explanation for this is
Kefass’s isolated location. Its restricted access has hampered the development
of the local market and of other trading and service activities. Furthermore,
settlers in Kefass reported having better access to land for cocoa farming than
is the case in Bonsu Nkwanta. Kefass therefore seems not to have been
‘pushed’ to diversify or shift livelihoods away from farming to the same extent.
 Small town development in the Ghanaian cocoa frontier 37
Bonsu Nkwanta has continuously managed to attract migrants both from
outside the district and region and from within the district. Initially this
was due mainly to opportunities within cocoa production, but also, and
increasingly so, due to the vibrant non-farm sector, which offers both business
and employment opportunities. Within the district, Bonsu Nkwanta attracts
many young people from the nearby rural settlements in search of jobs,
business opportunities or goods. Furthermore, the fact that Bonsu Nkwanta
offers mobile phone coverage, call centres, electricity, lively bars and a pulsat-
ing daily life, with people and goods travelling in and out of the settlement, is
indeed also an explanatory factor for why young people especially commute
or resettle there. While it may be an exaggeration to suggest that Bonsu
Nkwanta is a pool for rural migrants, it is certainly attractive for settlers in
nearby rural settlements. This is also the case for young people in nearby
Kefass, who demonstrate a noteworthy degree of mobility by travelling to
Bonsu Nkwanta often several times a week for leisure to take advantage of
the services offered there. Also, commuting for reasons of employment or
one’s own business activities within the non-farm sector in Bonsu Nkwanta is
intensifying. Some people from Kefass choose to move permanently to Bonsu
Nkwanta, but due to the relatively short distance between the two settle-
ments, most potential migrants choose to reside in Kefass due to lack of
financial resources with which to build a house or rent a room in Bonsu
Nkwanta and/or due to their having their main networks (family and friends)
in Kefass. Clearly the linkages between Bonsu Nkwanta and Kefass are
stronger than those between Bodi and Amoaya. This is not only a matter of
distance, but also of the diversity of the market, levels of services and
employment possibilities offered in Bonsu Nkwanta.

Concluding remarks
This chapter has provided empirical evidence and insight into why some
settlements in the Ghanaian cocoa frontier grow into small towns, while
others stagnate and hardly change in size and functions. First and foremost,
the chapter has demonstrated the significant development of small towns in a
predominantly rural area, far away from the major urban centres of Ghana,
and in settlements not laid out as district capitals. The chapter has identified
three main factors affecting settlement development and small town func-
tions: (1) location; (2) access to land; and (3) mobility patterns. The latter two
factors are closely related to the nature of the population (indigenous versus
migrants) and directly affect livelihood opportunities. Due to a combination
of these main factors, settlements in the frontier have developed differently.
For the two small towns, these factors have resulted in them having different
functions for their rural hinterlands in terms of market, access to services,
non-farm business opportunities and employment. The different functions
have implications for each small town’s ability to pull migrants from nearby
rural settlements and sustain them.
38 Michael Helt Knudsen
Looking specifically at small towns, an obvious conclusion is that neither
the development of small towns nor the function they have for nearby rural
settlements can be applied universally, as these are determined by a series of
factors that are particular to each small town. This suggests a need to analyse
small-town development and functions on the settlement level, as small towns
show great diversity, even within a limited geographical area with the same
economic base, as is the case in Juaboso District. This conclusion is in line
with those of other scholars, who have emphasised that the role of small
towns in rural development is not simply positive or negative, and who reject
generalisations and rather focus their analyses on what is specific and particu-
lar to each town and its rural hinterland (Hardoy and Satterthwaite, 1986;
Baker, 1990; Pedersen, 1997; Tacoli, 1998; Satterthwaite and Tacoli, 2003;
Owusu, 2008).
Most policy and planning in Africa is still inspired by the positive view of
the role of small towns in rural development, which is embedded in the
decentralisation programmes that have been implemented across Africa since
the 1990s (Satterthwaite and Tacoli, 2003). Furthermore, in Ghana, the focus
on the decentralisation of power and resources to lower levels of the urban
hierarchy seems to have exacerbated the focus on district capitals for the
location of public service investments. Consequently, small towns like Bodi
and Bonsu Nkwanta have been largely overlooked by official regional plan-
ning policies in Ghana, despite their potential importance for their large rural
hinterlands in providing access to markets, education and health services, as
well as income-generating opportunities. Bonsu Nkwanta is already fulfilling
some of these roles and could become an important focus for rural develop-
ment. However, to succeed in this, regional planning policies must acknow-
ledge and support the developments that are already taking place there.

Notes
1 COCOBOD granted licences to 19 private companies in the 2007–08 crop season
(Quartey, 2007).
2 PBC is also obliged to act as ‘the buyer of last resort’, thus guaranteeing all cocoa
farmers an outlet for their production.
3 In 2000, these two cities accounted for about 34 per cent of the total urban
population (GSS, 2002).
4 The survey covers households that live permanently in one of the four settle-
ments, and thus it does not do justice to the large numbers of seasonal migrants
entering the district in the harvest season.
5 This figure should be treated with caution, as it is based on the Population and
Housing Census of 2000 (GSS, 2002). Demographic development in the district
since then has undoubtedly resulted in a higher proportion of the population in
the district residing in areas which today would be defined as urban.
6 The non-farm ‘sector’ is defined as encompassing all activities outside the agri-
cultural sector, including services and manufacturing related to the transformation
and processing of agricultural produce, as well as non-related services and manu-
facturing activities, but excluding wage or exchange labour on other farms.
 Small town development in the Ghanaian cocoa frontier 39
Furthermore, it includes all forms of work taking place in a variety of locations
(Satterthwaite and Tacoli, 2003).
7 An indigenous household is defined here as a household that belongs to one
of the resident local Sefwi matrilineages, meaning that it has an inherited right
of access to land for farming (Awanyo, 1998). Correspondingly, a migrant
household that does not belong to any of the Sefwi matrilineages does not have
any such right.
8 Distance from the district capital of Juaboso to Sekondi (regional capital),
350km; to Kumasi, 225km; and to Accra, 495km. Transportation to these urban
areas mainly takes place by minibuses (so-called ‘tro-tros’) or share taxis. Only
a few households in the research settlements own a car for private use.
9 According to the 2000 Population and Housing Survey (GSS, 2002).
10 According to Boni (2005: 127), a ‘stool’ can be defined as: ‘symbolic artefacts
charged with religious and political significance, managed by stool-holders, mostly
chiefs, chosen from amongst the members of the matrilineage’. The land belong-
ing to the stool, or in practice to the matrilineage as a group, is termed ‘stool land’.
Juaboso District is covered by four stool-lands, i.e. Boisan (Bonsu Nkwanta and
Kefass), Bodi (Bodi), Amoaya (Amoaya and Kefass) and Benchima.
11 Mainly migration from small ‘villages’ located on the cocoa farms on the stool-
land. These ‘villages’ usually consist of not more than two to five households.
12 Most LBCs are represented in Bodi and Bonsu Nkwanta.
13 There is no public provision of water in Bodi, nor in any of the other research
settlements. Water is fetched from a few privately sponsored boreholes (mainly
sponsored by NGOs and LBCs) and from nearby rivers.
14 During the rainy season, transport time can be up to an hour.
15 Most transport to Juaboso District from other regions of Ghana passes through
Ghana’s second largest city, Kumasi, where the majority of goods traded in Bonsu
Nkwanta are bought.

References
Amoah, J. E. K. (1995) Development of consumption, commercial production and
marketing, Cocoa outline series 1, Accra: Jemre Enterprises Limited.
Awanyo, L. (1998) ‘Culture, markets, and agricultural production: A comparative
study of the investment patterns of migrant and citizen cocoa farmers in the
Western region of Ghana’, Professional Geographer, 50 (4): 516–30.
Baker, J. (ed.) (1990) Small town Africa: Studies in rural–urban interaction, seminar
proceedings No. 23, The Scandinavian Institute of African Studies.
Boni, S. (2005) Clearing the Ghanaian forest: Theories and practices of acquisition,
transfer and utilization of farming titles in the Sefwi-Akan area, Legon, Accra:
Institute of African Studies, University of Ghana.
Bulíř, A. (2003) ‘Can price incentive to smuggle explain the contraction of the cocoa
supply in Ghana?’, Journal of African Economies, 11 (3): 413–39.
Fold, N. (2002) ‘Lead firms and competition in “bi-polar” commodity chains: Grinders
and branders in the global cocoa-chocolate industry’, Journal of Agrarian Change,
2 (2): 228–47.
Fold, N. (2004) ‘Spilling the beans on a tough nut: Liberalisation and local supply
system changes in Ghana’s cocoa and shea chain’, in A. Hughes and S. Reimer
(eds), Geographies of commodity chains, London: Routledge.
Ghana Statistical Service (GSS) (2002) 2000 Population & housing census: Special
report on the 20 largest localities, Accra: Ghana Statistical Service.
40 Michael Helt Knudsen
Ghana Statistical Service (GSS) (2005) Population data analysis report. vol. 1: Socio-
economic and demographic trends, Accra: Ghana Statistical Service.
Gockowski, J. (2007) ‘Cocoa production strategies and the conservation of globally
significant rainforest remnants in Ghana’, ODI presentation, Sustainable tree
crops program, International Institute of Tropical Agriculture, Accra, Ghana.
Hardoy, J. E. and Satterthwaite, D. (eds) (1986) Small and intermediate urban centres:
Their role in national and regional development in the third world, London: Hodder
and Stoughton in association with the International Institute for Environment and
Development.
Hutchful, E. (2002) Ghana’s adjustment experience: The paradox of reform, United
Nations Research Institute for Social Development (UNRISD), Oxford: James
Curry.
Juaboso District Assembly (2006) Three year medium term development plan:
2006–2009, January 2006.
Knudsen, M. H. (2007) ‘Making a living in the cocoa frontier, western Ghana:
Diversifying incomes in a cocoa economy’, Danish Journal of Geography, 107 (2):
29–42.
Leitner, J. and Harding, S. (2004) ‘Trinidad, Brazil, and Ghana: Three melting
moments in the history of cocoa’, Journal of Rural Studies, 20: 113–30.
Mikell, G. (1989/1992) Cocoa and chaos in Ghana, Washington, DC: Howard
University Press.
Owusu, G. (2005) ‘Small towns in Ghana: Justification for their promotion under
Ghana’s decentralization programme’, African Studies Quarterly, 8 (2): 48–69.
Owusu, G. (2008) ‘The role of small towns in regional development and poverty
reduction in Ghana’, International Journal of Urban and Regional Research, 32 (2):
453–72.
Pedersen, P. O. (1997) Small African towns: Between rural networks and urban
hierarchies, Aldershot: Avebury.
Quartey, E. (2007) ‘Paper for COCOBOD presentation’, unpublished paper, Research
Department, COCOBOD.
Satterthwaite, D. and Tacoli, C. (2003) The urban part of rural development: The role
of small and intermediate urban centres in rural and regional development and
poverty reduction, Rural–urban working paper 9, London: IIED.
Tacoli, C. (1998) ‘Rural–urban interactions: A guide to the literature’, Environment
and urbanization, 10: 147–66.
Vigneri, M. and Santos, P. (2007) ‘Ghana and the cocoa marketing dilemma: What
has liberalization without price competition achieved?’, ODI project briefing 3,
December 2007.
US Department of State (2008) ‘Background Note: Ghana’ www.state.gov/r/pa/ei/
bgn/2860.htm, accessed 15 December 2008.
4 Living from coffee in Vietnam’s
Central Highlands
Susceptible livelihoods and
diverse settlement
transformations
Jytte Agergaard

Introduction
Vietnam’s economic reform process was manifested in the Doi Moi policies
of 1986, which in the late 1980s led to the dismantling of collective farming
and the expansion of family farming. In the 1990s, these policies were
accompanied by a break with the principle of regional self-sufficiency in rice
production and the stimulation instead of agro-forestry production directed
towards export. The expansion of coffee production is one example of this
development. In 2000, Vietnam reached its current position as the second
largest exporter of coffee in the world and at the same time became the largest
exporter of the ‘Robusta’ variety. This position was gained through a dram-
atic increase in coffee production during the 1990s. However, the economic
boom in the global coffee market in the 1990s became a bust at the turn of
the century, resulting in a far-reaching crisis in Vietnam’s coffee sector.
Around 60 per cent of coffee in Vietnam is produced in one single province,
Dak Lak Province, located in the Central Highlands of Vietnam. Coffee is
mainly produced by smallholders and is the dominant crop and source of
income in the central part of the province (‘the coffee belt’) and to a lesser
extent in the rest of the province. The area of land under irrigated coffee
cultivation has increased ten-fold since 1975 (D’haeze et al., 2005), especially
since 1995 (Agergaard et al., 2009). This remarkable rural expansion has
changed the Dak Lak landscape significantly and is visible in the ‘coffee belt’
as a near absence of natural tree cover, except for the rubber tree plantations
and uncultivable pockets of land. Closely linked to this development is more
than 30 years of continuous immigration by Vietnamese households, primar-
ily from the north. The resulting population growth has mainly been absorbed
in rural settlements and urbanisation has remained scattered, except for the
provincial capital and the district headquarters which have grown as centres
for administration, trade and services.
The main objective of this chapter is to explore how settlement transform-
ations relate to the dramatic changes in the coffee sector and how these
transformations and the coffee crisis have impacted on livelihoods in the
42 Jytte Agergaard
coffee settlements. The chapter draws on data collected in four settlements,
all affected by coffee expansion but characterised by different timings in
respect of immigration and/or (re-)settlement and, related to that, different
opportunities for benefiting from inclusion in the coffee sector. The chapter
starts with a short introduction to Vietnam’s role as a coffee producer, fol-
lowed by an outline of the policies and practices of immigration and resettle-
ment in Dak Lak. After describing settlement formation in the four survey
communes, the chapter turns to analyse the links between livelihood and
mobility dynamics in relation to settlement transformation. The conclusion
addresses how the coffee crisis impinges on livelihoods and how differences in
people’s ability to resist crises, reflecting their different locations and degree
and timing of their inclusion in the coffee sector, have affected urbanisation.

The coffee sector in Vietnam: boom, bust and slow recovery

Coffee was introduced as a plantation crop during the French occupation, and
its production continued partially during the reign of Dîem’s Republic of
Vietnam (Southern Vietnam, 1954–1975). After the unification of Vietnam,
coffee plantations were transformed into state collectives, and production was
improved with the aid of East German expertise and finance, turning Vietnam
into an exporter of coffee to its trading partners in the Eastern Bloc. By the
turn of the century, Vietnam had reached its current position as the world’s
second largest producer of coffee. In 2006 Vietnam produced 850,000 metric
tons of coffee, a figure exceeded only by Brazil, whose production of
2.59 million metric tons represents almost one third of the world’s production
(FaoStat, 2008). There are at least three reasons why Vietnam experienced
this rapid growth in coffee production (Ha and Shively, 2008). First, the policies
of privatisation and liberalisation (Doi Moi) that opened up smallholder
production eased imports of fertilisers, etc. and paved the way for the inter-
national marketing of the product. Second, state-sponsored migration from
the population-dense regions of the north for settlement in the Central
Highlands set the exceptional expansion in coffee-growing areas in motion
(see next section). Third, price increases in the world market, generated by a
substantial decrease in coffee supply due to frosts in Brazil, motivated
migrant and indigenous households to engage in coffee production.
Most of Vietnam’s coffee is of the ‘Robusta’ variety. Traditionally ‘Robusta’
was almost exclusively used as an ingredient in instant coffee, but due to
improvements in technological processing during the 1990s, roasters have
increasingly been prepared to include more ‘Robusta’ coffee in their blends
(World Bank, 2004). Accordingly, the demand for the much cheaper Robusta
coffee has increased. More than 90 per cent of Vietnam’s coffee is exported,
primarily as green beans, since roasted coffee is not suitable for long-distance
transport. After harvesting the coffee cherries, the most preliminary process-
ing takes place on the farm. Before the beans are collected by local assemblers
or buyers, most farmers dry their beans in the sun, and if they have access to
 Living from coffee in Vietnam’s Central Highlands 43
machinery, they also remove the green coffee beans from the shell (hulling) (de
Fontenay and Leung, 2002). The poorer segment of coffee farmers, however,
sells coffee cherries directly to the assemblers as ‘wet coffee’. Hence, only very
basic processing of the coffee takes place in Vietnam, while nearly all value
added in relation to roasting is taking place close to the market in the Global
North. At the producer level, minimal grading takes place, meaning that
settling prices mainly vary with seasonal fluctuations in coffee prices. Until
the late 1990s, state-owned enterprises (SOEs) functioned as the exclusive link
between the international coffee companies and local collectors commis-
sioned by the SOEs. However, in relation to liberalisation policies in the
1990s, international traders were invited to operate independently. Accord-
ingly, local collectors and buyers can now trade with both the international
traders and SOEs. Accordingly, the role (and the responsibility) of SOEs has
been decreasing.
Between 1993 and 2000, Vietnam’s coffee-producing area increased by
about 24 per cent per year, resulting in an average rise in production of about
20 per cent per year (Oxfam, 2002). This exceptional expansion in coffee
supply impacted greatly on world coffee-price fluctuations and eventually
resulted in a dramatic decrease in prices between 1999 and 2002, where prices
dropped by almost 75 per cent from the 1990s average of 180 US cents per kg
(Agergaard et al., 2009). While ‘fluctuating fortunes for smallholders is a
phenomenon neither unique to coffee nor Vietnam’ (Ha and Shively, 2008:
312), the fact that the dramatic decline in prices coincided with the exceptional
growth in outputs from Vietnam means that the country quickly found itself
in the middle of a coffee crisis. Although by the end of the decade coffee was
providing as much as 10 per cent of Vietnam’s annual export earnings (Ha and
Shively, 2008), national effects were modest, the crisis being much more pro-
found on the provincial, local and household levels (World Bank, 2004). Dak
Lak Province in particular was affected by the crisis in respect of: (1) a fall in
coffee revenues combined with increased government expenditure (support
for debt moratoriums and tax reductions); (2) a partial dismantling of ser-
vices, as the government did not have funding to replace the services in health
and education previously cared for by the coffee SOEs; and (3) post-crisis
changes in the business environment, for example, reduced credit availability
and a more modest use of purchased inputs, resulting in declining yields.
Since 2003, settling prices have slowly been increasing and stabilised at
around 65–70 US cents per kg. Although the average coffee-growing small-
holder in Dak Lak Province managed to attain incomes higher than house-
holds in other rural regions of Vietnam between 1999 and 2003, the province
and its coffee farmers are still suffering from the crisis (Ha and Shively, 2008).
During the crisis, the government offered direct support to farmers by halting
reimbursements of debts. More official policies to counter the effect of the
crisis have been to reduce the total coffee area, in particular on less suitable
land, and to secure crop diversification by introducing (and supporting) the
cultivation of other perennial crops such as cashew, rubber and pepper.
44 Jytte Agergaard
Before turning to the settlements chosen for the study and the analysis of the
households’ livelihood responses to the crisis and their spatial impacts, the
next section provides an overview of the immigration and resettlement policies.
This is of particular importance in understanding the setting for settlement
formation in Dak Lak Province.

Immigration and resettlement in Dak Lak Province

Since 1975, the population of Dak Lak Province has more than quadrupled
to the current figure of 1,667,000 (Dak Lak Statistical Office, 2004).
Immigration into the province has been continuous, having also taken place
both during French colonial rule and the American war between 1954 and
1975. However, the planned resettlement during the post-war decade (1975–
1985) represented a more systematic and organised effort at relocation. This
signals a new era of immigration to Dak Lak province and was part and
parcel of the government’s plan to extend its New Economic Zone (NEZ)
programmes, already embarked upon in the north. The core of the NEZ
programmes was to relocate households from regions of high population,
where people were competing for agricultural land, to regions with less pres-
sure; originally, the highland areas were by and large occupied by non-Kinh
ethnic groups living in longhouse communities practising shifting cultivation.
The first to move to Dak Lak Province in the late 1970s were young pion-
eers. Then followed the NEZ programmes, where families were offered the
opportunity to resettle on cooperative land, with the right to occupy a piece
of land for housing and gardening. Important instruments in pushing (in
some cases forcing) rice producers from the north in particular to move to the
highlands were economic incentives and privileges for resettling households,
as well as the household registration system (see note 2 in Chapter 6 (p. 88) for
an explanation). However, planned resettlement proved slow, and many reset-
tled households became dissatisfied with life on the frontier and returned to
their home villages (Dang et al., 1997; Hardy, 2000, 2003). Since the mid
1980s, settlement not linked to the NEZ programmes has caught up with
the planned resettlement, in particular in the first half of the 1990s when
state intervention in farming was reduced further and coffee became acknow-
ledged as a promising crop. Despite crisis in the coffee sector, Dak Lak
Province continues to attract new settlers.
The formidable tide of immigration has changed the ethnic composition of
the province. Today, indigenous minorities such as the Ede make up only
20 per cent or less of the total population of the province, compared to
47 per cent in 1975 (Oxfam, 2002). At settlement level, this has led to immi-
grants buying up land, coupled with growing landlessness or near landless-
ness on the part of indigenous households. But of equal importance has been
the post-unification sedentarisation programme, directed towards ‘fixed cul-
tivation and settlement’ of the minorities, whose lifestyle was seen as nomadic
(Salemink, 2003).1 In practice, indigenous households were allocated a piece
 Living from coffee in Vietnam’s Central Highlands 45
of land for habitation and cultivation in and around their hamlets, actually
copying the settlement structure of the immigrants. In the process, old ham-
lets were divided and some households resettled in new locations, ideally to
provide each household with 2 hectares of land near their houses. This dis-
solution of households and communities, and in particular the reduction of
traditional land rights, has contributed to indigenous groups today being
over-represented among the province’s poor households (Oxfam, 2002;
Doutriaux et al., 2008).
Hence, settlement transformations in Dak Lak Province have been greatly
affected by flows of immigrant households, and the planned and spontaneous
resettlement of indigenous households. Also, the high influx of people has
stimulated continuous demarcations, the most profound of these being the
division of Dak Lak Province in 2006. At the local level, these developments
have resulted in a pattern of indigenous groups and immigrants living separ-
ately. Nevertheless, most settlements share the practice of houses being located
in proximity to farming land rather than in village-like clusters (Illustration
4.1). On a regional scale, settlements differ in terms of time of settlement and
consolidation. Accordingly, some settlements were more consolidated in rela-
tion to developments in services, infrastructure and the business environment
prior to the coffee crisis. This will be illustrated in the next section.

Illustration 4.1 Typical coffee settler’s house, located close to the smallholder’s coffee-
growing land.
Photo by Jytte Agergaard
46 Jytte Agergaard
Settlement formation in the coffee belt
The districts of Dak Lak Province can be divided into three categories from
the point of view of their suitability for coffee: (1) highly favourable for coffee
growing; (2) moderately favourable for coffee growing; and (3) unfavourable
for coffee growing (Oxfam, 2002).2 All the districts selected for this study lie
within the first category, referred to as the ‘coffee belt’. By selecting districts
which are highly favourable for coffee production, the aim was to rule out the
major variations in production being due to differences in the overall suit-
ability of the area for coffee growing. Four districts were chosen at differing
distances from the provincial capital: Buon Ma Thout District includes the
provincial capital; Cu M’gar District is adjacent to the provincial capital
district; Krong Buk District lies to the northeast of the capital; and Ea H’leo
District is in the north of the province (Figure 4.1). Within each district, one

Figure 4.1 Map showing location of study communes in Dak Lak Province.
Source: Redrawn from the digital map provided by the Dak Lak Department of Agricultural
Development, Dak Lak Province. The map is not published.
 Living from coffee in Vietnam’s Central Highlands 47
commune was selected for the study according to the timing of settlement
and its ethnic composition (Table 4.1). In each commune, one to three villages
were selected to represent the commune (Table 4.2).
Rural settlement dominates the ‘landscape’ of habitation, urban clusters
being fewer. Clear exceptions are the urban concentration and sprawl of
the district centres. Another exception is the comprehensive growth of the
provincial capital, Buon Ma Thout. Growing affluence in the province has
stimulated all sorts of retail, services and private and public education, and
national and international coffee trading companies have continuously built
offices and storing facilities in the city. In 2005, it became a second-order
urban centre in Vietnam.3 On the other hand, industrial and handicraft
enterprises are almost absent, although since 2005 Buon Ma Thout has for
the first time laid out areas for industrial development. In the following, the
four survey communes are presented according to their settlement character-
istics and urbanisation trajectories.

Table 4.1 Background statistics for the four survey communes, 2004

Ea Tu Quang Hiep Ea Sien Ea Nam

Commune Commune Commune Commune

Households 2731 2310 1639 2259

Population 13,885 11,229 9040 10,726
Number of villages 13 11 16 17

Source: Based on information from People’s Party Committees at district and commune levels.

Table 4.2 Background statistics for the survey villages

Ea Tu Quang Hiep Ea Sien Ea Nam

Commune Commune Commune Commune

Survey villages Ko Tam Hiep Thang Village 1b Village 4

Hiep Tien Village 2a Ea Den
D’lung 2
Approximate age of villages > 100 35 25 15
(years)
Ethnic composition Ede Kinh Nung Kinh
Kinh Ede
Kinh
Immigrant households % (*) 41 100 86 100
Number of households 539 347 356 242
Number of households 182 121 149 122
surveyed

Source: Author’s household survey, 2006.

Note: * Households where head of household was not born in the surveyed commune.
48 Jytte Agergaard
Ea Tu: peri-urban coffee settlement
Ea Tu is located in the north-eastern part of Buon Ma Thout District. It
has developed around an old Ede hamlet and can date its history back more
than a hundred years. Immigration has been a continuous process, dating
back to before the planned resettlement (see Figure 4.2), though the majority
of households are still Ede. Compared to other settlements, many settlers in
Ea Tu have previously lived in other Dak Lak districts. The area is character-
ised by its flat terrain, and more than one third of the houses are either old
or modern versions of the Ede longhouse. Unlike the other settlements,
the density of houses is high and the average size of landholdings low. Feeder
roads into the settlement are paved, and the secondary roads are upgraded
dirt roads of good quality. Four coffee collectors are all well established in the
village, and most coffee smallholders are well serviced with hulling facilities.
Shops selling daily necessities are located along the main roads, which can be
reached by bike in less than 20 minutes by most households. Bus services
to and from the centre of Buon Ma Thout are frequent. Accordingly, it is
possible for both poor and more affluent households to have one or more
household members commuting on a daily basis. Access to health and edu-
cational services is also favourable, as is illustrated by the location of a primary
school in the settlement and secondary schooling located at the border of the
commune to the west. Growing urbanisation is also reflected in (primarily)
foreign aid-induced investments in water and drainage systems. Despite the
relatively high levels of services, the commune spans huge disparities in living

Figure 4.2 Head of household’s year of arrival in Dak Lak Province in cumulative
per centage of total number of households.
Source: Author’s household survey, 2006.
Note: Immigrant households (HH) make up 100% of all households in the Quang Hiep and
Ea Nam samples, while the same figure for Ea Sien is 86% and for Ea Tu only 41%.
 Living from coffee in Vietnam’s Central Highlands 49
arrangements (Table 4.3). Accordingly, Ea Tu can be described as a peri-
urban settlement that is gradually becoming integrated into the Buon Ma
Thout urban development (Illustration 4.2). Despite obvious signs of crisis
in the coffee sector, Buon Ma Thout continued to urbanise after 2000,
which has positively impacted on service provision in peri-urban areas such
as Ea Tu.

Table 4.3 Diversity in housing standards in the four survey communes (% of total
households)

Ea Tu Quang Hiep Ea Sien Ea Nam

Commune Commune Commune Commune

Large house (4+ rooms) 12 5 6 3

Medium house (2–3 rooms) 56 70 60 66
Small house (1 room) 24 19 32 24
Brick/cement 42 17 24 36
Timber walls 58 83 76 64

Source: Author’s household survey, 2006.

Illustration 4.2 The northern entrance to the provincial capital Buon Ma Thout,
which is gradually expanding to include rural coffee settlements on
the outskirts.
Photo by Jytte Agergaard
50 Jytte Agergaard
Quang Hiep: mature pioneer settlement
Quang Hiep has developed around one of the forest farms established shortly
after the unification of Vietnam and consists exclusively of immigrant house-
holds. Pioneers from the Da Nang area were offered various jobs in the new
enterprise, and the commune has continued to attract farmers, particularly
from Da Nang Province. Immigration to Quang Hiep has been gradually
increasing from 1976 onwards, and more than 40 per cent of households had
settled prior to the coffee expansion of the 1990s. The majority of houses are
made from timber, which partly reflects its availability on the forest edges, but
can also be explained by the temporality of houses and land in Quang Hiep.
The fact that more than 50 per cent of households do not formally hold land
certificates has resulted in semi-permanent housing. The road connecting
Quang Hiep to the district centre of Cu M’gar was upgraded and paved
before 2002. This development has resulted in better connections to traders
and services in the district centre and has also stimulated business in the small
trading community, located in proximity to the buildings of the old forest
enterprise now housing the People’s Party Committee of the commune’s
offices. Hence, Quang Hiep’s history as a pioneer settlement, the gradual
immigration and government investments in infrastructure from the mid
1990s onwards, have stimulated the gradual growth of a small local trading
centre that was put in place before the economic recession. Accordingly,
traders and shopkeepers have in most cases managed to survive the crisis,
while small craftsmen and traders in ‘luxury goods’ have either closed or
diversified, waiting for the smallholder economies to recover.

Ea Sien: ‘stretched’ settlement

The centre of the Ea Sien commune is inhabited primarily by Nung settlers
from Lang Son Province in the north of Vietnam. Settlement started in the
late 1980s, when soldiers from Lang Son were stationed in the area. The Nung
community was established around the cultivation of rice and maize, which,
together with poor development in infrastructure, explains why coffee has not
developed as a mono-culture in the area. Subsequently, more settlers have
arrived, many from Lang Son, but gradually also from other districts in Dak
Lak Province. Ede households, for example, have been seeking their fortunes
in the less-populated Ea Sien area. Originally, the northern part of the com-
mune was inhabited by an Ede community that became sedentarised and
divided in response to population growth in the early 1990s. In the early 2000s
the commune’s borders were changed, and Ea Sien became part of Krong Buk
District, whereas most of the commune was previously part of Krong Pak
District. In effect people travel both west and south for shopping. More
recently, the commune has acquired status as a ‘development commune’,
which means that Krong Buk District has committed itself to investments in
infrastructure, schooling and other services. Although the commune centre
 Living from coffee in Vietnam’s Central Highlands 51
has become better connected to its hinterland, the general upturn in coffee
has not stimulated the development of trading and services in Ea Sien. Hence,
coffee has not stimulated urbanisation in the commune, and it is doubtful
whether government investments in infrastructure alone will drive such
development because the stretched location of households will not stimulate
local shopping and trading.

Ea Nam: a recent coffee settlement

Ea Nam, located in the north of the province, is the most recent of the four
communes. It is almost exclusively inhabited by immigrants, the first settlers
having arrived in the mid 1980s. Immigration, stimulated by the high expect-
ations of growing coffee, gained momentum from the mid 1990s onwards.
Poor-quality gravel roads radiate out from the paved north–south road,
meaning that many households have difficulties in reaching the main road
during the rainy season. However, once the main road has been reached,
people have easy access to the small markets that have developed five kilo-
metres to the south and close to the district centre to the north. This relatively
disadvantageous location is compensated for by the more affluent farmers,
who have built an extra house near the road, from where their children can
easily go to school. Coffee is the main activity in the area, and these farmers
have benefited from relatively large coffee plots. However, a majority of coffee
smallholders suffered directly from the falling coffee prices because they were
about to take their first full coffee harvest at the time the prices collapsed.
This is probably one important reason why shopping and trading within the
commune is limited to one trader and two collectors. Consequently, most
farmers depend on travelling collectors and traders, and they also have to
spend more time acquiring daily necessities. Currently, there are no official
plans to upgrade the village infrastructure in general, nor are there any plans
to strengthen service provision. Also, it is unlikely that the local business
environment will flourish in the near future. The development of Ea Nam, with
its high dependence on coffee and young and/or immature institutions, has
suffered from the economic recession. Hence, in spite of Ea Nam’s ‘favourable’
location along the north–south main road, urbanisation has not been stimu-
lated in the same way as in the older settlements of Ea Tu and Quang Hiep.

Livelihoods and mobility in the coffee settlements

The preceding section has illustrated in some detail the considerable diversity
in settlement histories and urbanisation trajectories. While settlement trans-
formations obviously differ, there are also some obvious similarities due to
the fact that the majority of households depend directly or indirectly on the
coffee crop, whether as smallholders or as coffee labourers (Table 4.4). In
the following, some of the characteristics of how livelihoods have come under
pressure during the preceding ten years will be presented.
52 Jytte Agergaard

Table 4.4 The significance of households’ engagement in coffee farming

Ea Tu Quang Hiep Ea Sien Ea Nam

Commune Commune Commune Commune

Smallholder coffee farming 85 75 73 95

households (%)
Employed in coffee farming (%) 18 19 41 21
Average area of smallholding (ha) 0.7 1.5 1.0 1.8
Average area of land under coffee 0.6 0.8 0.5 1.5
cultivation (ha)
Average coffee yield (kg/ha) 1479 1379 1382 1436
Percentage of coffee smallholders 39 37 45 28
with yields < 1000 kg/ha
Percentage of coffee smallholders 28 27 17 18
with yields > 2000 kg/ha

Source: Author’s household survey, 2006.

The majority of households have a low degree of self-sufficiency in food

crops and depend on cash incomes to buy almost all their necessities. The
cultivation of food crops is limited, while commercial agriculture (primarily
coffee, but also cashew, pepper and rubber) predominates. Rice farming is
almost exclusively practised by indigenous households, though seldom in
quantities sufficient to feed the family. Nonetheless food crops are not
neglected, as is illustrated by most farmers’ cultivation of food crops in the
coffee fields during the three to four years between planting coffee and start-
ing to harvest. Currently, most farmers also cultivate green leaves, etc. to
garnish their daily meals on the small pieces of garden land around their
homes. Also, some households manage to cultivate crops like pulses and
cassava on patches of communal land, including ditches and borderlands,
during the spring. However, these outputs are never reported in official data
on farming portfolios.
Another feature typical of the coffee farmers is their almost complete lack
of knowledge of coffee farming previous to their settlement. Most farmers
report that they learned about coffee cultivation through the grapevine, and a
few have participated in training and/or have acquired agronomic knowledge
though books and pamphlets. What most farmers also note is that, although
sufficient water and fertilisers are decisive for acquiring high yields, skills and
knowledge regarding the amount and timing of these additives and the gen-
eral care of the plants are of equal importance (see also D’haeze et al., 2005).
A qualified extension service has been provided through the local Farmers’
Unions. However, membership of these is dependent on membership of the
Party, to which you are appointed; hence, this is not an option for the major-
ity of households. Consequently, most farmers rely on a locally provided
extension service (through the commune) or project and programme officers
 Living from coffee in Vietnam’s Central Highlands 53
visiting the farmers. As one farmer said, expressing his frustration: ‘They
come in a car and say they will help you. Then they drink your tea, chat for
a while, close the door and say goodbye’ (interview, Ea Nam, 2 June 2006).
The importance of producing high(er) yields has grown dramatically since
the coffee prices started their collapse. Hence, farmers have been eager to
improve their outputs in order to compensate for the falls in prices. While the
government response to the crisis has been to try and reduce coffee-growing
areas, forcing coffee farmers with plots of poor soils or poor conditions for
coffee farming to abandon coffee production, the farmers’ response has been
to make investments in irrigation and fertilisers. Often this has resulted in
increased borrowing on top of the loans that farmers have already taken out
to acquire their plots. Hence, the coffee crisis has opened up a debt spiral for
many coffee smallholders and led to increasing dependencies on traders,
who are commonly the ones to offer the loans; often the smallholders have
to provide security in the form of the coming harvest. On top of the difficult
debt economy, many farmers have not been able to benefit from the moder-
ately increasing prices in coffee in the 2000s because their settling prices have
been fixed at a low level. For the poorest households, which could not obtain
loans, the crises have simply resulted in a neglect of additives, which has
produced very low yields as a result. The low yields in the survey communes
can be observed by comparing the average yields of around 1.4 tonnes/ha
(Table 4.4) with the estimated yields of 2.7 tonnes/ha for moderately suitable
coffee land (D’haeze et al., 2005: 67). Given that up to 45 per cent of coffee
smallholders in some communes produce less than 1 tonne/ha, it should be
acknowledged that coffee production is far from optimal.
The fact that smallholder economies have come under pressure is also
illustrated by the increase in the number of households supplementing and/or
building their incomes on employment (the majority doing farm work or
unskilled work; see Table 4.5). Accordingly, the huge influx of labour migrants

Table 4.5 Patterns in income diversity 1996 and 2006: number of households acquiring
their income from various income activities

Ea Tu Commune Quang Hiep Ea Sien Ea Nam

Commune Commune Commune

1996 2006 1996 2006 1996 2006 1996 2006

Smallholder 153 162 104 113 143 147 116 122

farming
Employment 49 102 29 51 40 76 12 41
Trade 8 25 9 21 0 4 2 4
Livestock 3 22 7 26 1 24 0 18
keeping
Sample size 182 121 149 122

Source: Author’s household survey, 2006.

54 Jytte Agergaard
from the north taking up employment during the coffee harvest from late
November to late January, known from the 1990s, has been replaced by the
employment of ‘local’ workers. In 2006, 25 per cent of coffee farmers
employed farm labourers, of whom less than 10 per cent were migrants
(author’s household survey). In Ea Tu and Ea Sien communes, where more
than 50 per cent of households are engaged in farm work or unskilled work, it
is particularly the Ede households that follow this option. In Ea Sien, which
also has the highest number of households employed in coffee, a particular
relationship has developed between more affluent coffee farmers near the
district centre in Krong Buk and poor farmers from Ea Sien.
Keeping livestock is another strategy for diversifying incomes. Farmers
mainly keep chickens, ducks or pigs, while the tradition among the Ede of
keeping buffaloes has almost vanished, not least as a consequence of the
sedentarisation of the indigenous households, with their consequent loss of
extended areas for grazing. However, raising livestock for sale is still rather
limited, as is the propensity of households to engage in trading. As shown
in Table 4.5, trading has mainly developed in Ea Tu and Quang Hiep, but is
almost absent from Ea Sien and Ea Nam due to their limited urbanisation.
Thus, even though livelihood diversification has increased since the mid
1990s, opportunities for income diversification vary.
Diversification in farming is another strategy that has been adopted to
counter dependence on price fluctuations in coffee. In Quang Hiep, concerted
government efforts have been invested in introducing alternative crops such as
pepper and cashew, as is reflected in Table 4.4, with the difference in average
landholdings, especially those under coffee cultivation. While pepper has
shown itself to be an even more precarious crop than coffee, cashew cultivation
has been more successful, and many poor households have been able to change
from coffee to cashew. Likewise, Ea Sien is characterised by diversification in
farming, more as a consequence of its slow inclusion in the coffee sector
than in response to the crisis in that sector. In contrast, Ea Nam is charac-
terised by an almost total absence of crop diversification. More recently,
however, poor farmers have been offered compensation if they drop out
of coffee and plant Melia trees, which might produce good timber in ten
years’ time.

Migration and mobility

In analyses of the livelihood effects of the coffee crisis, the importance of
migration and mobility is usually neglected. Given the high number of
migrant households, remittances could play a significant role as an economic
safety net for coffee households under pressure. However, our data indicate
that there are few regular economic transfers to and/or from family members
living elsewhere. On the other hand, contacts with immigrants’ homes con-
tribute to livelihoods in more subtle ways. Favourable borrowing from family
and kin in relation to household separation is an obvious case. Additionally,
 Living from coffee in Vietnam’s Central Highlands 55
interviews with coffee farmers reveal that, at various points in time, by ‘play-
ing’ the household registration system, they have been able to claim services
and opportunities in both locations. The most striking feature, though, is
educational strategies, where households have been schooling their children at
‘home’, not only in response to poor schooling opportunities in Dak Lak,
but also due to the lack of economic and social resources with which immi-
grants can support their children’s schooling. In Ea Nam, even more flexible
arrangements were observed; some immigrants have been able to ‘commute’
between their coffee farms in Dak Lak and their rice farms in Binh Dinh
Province (8–9 hours away by bus). Due to daily bus services, farmers travel
frequently between their two homes. It is difficult to value the significance
of these multi-local arrangements. However, they clearly privilege certain
groups of migrants, while indigenous households seldom have access to such
safety nets.
Access to daily mobility is of equal importance. For the poorer segment of
households, mobility for work is crucial. Such mobility is supported by public
transportation and/or access to motorised vehicles to overcome the distances
between home and work. As shown in Table 4.6, between 20 per cent and
40 per cent of households do not possess their own means of transportation,
which is particularly problematic in Ea Sien and Ea Nam, where there is only
limited access to public transportation, and from where people have to travel
longer distances not only to work, but also to access services. The need for
transportation has increased in the past five to ten years, as prices in the
local markets have increased. Government subsidies for foodstuffs have dis-
appeared, so local prices have come to reflect actual market prices, meaning
that households in poorly serviced areas such as Ea Sien and Ea Nam have
a stronger incentive to travel to more urbanised markets. Thus either people
are having to spend time and money travelling to buy goods or will have to
spend more money on their local purchases. Hence, in these settlements, the
households that are suffering most from the crisis in the coffee sector are
those lacking proper means of transportation.

Table 4.6 Percentage of households possessing selected means of transportation

Ea Tu Quang Hiep Ea Sien Ea Nam

Commune Commune Commune Commune

Bicycle 74 55 56 36
Motorbike 61 77 57 76
(Xe) may cai 36 25 37 21
Vehicle 6 11 3 5
Having neither motorbike, 32 22 39 20
may cai nor vehicle

Source: Author’s household survey, 2006.

Note: A may cai is a three-wheeled motorised scooter with a detachable trailer.
56 Jytte Agergaard
Conclusion
This chapter has described how, during the 1990s, Vietnam, and in particular
Dak Lak Province, became a giant producer of coffee for the world market
and explained how the role of coffee as the dominant crop made the region
susceptible to fluctuating world market prices. Area expansion in coffee from
the early 1990s has been an extreme generator for literally all household
types in Dak Lak’s coffee belt. This development is closely related to the
post-unification immigration of households from the north. While emphasis-
ing that coffee has probably been the most important driver of overall settle-
ment formation in Dak Lak Province, it has been shown that population
policies, such as planned settlement, sedentarisation and new demarcations,
have strongly impacted on the overall pattern of settlement formation in the
province.
Physical location and timing are important factors in explaining the
marked differences in the settlements’ urbanisation trajectories. It has been
shown that the older settlements have developed small trading centres in
relation to commune centres, while settlements that developed from the mid
1990s onwards have not urbanised much and hardly will in the near future.
This is not least because they will not benefit from the booming coffee
economies that have stimulated pre-crisis market developments in the older
settlements. This situation is greatly affected by the abilities of different
households to cope with economic crisis from the early 2000s onwards. Coffee
smallholdings, established from 1994/95 onwards, have been particularly hit
by the crisis because their first full coffee harvest coincided with dramatic
falls in prices. To counter the effects of the crisis, opportunities for the
diversification of incomes are crucial. In that respect, households in the old
settlements have greater opportunities than in the two younger ones. The
economic crisis and the government’s withdrawal of subsidies have further
disadvantaged small local markets, where prices now reflect transport costs
and limited opportunity costs. Hence, the undeveloped infrastructure in the
young settlements further hampers livelihood diversification and also
impedes market access and service provision.
Migration, multi-locality and mobility are also important livelihood
resources and encompass more than ‘access’ to markets. It has been shown
that many migrant households maintain strong networks with family and kin
in the north, and in some cases stretch their households between the north
and Dak Lak Province. Regular remittances are not significant, but multi-
local practices provide these households with extra resources. However, these
extra resources are not available to most indigenous households, which,
already dispossessed as a consequence of many years of sedentarisation
policies and practices, risk being further disadvantaged.
 Living from coffee in Vietnam’s Central Highlands 57
Notes
1 Competition for land on the frontier was far from new. While the French were
eager to establish plantations, the process of territorialisation of the highlands
between 1954 and 1974 was more mixed. On the one hand, the American presence
was concerned with resettlement of indigenous groups in strategic hamlets. On the
other hand, South Vietnamese nation-building attempts were directed towards the
sedentarisation of indigenous groups (Salemink, 2003).
2 Oxfam developed this classification based on advice from coffee experts in the Dak
Lak Department of Agriculture and Rural Development.
3 Urban centres in Vietnam are classified into six grades: special grade (Hanoi and
Ho Chi Minh City), and grade-I to grade-V urban centres. The grade-II urban
centre is defined as: ‘1) Being an urban center with the function as a political,
economic, cultural, scientific-technical, tourist and service center, traffic hub and
an exchange center of a province, inter-provincial region or in the whole country,
and playing the role of boosting the socio-economic development of an inter-
provincial region or the whole country in several domains; 2) The non-agricultural
labor represents 80 per cent of the total labor force or higher; 3) Having an infra-
structure built in many sectors becoming synchronous and complete; 4) Having a
population of 250,000 or more; 5) Having an average population density of 10,000
people/km2 or more’ (Decree No. 72/2001/ND-CP of October 5, 2001 on the
classification of urban centers and urban management levels).

References
Agergaard, J., Fold, N. and Gough, K.V. (2009) ‘Global–local interactions: Socio-
economic and spatial dynamics in Vietnam’s coffee frontier’, Geographical Journal
(forthcoming), published online, DOI 10.1111/j.1475-4959.2009.00320.x.
Dak Lak Statistical Office (2004) Dak Lak Statistical Province Yearbook 2004, Buon
Ma Thout City: GSO.
Dang, A., Goldstein, S. and McNally, J. (1997) ‘Internal migration and development
in Vietnam’, International Migration Review, 31(2): 312–37.
de Fontenay, P. and Leung, S. (2002) ‘Managing commodity price fluctuations in
Vietnam’s coffee industry’, Working Paper 02-4, International and Development
Economics, Asia Pacific School of Economics and Government, The Australian
National University.
D’haeze, D.D., Deckers, J., Raes, D., Phong, T.A. and Loi, H.V. (2005) ‘Environ-
mental and socio-economic impacts of institutional reforms on the agricultural
sector of Vietnam: Land suitability assessment for Robusta coffee in the Dak Gan
region’, Agricultural Ecosystems and Environment, 105: 59–76.
Doutriaux, S., Geisler, C. and Shively, G. (2008) ‘Competing for coffee space:
Development-induced displacement in the Central Highlands of Vietnam’, Journal
of Rural Sociology, 73(4): 528–54.
FaoStat (2008) online at http://faostat.fao.org/site/567/default.aspx#ancor, accessed
5 November 2008.
Ha, D.T. and Shively, G. (2008) ‘Coffee boom, coffee bust and smallholder response in
Vietnam’s Central Highlands’, Review of Development Economics, 12(2): 312–26.
Hardy, A. (2000) ‘Strategies of migration to upland areas in contemporary Vietnam’,
Asia Pacific Viewpoint, 41(1): 23–34.
Hardy, A. (2003) Red hills: Migrants and the state in the highlands of Vietnam.
Copenhagen: NIAS Press.
58 Jytte Agergaard
Oxfam (2002) The impact of the global coffee trade on Dak Lak Province, Vietnam:
Analysis and policy recommendations. Oxfam Great Britain, Oxfam Hong Kong,
ICARD, available online at http://www.isgmard.org.vn/Information%20Service/
Report/Agriculture/Oxfam-ICARD-Full%20report-final-e.pdf, accessed May
2009.
Salemink, O. (2003) The ethnography of Vietnam’s Central Highlanders: A historical
contextualization, 1850–1990. Honolulu: University of Hawai’i Press.
World Bank (2004) The Socialist Republic of Vietnam: Coffee sector report. The World
Bank Agricultural and Rural Development Department Report No. 29358-VN,
available online at http://web.worldbank.org/external/default/main?pagePK=
51187349&piPK=51189435&theSitePK=387565&menuPK=64187510&search
MenuPK=387606&theSitePK=3875, accessed May 2009.
5 Rise and fall of smallholder
pineapple production in Ghana
Changing global markets,
livelihoods and settlement
growth
Katherine V. Gough and Niels Fold

Introduction
The export of pineapples from Ghana has been one of the success stories for
a country struggling to find alternative sources of foreign capital. Initially
grown by smallholders in the Akwapim Range, production has been increas-
ingly dominated by large-scale pineapple farms located to the west. This shift
has coincided with the development of a new type of pineapple (MD2) which
has come to dominate the global pineapple market. Smallholders have
experienced the bottom falling out of the market for the Smooth Cayenne
variety that they grow, but have been unable to switch to the new variety as
MD2 pineapples require capital-intensive investment. Some farmers have suf-
fered heavy financial loss as they have been unable to sell the pineapples they
have grown, whilst others have lost the land that they have been cultivating
for generations to large commercial farms. A few of the most successful
farmers, who all started off as smallholders, have been able to expand their
pineapple production and maintain sales to pineapple-processing companies
located nearby.
This chapter focuses on the changes taking place in livelihoods and settle-
ments located in the Ghanaian pineapple frontier, exploring the ways in
which the explanatory factors for these changes lie in both global processes
and local particularities. Correspondingly, the chapter is divided into two
sections. In the first section we discuss the dynamics of the global market
for fresh pineapples, highlighting how supermarkets and multi-national com-
panies are increasingly controlling the export market for fresh fruit. In the
second section we turn to focus on changing livelihoods, and their impact
on settlement growth, in two settlements which have been dominated by
pineapple production in Ghana. We show how both settlement growth and
livelihood opportunities can be highly determined by the changing fortunes
of one crop though in varying ways. In particular, access to land associated
with membership of an ethnic group influences livelihood opportunities in
the pineapple frontier which in turn affects settlement growth and contrac-
tion. Hence, we argue that although all residents in the pineapple frontier
60 Katherine V. Gough and Niels Fold
have been affected by the changes in the global pineapple market, the ways in
which these are worked out at the local level vary not only between but also
within settlements.

Dynamics in the global market for fresh pineapples

Since the mid 1980s consumption in the industrialised countries of fresh fruit
from the Global South has increased substantially in both scope and diver-
sity. Up till then, only bananas, previously the archetypical example of a
tropical fruit, were internationally traded on a significant scale. Bananas were
gradually supplemented by other tropical fruits, including fresh pineapples,
and counter-seasonal temperate products from countries such as Chile,
Argentina and South Africa. The shift in supplies was primarily caused by
new developments in cooling techniques and logistics that allowed the culti-
vation of fresh produce (fruit and vegetables) to be increasingly geographically
separated from consumption. In addition, the dietary patterns of consumers
in the Global North changed towards a more healthy composition with a
higher share of fresh produce. Initially this trend was most pronounced
among more affluent consumers but it progressively became more widespread
(Friedland, 1994).
This shift in dietary patterns took place in the same period as the retail
sector in the Global North was significantly restructured (Busch and Bain,
2004). As a general trend, supermarkets have taken over most of the food
provisioning business from traditional wholesalers and local retailers. At
the same time, consolidation processes within most national retail sectors
resulted in structurally similar situations whereby only a handful of super-
market chains control the bulk of food retail marketing and sales. Some
observers argue that supermarkets have transformed their functions from
mere food distribution to strongly influencing consumption patterns by con-
structing new demand patterns via product innovation and differentiation,
for example by introducing convenience foods such as ready-meals and
different cuts of (seasoned) meat (Dixon, 2007). Supermarkets have also
expanded their involvement upstream with suppliers and strongly influence
production via so-called own-brand products and the implementation of
standards for suppliers in terms of requirements for food safety and quality
(Burch and Lawrence, 2007). In this sense, geographically dispersed pro-
duction and consumption of food is aligned via the supermarkets, in turn
positioning them as guarantors of ‘consumer interests’ by, for example,
promoting particular products as part of a healthy diet.
Even though there are significant differences in the scope and nature of
supermarket dominance between countries in the European Union (EU), the
general picture painted above applies to the provisioning of fresh produce.
During the past two decades, consumers in the EU have purchased increasing
volumes of fresh produce, including fruit from the Global South, in super-
markets at the expense of local greengrocers and operators at wholesale
 Rise and fall of smallholder pineapple production in Ghana 61
markets. Initially, wholesalers maintained a role in the supply chain as medi-
ators between growers/exporters and supermarket chains. However, as part of
a general process of rationalisation and supply chain re-organisation, super-
markets have increasingly come to rely on a small number of preferred
suppliers (Humphrey, 2006). These companies have the ability to expand and
upgrade their functions in accordance with the outsourcing strategies of the
supermarkets. In addition, they are able to monitor the product standards
set by the supermarkets so that quality requirements are controlled upstream
in the supply chain. Usually, these companies assume full responsibility for
sourcing specific categories of closely related products to supermarkets, for
instance a subgroup of tropical fruits. The companies thus become so-called
category managers.
One of the consequences of the comprehensive introduction of standards
into supply chains and the appearance of category managers has been a
corresponding concentration of producers away from smallholders to larger
grower/exporters in the Global South (Dolan and Humphrey, 2004). For
the category managers, these large-scale producers can more readily comply
with given standards as they are easier to monitor, transaction costs are
smaller, and they have the capacity to accommodate lower-profit functions
that are outsourced such as cutting, packing, bar-coding, etc. It is still pos-
sible, however, for smallholders in the Global South to enter and main-
tain a position as suppliers of fresh produce to supermarkets or traditional
wholesalers in the EU. Fresh produce is considered a non-traditional, high-
value crop with considerable export potential for smallholders. Numerous
activities and aid projects have been implemented by multilateral and
bilateral donors, often in cooperation with NGOs, with the aim of incorpor-
ating African smallholders in supply chains to buyers in the EU. Some of the
large-scale growers/exporters operate schemes in which selected smallholders
are trained and monitored to produce products according to the prevailing
product and process standards. In general, the increasingly global nature of
procurement areas and the logistic capacity of sourcing agents result in new
opportunities for smallholders. The sourcing networks may potentially act
as ‘springboards’ to global markets and previous arm’s-length supplies to
distant wholesale markets may be replaced by a close relationship with active
marketing agents. In addition, some of the standards on product quality and
food safety may be used for a proliferation of existing markets via product
differentiation (Henson and Jaffe, 2006; Masakure and Henson, 2005).
The nature and importance of smallholder involvement, however, seems to
fluctuate in phases as imports of fresh produce from the Global South tend to
traverse through three successive phases (Reardon and Flores, 2006). In the
initial phase, the new product exploits the benefits of a narrow niche market
where profits are relatively high for early entrants. Even though high profits
seldom trickle down to smallholders, they are nevertheless an important
part of the supplier base in this phase. If the product is successful and
the market expands, large-scale suppliers will start production, in turn
62 Katherine V. Gough and Niels Fold
transforming the product to a basic commodity. In this second phase, small-
holders tend to be marginalised in the supplier base as price becomes the
main competitive parameter. Finally, in the third phase, the basic commodity
is subject to innovation and product differentiation resulting in the creation
of a series of new niche markets, not least due to the implementation of
process standards (e.g. fair trade and organic products).
This brief outline of the dynamics of the retail sector in the EU and the
conditions for smallholder involvement in fresh produce exports from the
Global South serves as a broad framework for understanding the globalisa-
tion process of fresh pineapples and the incorporation and subsequent mar-
ginalisation of Ghanaian smallholders (Fold and Gough, 2008). Consumption
of fresh pineapple took off during the 1980s, particularly in the US where
production was re-located from Hawaii to Costa Rica and on a smaller
scale in other Central American countries. Traditionally, pineapple pro-
duction for the US market was dominated by large plantation companies,
notably Dole and Fresh Del Monte, who had an economic basis in banana
trade. These plantation companies dominated production in Costa Rica but
other companies and local large-scale producers were also involved. The
pineapple sector in Costa Rica was rapidly developed with infrastructure
(including roads and ports) and the necessary cooling facilities along the
chain. Production and exports increased tremendously during the 1990s
and after the turn of the century, resulting in Costa Rica dominating pine-
apple exports on the world market; for example, pineapples from Costa
Rica constituted almost 70 per cent of total imports to the EU by 2007
(MIR, 2008). This has been mainly due to the successful development of a
new variety of pineapple (MD2) that was massively planted in Costa Rica,
initially by Fresh Del Monte who managed to get a 10-year patent in 1993.
MD2 is considered to be better adapted to consumer preferences in terms of
taste (sweet), colour (golden), shape (rounded) and size (‘family-meal’ with
no need for storing leftovers in the fridge). Whether these characteristics
appeal to ‘intrinsic’ consumer preferences is hard to tell but comprehensive
marketing campaigns certainly stressed these virtues. Almost all large-scale
pineapple companies have switched to produce and export MD2 whilst
other varieties have concurrently experienced a dramatic decrease in global
demand.
In the EU, most of the booming demand in the 1980s for pineapples
was covered by exports from French plantation companies located in Côte
d’Ivoire. In the late 1990s and just after the turn of the century, exports from
Ghana also increased substantially, coinciding with the growing demand for
fresh pineapple among European consumers. The majority of pineapples
from West Africa were of the Smooth Cayenne variety, a bigger pineapple
with a somewhat bitter taste compared to MD2. In contrast to MD2, Smooth
Cayenne is more sturdy and less perishable, hence easier to cultivate and
harvest for smallholders with limited access to cooling facilities (Illustration
5.1). At the same time as Ghanaian exports increased, the major US
 Rise and fall of smallholder pineapple production in Ghana 63

Illustration 5.1 Freshly harvested pineapples lie by the roadside ready to be packed
into containers and loaded into a waiting truck.
Photo by Katherine V. Gough

plantation companies started an offensive marketing campaign for MD2

in the EU. The US market had gradually become saturated and production
capacity in Central America was sufficient to cover both the US and EU
markets. As part of the strategy, the plantation companies also acquired
existing European wholesale and distribution companies thereby expanding
their facilities and logistic capacity in the EU market. Some of the French
plantation companies, including shipping lines, were also taken over by
the US plantation companies. In essence, the US companies were able to
catapult their pineapples directly onto the supermarket shelves, bypassing
existing category managers or traditional wholesale markets where neces-
sary. Direct access to consumers with product promotion and marketing
was supported via in-store campaigns. The swift conquest of the EU market
for fresh pineapples took place at the expense of traditional smallholder-
based exports of Smooth Cayenne pineapples from Ghana and Côte d’Ivoire,
the latter marred by civil war. As we shall now turn to show, the dram-
atic decline in exports from the West African countries had severe impacts
on smallholder households and settlements in the Ghanaian pineapple
frontier.
64 Katherine V. Gough and Niels Fold
Changing livelihoods and settlement growth in the
pineapple frontier
The livelihoods of the inhabitants and the nature of settlement growth in the
Ghanaian pineapple frontier can broadly be divided into three phases
roughly corresponding to the production and export trends for pineapples
outlined above: a period before pineapples became the main income source
(pre-1990s), a boom period in the pineapple sector (1990s–early 2000s) and a
post-pineapple boom period (post-2005). For each phase, the changing and
diverse nature of livelihoods and their impact on settlement growth will be
discussed. First, however, the nature of the pineapple frontier and the profiles
of the settlements studied will briefly be outlined.

Ghanaian pineapple frontier

Pineapples are predominantly grown in the south of Ghana in the area where
the Eastern, Central and Greater Accra regions converge. Pineapples were
introduced into the Akwapim Hills in the Eastern Region in the 1980s by
smallholders from the indigenous landholding groups. During the 1990s,
a number of international NGOs assisted the smallholders in organising
cooperatives to streamline production, collection, transport and shipping of
their pineapples. When the demand for exports rose during this period, large-
scale cultivators became interested in entering the market but were unable to
obtain large tracts of land in the Akwapim Hills. They headed west and
obtained leaseholds for land in the Central Region and Greater Accra Region
where they set up extensive commercial pineapple farms. The nature of the
expansion of pineapple production is closely related to the customary system
of land tenure in Ghana which is central to understanding access to liveli-
hoods and settlement growth (Alhassan and Manuh, 2005). Most land in
Ghana is held under customary land tenure arrangements. The ownership
of land is communal, with traditional authorities (usually chiefs and elders)
acting as trustees having the right to allocate land. Members of the indigen-
ous land-owning group who are allocated land usually continue to have the
rights to the land as long as they are cultivating it. Migrants in host com-
munities, referred to as ‘strangers’, do not have the same rights of access to
land. Regardless of how many generations have been resident in a settlement,
they cannot become indigenous and do not have access to communal land.
Strangers can acquire access to land, however, if some is available, through
a range of means (Kasanga, 2001; Kasanga and Kotey, 2001). A common
method in the pineapple-growing areas is through the system of abusa
whereby the landowner receives one third of the crop as payment for the use
of the land. In situations of land scarcity, indigenous people may also expand
their access to land under the abusa system.1
In order to explore changing livelihoods and settlement patterns in the
pineapple frontier, one settlement was studied in the original Akwapim Hills
 Rise and fall of smallholder pineapple production in Ghana 65

Figure 5.1 Map showing location of study settlements in the pineapple frontier.

pineapple-growing area (Pokrom Nsaba, hereafter referred to as Pokrom)

and one in the newly expanding area further west (Obom) (see Figure 5.1).
Pokrom is located in Akwapim South District in the Eastern Region,
approximately halfway between the district capital towns of Nsawam and
Aburi. The road to Nsawam has recently been paved and tro-tros (mini-
buses) ply the route frequently. The inhabitants are predominantly from
the indigenous land-owning Akan group (87 per cent) supplemented by Ewe
migrants from the Volta region (Table 5.1). The 383 houses are mainly
family-based compound houses built with concrete bricks and roofed with
corrugated iron. According to the official Government of Ghana census,
2181 people lived in Pokrom in 2000. Obom is located in Ga West District in
Greater Accra Region, equidistant from the district capital towns of Nsawam
and Amasaman. Despite being located close to Accra, there is no electricity
in Obom and all the roads leading there are poorly maintained dirt tracks. No
tro-tros serve Obom but a number of shared taxis brave the roads. According
to the 2000 census, the population of Obom is 1179 but only 18 per cent
belong to the indigenous land-owning group Ga-Dangbe. The inhabitants are
66 Katherine V. Gough and Niels Fold

Table 5.1 Ethnic groups of the inhabitants of

Pokrom and Obom (%)

Ethnic group Pokrom Obom

Akan 87 27
Ga-Dangbe >1 18
Ewe 10 41
Northerners 1 12
Other 1 2

Source: Ghana Census Survey, 2000.

thus predominantly strangers (non-indigenous) with the largest group being

Ewe followed by Akan and Northerners (Gurma, Mole-Dagbon and Grusi).
They live in a total of 272 houses which are mainly small, sometimes com-
pound houses with a predominance of mud walls and corrugated iron roofs,
though a third still have thatched roofs (Illustration 5.2).

Pre-pineapples phase
According to oral history, the settlement of Pokrom dates back to around
200 years ago when the indigenous Akwapim moved down from the hills to
live by the new road which had been constructed along the valley bottom. In
contrast, Obom was settled primarily by strangers who started migrating
from the east and north of Ghana, about a hundred years ago, looking
for land to cultivate. The original inhabitants of both Pokrom and Obom
grew basic foodstuffs, mainly for subsistence though some surplus was sold.
In Pokrom, some farmers turned to growing cocoa as a more lucrative
income-generating activity. By the 1950s, however, swollen shoot disease
had devastated cocoa production (Goldstein and Urry, 2006) and a number
of the inhabitants migrated west to become tenant cocoa farmers in the new
cocoa frontier (see Chapter 3). Growing food crops again became the major
activity in Pokrom, with maize and cassava being especially important crops.
Incomes were supplemented with remittances from family members who had
migrated. Pokrom was thus a settlement of out-migration for many years
with people migrating further west to work on the cocoa farms, into Accra to
search for employment in the city, or venturing as far as Nigeria to try their
luck abroad.
Strangers were allowed to settle and cultivate land in Obom without mak-
ing any formal payment or receiving any papers as the land was not being
used. Hence, many inhabitants have been living off the land for several gener-
ations. The opportunities for selling their produce rose dramatically when the
main Accra–Takoradi railway line was built nearby in the 1950s. A number of
residential quarters were constructed near the railway line for railway
employees and the chief moved his palace to be close to the railway as this
 Rise and fall of smallholder pineapple production in Ghana 67

Illustration 5.2 The street running through the centre of Obom where most of the
stores are located.
Photo by Katherine V. Gough

had become the centre of activity in the settlement. Obom played an important
role as a small market town for over 40 years with goods and people travelling
to and fro by rail. When the railway line was closed in the mid 1990s it was a
serious blow for Obom. The once thriving market soon collapsed as the very
poor condition of the roads meant that there was no alternative mode of
transport to the settlement. As there were no longer any railway employees
the houses had to be vacated, and now lie derelict. A number of the former
railway employees, however, remained in Obom as they had settled there with
their families. Their job opportunities were limited by the lack of electricity in
the settlement which hampers many potential income-generating activities.
Another important influence on settlement growth in Obom was the arrival
of the district health centre in 1974. The centre serves a wide area including
more than 100 villages and its main role is to provide primary healthcare. It is
hampered, however, by the lack of electricity which makes storing drugs very
difficult. The 30 or so employees reside in the accommodation specifically
built for them alongside the hospital. As well as these professionals, who are
migrants, there has also been a stream of migrants from the north of Ghana.
For generations they have travelled down from the north when there is no
work on their own farms, to the south to work as labourers for smallholders
68 Katherine V. Gough and Niels Fold
growing food crops for sale. Although most of them have been seasonal
migrants, some have settled in Obom providing in turn a home for relatives
and friends coming to work for short periods of time.
Summing up, prior to the boom in pineapple production, Pokrom was a
settlement of primarily indigenous inhabitants some of whom had migrated
out of the settlement to search for better livelihoods elsewhere. Obom on
the other hand was primarily settled by migrants who were growing food
crops and by employees of the railway and health clinic. They had already
experienced a boom-and-bust period with the arrival and departure of the
railway.

Pineapple boom years

Pineapples, of the Smooth Cayenne variety, started to be grown in Pokrom in
the early 1980s and were sold on the local market. The number of pineapple
farmers increased slowly at first but then accelerated in the late 1990s as
demand for pineapples for export grew. Many of the farmers began to grow
pineapples after the start of the new millennium. By 2006, 83 per cent of the
households interviewed were engaged in pineapple farming in some way and
it formed the main source of income for over half (56 per cent). Almost all
(97 per cent) had a second source of income and for 22 per cent of these it
was pineapple farming. This illustrates how central pineapple farming had
become to livelihoods in Pokrom. The farmers were predominantly small-
holders and mainly members of the indigenous landholding group; a third
farmed less than one hectare, 60 per cent farmed 2 hectares or less, and only
11 per cent had farms which were larger than 20 hectares. Even the larger
farms had started small; the largest farm of 210 hectares started in 1988 with
just a tenth of a hectare and has gradually expanded over the years.
From the 1980s, and increasingly during the 1990s, some of the inhabitants
of Pokrom started to become relatively wealthy from cultivating pineapples.
Many households invested in improving their homes, replacing the mud walls
with concrete blocks and adding extra rooms, thus expanding the size of the
house. Some chose to rent out rooms to migrant workers who were needed to
work on the pineapple farms, especially during weeding and harvesting. The
population of Pokrom grew not only due to the influx of migrants but also
as some members of the indigenous group returned to their hometown.
Encouraged by stories of the amounts of money that could be made cultivat-
ing pineapples in Pokrom, they returned to start pineapple farms themselves.
Pokrom thus became a settlement of in-migration during the 1990s and in the
immediate years after the turn of the century.
During the 1990s, large-scale pineapple plantations began to be established
to the west of the Akwapim Range where land was more widely available. As
much of the land was being farmed by strangers who had no legal right to the
land, the chiefs and elders were in a position to sell lucrative leaseholds for
large areas of land. Often with the backing of foreign capital, including
 Rise and fall of smallholder pineapple production in Ghana 69
German, French and Italian, these farms were planted with large areas of
pineapples, initially with the Smooth Cayenne variety, later with MD2. For
example, a farm on the outskirts of Obom had 1000 hectares of pineapple
under cultivation resulting in an annual yield of 16,000–17,000 tonnes by
2007 and was planning to expand. Many of the workers employed on the
farms came from the surrounding settlements, including Obom. Having
gained skills and knowledge of pineapple farming, and encouraged by reports
of the profits that smallholders were making in the Akwapim Range, some
smallholders in Obom switched from growing food crops to investing in pine-
apples in the early years of the new millennium. Initially the smallholders
made reasonable profits but there were many cases of farmers who entered
the market just as the demand for pineapples fell and who subsequently lost
considerable amounts of money. In 2006, 21 of the 75 households inter-
viewed were still growing pineapples but a further 14 households had already
stopped. Pineapple farming was recorded as being the most important source
of income for only 15 per cent of households, with a further 6 per cent naming
it as their most important secondary source of income. Crop farming was the
most important source of income overall (for 35 per cent of households) with
a further 24 per cent naming unskilled work, including casual work, as their
most important income source.
As the account above shows, Obom was less reliant than Pokrom on pine-
apple farming as an income source, and livelihoods in Obom were more
diverse. This is reflected in data for 2000 from the Ghana Census Survey,
which records that 75 per cent of income earners in Pokrom were engaged in
agriculture whereas the corresponding figure for Obom was 46 per cent. As
Table 5.2 shows, the majority of workers in both settlements were working in
the informal sector. The higher proportion of workers in the formal private
sector in Obom reflects labourers employed on the pineapple farms.

Post-pineapple boom
Pokrom is a changed settlement since Smooth Cayenne pineapples are no
longer in demand on the export market. Whereas until 2005 the place was
a hive of activity with people walking to their fields, working in the packing

Table 5.2 Employment sector for workers in

Pokrom and Obom (%)

Employment sector Pokrom Obom

Public 5 14
Private – formal 16 29
Private – informal 78 57
Sample size 979 460

Source: Ghana Census Survey, 2000.

70 Katherine V. Gough and Niels Fold
house, and trucks laden with pineapples were constantly passing by, by 2007
the settlement had an empty feel. Many of the fields surrounding the settle-
ment, which were once carefully tended pineapple plots, are now overgrown.
Although Smooth Cayenne is no longer exported, it is still bought by pine-
apple processors, mainly small- and medium-scale companies producing juice
for the local market, and local market women. The amount that can be
earned is much less, with a market woman typically paying only half
of what a processor will pay which in turn is half the amount that exporters
used to pay. An important processor in the region is Blue Skies, a foreign-
owned company that produces sliced fresh fruit for the EU market. It only
buys, however, from farmers who have been trained in production practices
according to EurepGAP2 and can supply a minimum of 2000 pineapples
weekly. Many of the farmers who were still growing pineapples by 2007
were from the larger farms which had a regular order from Blue Skies. The
majority of the smallholders have not been able to afford the necessary inputs
to continue farming pineapples or switch to another crop. For many, their
capital is tied up in outstanding payments owed by exporters who had col-
lected pineapples which they had never paid for, claiming that they had not
themselves received payment. Many of the farmers could produce invoices of
up to several million cedis3 for pineapples which had been collected by
exporters up to two years previously but for which they had not yet been paid.
Switching to growing MD2 is not an option for smallholders because the
investment needed in cooling and packing facilities is beyond their means,
and because the price of seedlings4 is so high.
The number of people living in Pokrom has fallen since the demand for the
pineapples dropped. Most of the migrants have left as there is no more need
for their labour, and some of the indigenous inhabitants have started to leave
to look for work elsewhere, principally in Accra. The young people talked
about feeling trapped in Pokrom without the financial capacity or contacts to
move elsewhere to look for work; the mood in Pokrom was one of widespread
despair. Obom has similarly become a settlement of out-migration. Many of
the smallholders who have been growing foodcrops, but as strangers without
any formal rights to the land they farm, have lost their land to the large-scale
pineapple plantations. Not entitled to compensation, and with no land in the
vicinity left for them to farm, they are moving elsewhere in search of new
land. Some of those who remain in Obom have found employment as labour-
ers on the large pineapple farms. However, they complain that the work is
hard, the hours long and the pay low. Others are attempting to find new ways
of earning an income in Obom but the lack of electricity restricts the estab-
lishment of most income-generating activities. The circular flow of migrants
from the north, who worked for smallholders growing food crops, has
virtually stopped as the number of these smallholders has dropped. The
large pineapple companies are not interested in employing casual workers,
who will only stay for a few months at a time, preferring instead to employ
workers on a permanent basis. The smallholders who still have access to land,
 Rise and fall of smallholder pineapple production in Ghana 71
predominantly the indigenous inhabitants, have found that, as in Pokrom,
they can no longer sell their pineapples to exporters since the demand
switched to MD2. Many are now switching back to growing foodcrops
once again.

Conclusions
This study of two settlements in the Ghanaian pineapple frontier has shown
how the livelihoods of smallholders can be greatly affected by changes taking
place at the global level. A thriving smallholder-based pineapple sector in
Ghana was gradually eroded by competitive strategies of transnational com-
panies with global activities in production, processing and export of tropical
fruits. These global changes are, however, not the only forces operating,
with local processes affecting the ways in which these global processes are
worked out at the local level. As well as being affected by the changing
fortunes of the pineapple trade, access to livelihoods in Pokrom and Obom is
closely tied to the customary land tenure system. The differing accounts that
emerge from the two settlements are directly linked to the inhabitants of
Pokrom being predominantly indigenous whereas those in Obom are mainly
‘strangers’. The former are smallholders who increasingly specialised in pine-
apple production until the demand for the Smooth Cayenne variety fell.
Although the latter had farmed the land for generations, they had no right to
the land and have lost their smallholdings since the land they had been culti-
vating was leased to large-scale pineapple farms. Some have become wage
labourers on these farms.
In terms of settlement processes, Pokrom expanded during the boom years
in the pineapple trade as indigenous people returned and migrants came
looking for work, but has subsequently contracted with the young indigenes
leaving to look for work elsewhere. Obom suffered a drop in population
following the closure of the railway which effectively closed the market that
had grown up around the station. The population has subsequently fallen as
some of the ‘strangers’ who have lost their land have moved elsewhere and
because fewer short-term migrants pass by since there is no longer a demand
for casual labourers to help with foodcrops. Hence, it has been shown how
both processes of settlement and livelihood opportunities in frontier regions
can wax and wane with the changing demand for their products on the world
market, but that the ways in which these are worked out in particular local-
ities can vary widely.

Notes
1 See Agbosu et al. (2007), Aryeetey et al. (2007a) and Aryeetey et al. (2007b) for a
more detailed discussion of the land tenure system in Ghana.
2 EurepGAP is a set of quality standards developed by European supermarket chains
and their suppliers to secure ‘Good Agricultural Practices’ in farm management.
The comprehensive standards establish traceability and cover issues related to
72 Katherine V. Gough and Niels Fold
environmental impact, animal welfare, use of chemical inputs, and worker health
and safety. In September 2007, EurepGAP changed its name to GLOBALGAP,
reflecting the expanding international role in establishing Good Agricultural Prac-
tices between multiple retailers and their suppliers on a global level. Detailed
information is available at www.globalgap.org.
3 There were approximately 10,000 cedis to the dollar at the time.
4 Vegetative multiplication of pineapple in Ghana is usually based on suckers
(i.e. shoots arising from the base of the plant at ground level) which are cheaper
than seedlings. So far suckers are not produced in sufficient quantities to be avail-
able on the market.

References
Agbosu, L., Awumbila, M., Dowuona-Hammond, C. and Tsikata, D. (2007) Custom-
ary and statutory land tenure and land policy in Ghana, Technical Publication No.
70, ISSER, Legon: University of Ghana.
Alhassan, O. and Manuh, T. (2005) Land registration in Eastern and Western Regions,
Ghana, Research Report 5, London: IIED.
Aryeetey, E., Al-Hassan, R.M., Auming-Brempong, S. and Twerefou, D.K. (2007a)
The organization of land markets and production in Ghana, Technical Publication
No. 73, ISSER, Legon: University of Ghana.
Aryeetey, E., Al-Hassan, R.M., Auming-Brempong, S. and Twerefou, D.K. (2007b)
Legal and institutional issues in land policy reform, Ghana, Technical Publication
No. 74, ISSER, Legon: University of Ghana.
Burch, D. and Lawrence, G. (2007) ‘Supermarket’s own brands, new foods and the
reconfiguration of agri-food supply chains’. In Burch, D. and Lawrence, G. (eds),
Supermarkets and agri-food supply chains, Cheltenham: Edward Elgar.
Busch, L. and Bain, C. (2004) ‘New! Improved? The transformation of the global
agrifood system’, Rural Sociology, 69(3): 321–46.
Dixon, J. (2007) ‘Supermarkets as new food authorities’. In Burch, D. and Lawrence,
G. (eds), Supermarkets and agri-food supply chains, Cheltenham: Edward Elgar,
pp. 29–50.
Dolan, C. and Humphrey, H. (2004) ‘Changing governance patterns in the trade
in fresh vegetables between Africa and the United Kingdom’, Environment and
Planning A, 36(3): 491–509.
Fold, N. and Gough, K.V. (2008) ‘From smallholders to transnationals: The impact
of changing consumer preferences in the EU on Ghana’s pineapple sector’,
Geoforum, 39: 1687–97.
Friedland, W., (1994) ‘The new globalization: The case of fresh produce’. In Bonanno,
A., Busch, L., Friedland, W., Gouveia, L. and Mingione, E. (eds), From Columbus
to ConAgra: The globalization of agriculture and food, Lawrence: University Press
of Kansas, pp. 210–31.
Goldstein, M. and Urry, C. (2006) The profits of power: Land rights and agricultural
investment in Ghana, ISSER discussion paper no. 32, Legon: University of Ghana.
Henson, S. and Jaffe, S. (2006) ‘Food safety standards and trade: Enhancing com-
petitiveness and avoiding exclusion of developing countries’, European Journal of
Development Research, 18(4): 593–621.
Humphrey, J. (2006) ‘Policy implications of trends in agribusiness value chains’,
European Journal of Development Research, 18(4): 572–92.
 Rise and fall of smallholder pineapple production in Ghana 73
Kasanga, K. (2001) Land resource management for agricultural development in Ghana,
London: RICS Foundation.
Kasanga, K. and Kotey, N.A. (2001) Land management in Ghana: Building on tradition
and modernity, London: IIED.
Market Intelligence Report (MIR) (2007) online at http://www.ghanafreshproduce.
org/images/stories/pdf/MIR/mirpineapplejandec2007.pdf.
Masakure, O. and Henson, S. (2005) ‘Why do small-scale producers choose to produce
under contract? Lessons from non-traditional vegetable exports from Zimbabwe’,
World Development, 33(10): 1721–33.
Reardon, T. and Flores, L. (2006) ‘Viewpoint: “Customized competitiveness” strat-
egies for horticultural exporters: Central America focus with lessons from and for
other regions’, Food Policy, 31(6): 483–503.
6 Transforming livelihoods
and settlements
Fruit production and
urbanisation in Vietnam’s
Mekong Delta
Hoang Xuan Thanh and Cecilia Tacoli

Introduction
This chapter describes the multiple impacts of rapid urbanisation and indus-
trialisation on three rural settlements in the area often described as Vietnam’s
rice bowl and where livelihoods and the nature of the local economic base,
including farming, have changed radically in the past two decades. Rice no
longer dominates crop production in the settlements; in fact, it has virtually
disappeared, giving way to fruit orchards. Fruit is the key driver of the
diversification of the economic base in the study settlements. This, in turn, is
triggered by the rising demand from urban consumers as well as by the
growth of industry and services in local urban centres. Employment in these
sectors absorbs large numbers of increasingly mobile rural dwellers, especially
but not only the younger generations, and contributes substantially to rural
households’ incomes and capacity to invest in fruit production. Within the
settlements, the expansion of activities related to the fruit trade has opened up
opportunities for wealthier and poorer groups alike, and contributes to local
economic growth and poverty reduction, although with widening income
inequality.
These transformations have not taken place in a vacuum, however, and in
order to understand future opportunities and challenges it is important to
consider the wider landscape beyond the local dimension. This includes
national policies in relation to economic and infrastructural development
strategies, as well as broader regional and international trade and investment
patterns and agreements.

The fruit sector in Vietnam: expanding domestic demand and

unstable export markets
Fruit production in Vietnam more than doubled between 1995 and 2005,
from 3 million metric tons to over 6 million metric tons, much as the result of
changes linked to the doi moi (innovation) reforms of the mid 1980s (MARD,
2005). The allocation of land use rights to households is widely acknowledged
 Fruit production and urbanisation in Vietnam’s Mekong Delta 75
to have played a key role in the reduction of poverty in Vietnam by increasing
agricultural productivity and stimulating investment. It also has an important
influence on farmers’ decisions to switch to fruit trees that require significant
investment and only become fully productive a few years after planting.
Other supply-side factors that have stimulated the growth of fruit farming
include the implementation of specific agricultural policies in support of fruit
production, and the relaxation of land use regulations which previously made
it difficult, if not illegal, to switch from rice to other crops without official
permission (Vietnam Development Report, 2004). As a result, in 2005 over
750,000 hectares of farmland in Vietnam were planted with fruit trees, up
from 360,000 hectares in 1995. The Mekong Delta concentrates 35 per cent of
the total fruit tree area, and more than half of the total fruit production in
Vietnam (GSO, 2004).
On the demand side, consumption of fruit in both rural and urban areas
has doubled in the past decade, making it the key driver of the sector’s fast
growth. Per capita expenditure on fruit is higher in urban centres, especially
in the larger cities, Hanoi and Ho Chi Minh City (HCMC). Nevertheless, in
both rural and urban areas the consumption of high-value fruit, such as
mango and citrus, has increased while that of banana, the cheapest fruit, has
remained stable. Dietary transformations and changes in food expenditure
patterns, in turn, are closely related to the generalised and rapid improvement
of the living standards of Vietnamese households (VNLSS, 1993;
VLSS,1998; VHLSS, 2002, 2004).
In this context, exports appear to play a secondary role compared to internal
demand in stimulating production, although with regional variations. Fruit
exports have fluctuated widely in the past decade. After reaching a peak of 13
per cent in 2001, the share of fruit in the total agricultural export value of the
country had fallen to just 4 per cent in 2004 (GSO, 2004). Until 2001, China
was the main export market for Vietnamese fruit, but since its accession to the
World Trade Organization (WTO) its share of the export value has fallen
from 80 per cent to 40–50 per cent. At the same time, Vietnam has become an
important market for fruit imports from China, the US and Thailand. The
accession of Vietnam to the WTO in 2007 is likely to increase this further as
from 2010 import tariffs will be reduced from the current 40 per cent for all
fruit to specific tariffs for different fruit. Growing integration in international
markets is therefore important but, at least so far, domestic demand has
proved to be more substantial and especially more stable.
In the Mekong Delta, the change in land use from paddy to fruit gardens
has occurred faster than in the rest of the country. The share of rural house-
holds’ productive land used for growing fruit has increased from 12 per cent in
1993 to over 32 per cent in 2004, compared to the national average of just
under 8 per cent and over 13 per cent for the same years. Fruit producers in
the Mekong Delta also have larger orchards than the national average, with
39 per cent using more than 1000 m2 (an increase from 25 per cent in 1993)
compared to 15 per cent nationally (from 10 per cent in 1993). The trend
76 Hoang Xuan Thanh and Cecilia Tacoli
towards land concentration, the result of an emerging land market, goes
hand in hand with increasing landlessness (from 17 per cent in 1993 to
29 per cent in 2002) which in turn is closely linked to poverty in the region
(VNLSS, 1993; VLSS,1998; VHLSS, 2002, 2004). However, it should be
noted that this correlation between poverty and landlessness does not neces-
sarily apply to other regions of Vietnam. In the Red River Delta, for example,
landlessness appears to be higher among rural households in the richer
expenditure quintile, possibly reflecting the fact that the rural non-farm
sector is more developed there (Hoang et al., 2005).
Due to market fluctuations and increases in non-agricultural employment,
the share of revenue from fruit in Vietnam’s rural household incomes
remained unchanged between 1993 and 2004, with an average of 2.5 per cent,
although it was slightly higher in the Mekong Delta, on average just over
4 per cent. Nevertheless, during the same period, poverty has declined faster
among fruit-growing rural households than among those not growing fruit in
the region, although this is not the case for the whole of the country (VNLSS,
1993; VLSS, 1998; VHLSS, 2002, 2004).
The findings from this case study suggest that changes in farming produc-
tion systems – in this instance, from rice to high value-added and less
labour-intensive fruit farming – are best understood as a component of
fundamental transformations in the livelihoods of rural households and in
the economic base of the settlements where they live. Key to these transform-
ations is the diversification of income sources and economic activities which,
in turn, builds on stronger links between rural areas and urban centres. The
three study settlements also suggest that there are many different ways in
which these transformative processes take place, depending on a wide range
of factors – geographical, social, economic and ecological. Hence, while
one settlement is rapidly growing into a dynamic small market town, another
one retains its agricultural vocation but is supplemented by significant
migrant remittances, while the third is affected by its proximity to one of the
fastest-expanding cities in the delta.

The study locations

In order to explore the impacts of urbanisation and industrialisation on high
value-added fruit production systems and, as a result, on rural settlements
and livelihoods, we used the following selection criteria for our study loca-
tions: the established production at the settlement level of speciality fruit
(likely to have better marketing opportunities); access to both export and
domestic markets (providing us with the opportunity to investigate value
chains); and proximity to growing urban centres (likely to have an impact on
employment and mobility patterns).
The three settlements in which this study was conducted are located in two
of the major fruit-producing provinces in the Mekong Delta. Fruit-growing
areas in Tien Giang and Vinh Long provinces have increased by 30 per cent
 Fruit production and urbanisation in Vietnam’s Mekong Delta 77

Illustration 6.1 The Mekong Delta’s extensive waterways system links orchards to
market nodes, from where produce is shipped to domestic and inter-
national markets.
Photo by Niels Fold

and 55 per cent respectively since 2000–2001. Tien Giang Province is well
known for its production of speciality mango and milk fruit, while Vinh
Long is a major producer of speciality pomelo fruit. Both provinces are
strategically located and well connected to road, river and transport networks
(Illustration 6.1). Tien Giang province stretches 120 km in length along the
Tien River. My Tho city, the provincial centre, is only 60 km south of
HCMC. Vinh Long Province is located in the centre of the Mekong Delta,
between the Tien and Hau branches of the Mekong River. It is 135 km south
of HCMC and adjacent to Can Tho, the main city in the southern part of the
delta. With the construction of the My Thuan bridge across the Tien River
and the rehabilitation of Highway No.1A, road transportation has much
78 Hoang Xuan Thanh and Cecilia Tacoli
improved. All the communes of the two provinces are connected to the
national electricity grid and have paved roads.
Population density is the highest in the delta, and while the level of urban-
isation is increasing, at 15 per cent it is still much lower than the national
average of just below 28 per cent at the last census in 1999 (CCSC, 2000). Both
provinces are undergoing processes of economic transformation but are
still predominantly agricultural, as shown in Table 6.1. In 2005, the total
agricultural export values of Tien Giang and Vinh Long provinces were
US$72 million and US$133 million respectively. In the same year, official
income poverty rates were 23 per cent in Tien Giang and 13 per cent in Vinh
Long (Tien Giang Province, 2005; Vinh Long Province, 2005).
Table 6.1 The two study provinces: essential information

Vinh Thoi Hoa (mango) My Thoi 1

(milk fruit) (pomelo)

Agriculture (own farm) 19 62 46

Trade and services (own business) 68 18 27
Wage labour (skilled and unskilled) 10 17 22
Civil servants, pension, social welfare 2 2 2
Others 1 1 2
Sample size 100 100 100

Source: 2005 Statistical Yearbooks, Tien Giang and Vinh Long Provinces.

The three study settlements1

Vinh Thoi (Tien Giang province)
Vinh Thoi is the largest of the six settlements of Vinh Kim commune, which
is the smallest administrative unit in rural Vietnam. It is at the commune level
that data are collected and services provided. We start therefore by describing
the commune as the backdrop to the study settlement. Vinh Kim commune is
10 km from My Tho city, and has a total population of 11,000. Population
growth in the commune is a combination of natural increase (1.5 per cent
per year) and in-migration (1 per cent per year). Vinh Kim is one of seven
communes in the district growing the speciality Lo Ren milk fruit. Limited
land availability due to high population density and the decay of old and
stunted milk fruit trees have, however, contributed to a decline of farming in
favour of trade and services.
A market node for three surrounding districts, with good road and canal
links to HCMC and other provinces, Vinh Kim has a long-standing central
market and a newly built fruit market, around 150 fruit traders (vua trai cay)
and 6 specialised fruit packaging suppliers, all of which employ several per-
manent and seasonal workers. In addition, there are hundreds of associated
small service providers (tap hoa), a pool of 300 porters (boc vac), several
motorbike taxis (xe om) and parking attendants (giu xe). On average, each
 Fruit production and urbanisation in Vietnam’s Mekong Delta 79

Figure 6.1 Map showing location of study settlements in the Mekong River Delta.

day 30–40 trucks bring fruit from the commune markets to HCMC and other
provinces. A four buses per day service connects the commune centre to My
Tho city. In short, Vinh Kim can be described as a dynamic ‘urbanising’
commune with a diversified economic base, whose residents’ lifestyles, with
widespread use of mobile phones, cooking gas and bottled water, are more
‘urban’ than typically rural. Environmental issues, especially the disposal of
the growing commercial and domestic waste, are an increasing concern.
Vinh Thoi is a densely built settlement at the physical and economic centre
of the commune, adjacent to the two markets. It has 334 households with a
total population of 1766 people, plus 15 in-migrant households that do not
yet have permanent residential status.2 Since 2000, all rice fields in the village
have been converted into orchards. Trade and services linked to fruit produc-
tion are the main source of income for the residents, in combination with
fruit farming and husbandry. Although there are many opportunities within
the commune for employment in petty trade and services, these attract in-
migrants while young people prefer to migrate to HCMC and the industrial
zones in the delta. To poor people who do not have the resources to migrate,
the market offers income-earning opportunities such as small services, porter-
ing and fruit grading and packaging.

Hoa (Tien Giang province)

Hoa settlement is part of Hoa Hung commune. The commune’s location
along Highway No. 1A, right on the My Thuan bridgehead, makes it one of
80 Hoang Xuan Thanh and Cecilia Tacoli
the main fruit market nodes in the Mekong Delta. In 2005, the commune had
a permanent population of 16,139 people. Hoa Hung is the homeland of the
speciality Hoa Loc mango variety, and since 1996, all rice fields have been
converted into orchards. In 2001, the My Thuan bridge replaced the ferry
crossing of the Tien River. The increased traffic stimulated the expansion of
services for both residents and visitors, such as cafés, food stalls and grocery
shops supplying gas bottles and bottled water. This, in turn, contributed
to the transformation of local habits: for example, it is estimated that around
60 per cent of households in the commune now use gas cookers.
Hoa, a large settlement with 2600 residents, is well known for the produc-
tion of Hoa Loc mango, which is the main local income source and fetches
high prices. The village is well connected by river transport to the market node
of An Huu, just 3 km away. A Hoa Loc mango producer cooperative with 63
members was established in 2003. While Hoa can be described as an essen-
tially agricultural settlement, employment diversification is important for
local livelihoods, although less so than in Vinh Thoi. An estimated 50 house-
holds engage in petty trade and services, 20 provide private transport services
and more than 100 households have at least one member working away from
the settlement, mostly in the cities. Seasonal immigrants, typically around
30–40 people, come to the village for agricultural labour in the orchards.

My Thoi 1 (Vinh Long province)

My Thoi 1 settlement is part of My Hoa commune, which is located 7 km
from Cai Von district town and across the river from Can Tho city. The
commune has a population of 16,962. In 2004, a bridge was built to link
the commune with Cai Von district town by road. Before that, goods trans-
portation was mainly by boat. There is no commune market, and people
often go to Cai Von district town or to Can Tho city to buy their necessities.
My Hoa is famous for its Nam Roi pomelo, which is grown in almost all
garden land in the commune. However, due to the excessive expansion of
pomelo production, both in the province and the delta, prices have gone
down in the past few years. Moreover, the pomelo orchards area in the com-
mune is likely to be much reduced due to urbanisation and industrialisation.
The planned construction of the Can Tho bridge, the Binh Minh river port,
the Binh Minh industrial zone and a ‘green tourist centre’ will claim several
hundred hectares of farmland. This has already started: in 2002, 162 hectares
of orchards were cleared for the first phase of the Binh Minh industrial zone
construction.
My Thoi 1 has 1396 registered residents. It has fertile land and a stable
fresh water supply. Since 2003, a surrounding dike has much reduced the risk
of flooding and improved irrigation. Due to its location within the commune,
My Thoi 1 will not be directly affected by the planned construction of the
bridge, port and industrial zone, at least in the immediate future. However,
although the settlement’s land has not been claimed for infrastructure
 Fruit production and urbanisation in Vietnam’s Mekong Delta 81
construction, it is likely that the proximity of industrial activities will
somehow affect farming, and measures to control and reduce air and water
pollution will need to take into account their impact on peri-urban areas and
settlements such as My Thoi 1.
The conversion from paddy fields to Nam Roi pomelo orchards in My
Thoi 1 started in the early 1990s and was completed by 2001. Currently, the
main income sources of the villagers are a combination of fruit production,
husbandry, small trade, services and out-migration to Can Tho and HCMC.
One third of households in My Thoi 1 have at least one member working in
trade or service jobs. One third also have at least one migrant member, either
seasonal or permanent.

From fruit production to income diversification:

increasing wealth, widening inequality
In all three settlements, the majority of residents have radically transformed
the nature of their farming activities and, in most cases, the mix of income
sources that support their livelihoods. Although some fruit production has
long been part of farming in the area, the more drastic switch from rice
cultivation in the settlements started in the 1990s.
These changes in land use have affected husbandry, making it much less
important than in the past. Oxen and buffaloes, which were essential for
ploughing paddy fields, have all but disappeared and only a minority of
households raise pigs. The switch from rice to fruit has also created new
opportunities in trade and services. In part, this is because fruit production is
less labour-intensive than rice farming and, therefore, allows farmers to
engage in additional non-farm activities. The second reason is that fruit is
a perishable commodity with high value-added that requires a number of
services such as grading, packaging and daily transport to markets, all activ-
ities which are not necessary with dry, bulk commodities such as rice. At the
lower end – that of unskilled and semi-skilled wage labour – these activities
provide employment to poor local residents and seasonal migrants from
the surrounding areas. At the same time, the concentration of people and
activities around market nodes creates demand for services such as cafés and
restaurants, hairdressers, small shops, and transport.
As the economic base of the settlements becomes more diversified, the
proportion of household incomes derived from non-farm activities increases.
In Vinh Thoi, the most ‘urbanised’ of the three study settlements, trade and
services account for over two-thirds of the survey respondents’ household
incomes, against less than one fifth from agriculture. The income share of
trade and services is lower in both Hoa and My Thoi 1 (see Table 6.2), but
in all three settlements, including Vinh Thoi, respondents still perceive them-
selves primarily as farmers, perhaps reflecting the speed in the transformation
of their livelihoods. Overall, this change has been positive: living conditions
in the settlements have greatly improved in the past decade. Basic indicators
82 Hoang Xuan Thanh and Cecilia Tacoli

Table 6.2 Structure of respondents’ average household incomes in 2005 (%)

Tien Giang Vinh Long

Natural land area (km2) 2482 1478

Agricultural land 74 79
Population density (person/km2) 685 714
Urban population 15 15
Economic structure (2005): Agriculture 48 53
Industry and transports 22 16
Services 30 31

Source: Authors’ household survey, 2006. Figures represent the average income composition in
each settlement and reflect their economic base.

of assets, housing and facilities are much better than the average for
the Mekong Delta rural region, which includes more remote areas and
settlements that produce primarily paddy. For example, more than half of the
respondents live in houses with permanent walls and roof, compared to an
average of 8 per cent for the Mekong Delta rural region (VHLSS, 2004).
At the same time, however, income inequality has widened, against a
national average ratio of highest to poorest quintiles of just over 8; in the
study settlements it ranges from 20.5 in Vinh Thoi, the most ‘urban’ of the
three, to 13 in My Thoi 1 and 9 in Hoa. Households that rely exclusively on
agriculture are more likely to be in the poor and lower middle-income groups,
although access to land, labour availability and household composition are
equally important factors. Young families with small children may find it
difficult to diversify their income sources and occupations, unlike households
with lower dependency ratios. While in Vinh Thoi land is relatively scarce and
the wealthier groups are those whose income is derived mainly from trade
and services, in My Thoi 1 the wealthier households have larger farms and
their income is mainly from fruit production.
Income diversification into non-farm activities directly impacts on farm-
ing. Investment in fruit trees involves risk as trees can take several years to
mature and become fully productive, and in the meanwhile market prices
may drop. A common strategy is to convert land gradually, intercropping
immature trees with shorter-term crops. However, non-farm sources can
also provide income stability and encourage risk-taking. An example of this
is the GAP (good agricultural practice) production systems, which impose
strict limitations on pesticide and fertiliser use and can be costly compared
to usual systems, but provide benefits in the long term by reducing plant
stress and ensuring access to more strictly regulated (and often better
paying) markets.3 The adoption by farmers of elements of the GAP tech-
niques is still limited in the settlements to better-off farmers who can afford
this long-term investment and can hire labour for the additional work it
requires.
 Fruit production and urbanisation in Vietnam’s Mekong Delta 83
Expanding livelihoods’ spatial boundaries:
migration, mobility and remittances
Income diversification in the study areas involves significant levels of mobility.
It certainly helps that all three settlements are located in close proximity and
within easy access of rapidly growing urban centres such as Can Tho city and
metropolitan regions (HCMC) where Vietnam’s booming manufacturing,
services and construction are concentrated, offering substantial employment
opportunities especially to young people.
About half of the survey respondents’ households have at least one
member who migrates, temporarily or over long periods of time. The over-
whelming majority of these migrants started moving after 1998 when the
transformation of paddy fields into orchards began in earnest. Clearly, the
less labour-intensive fruit played a key part in increasing mobility in the area;
nevertheless, other important factors should not be overlooked. First is the
vertiginous growth of manufacturing, services and construction in Vietnam,
much of it concentrated in the south-eastern region around HCMC. Second
is the dramatic improvement in transport and communications infrastructure,
including the widespread availability of mobile phones that make access to
information on job opportunities in distant locations so much easier. Last but
not least, the educational qualifications of younger generations have much
improved, raising both their skills levels and, perhaps more crucially, their
expectations as workers. As in many other parts of the world, farming is less
and less attractive to young people who see manufacturing as a relatively
more secure job and as an opportunity to see the world beyond the village
boundaries. Indeed, over 70 per cent of migrants in respondents’ households
are young and single, with no major differences in the proportion of men and
women. Demand for women workers has expanded substantially in the past
decade and concentrates in the manufacturing sector, especially in export
industries, and in domestic service, the latter triggered by the growth of a
wealthy urban middle class.
Table 6.3 shows these trends amongst migrants engaged in long-term
mobility (over three months, with an average of over two years for first-time
movers). The increase in the number of movements in the period 2001–2005
is clearly related to the expansion of manufacturing and the demand for
domestic and service workers in urban centres. Between them, HCMC
and provincial towns attract over 80 per cent of first-time migrants and all
third-time migrants. And while income-generating activities rank first as the
purpose of migration, a remarkable one quarter of first-time migrants move
for educational reasons (see Table 6.3). Education is an important investment
in all three settlements, and indeed around 40 per cent of respondent house-
holds that receive remittances use them to fund their children’s schooling.
An emerging trend among young women from poor households is mar-
riage to foreign men, often in Korea and Taiwan. This pattern has increased
since 1999–2000. About 60 women from My Hoa commune have married
84 Hoang Xuan Thanh and Cecilia Tacoli

Table 6.3 Long-term mobility

First trip Second trip Third trip

(%) (%) (%)

Year of moving/ From 1978 to 1990 1 2 –

migrating
From 1991 to 1995 1 2 –
From 1996 to 2000 20 3 –
From 2001 to 2005 78 94 100
Destinations Other village, same 2 2 –
commune
Other commune, same 3 2 –
district
Other district, same province 8 5 –
HCMC 55 56 75
Other province, urban 25 37 25
Other province, rural 6 – –
Abroad 2 – –
Purpose of For business/work 65 83 79
migrating
For market exploration 1 – –
Funeral, wedding, feasts 1 – –
For studying 25 13 21
Other 7 3 –
Average length 24 20 11
of stay (months)
Total 233 63 24

Source: Authors’ household survey, 2006

Taiwanese men, and a similar number of women from Hoa Hung commune
have married Taiwanese and Korean men. In Hoa settlement, at the time of
the fieldwork, ten women were preparing to marry Taiwanese, American and
Canadian men (often of Vietnamese origin). In Vinh Kim commune, about
ten households were marrying off their daughters to foreigners to whom they
had been introduced by local matchmakers or by friends.4
Compared to women, men appear to have a wider range of employment
opportunities, including construction, trade and factory work, and as a result
their movement encompasses a wider geographical area – from neighbouring
communes to overseas destinations – and a variety of forms – from daily
commuting to semi-permanent migration. Household wealth does make a
difference among both men and women, however, with poorer migrants
more likely to work as seasonal wage labourers in agriculture or construction.
Reliance on friends and relatives to provide access to jobs probably reinforces
these differences.
All migrants in the respondent households send remittances to their
relatives at home, regardless of household income level, and, in many cases,
 Fruit production and urbanisation in Vietnam’s Mekong Delta 85
their contribution is a substantial proportion of household income, often as
high as, if not higher than, local earnings. Women migrants are more likely
than men to send money regularly to their relatives (82 per cent against
76 per cent), reflecting what seems to be a universal trend. In relative terms,
remittances are much more important for the poorest groups, where they can
be as much as 3–4 times the local earnings. As may be expected, remittances
are a lower proportion of the incomes of the wealthier groups. In absolute
terms, however, they can be much higher, especially when the migrants are
skilled and have well-paid jobs. Considering the wide fluctuations in fruit
prices in the past few years, it is clear that remittances have a key role
in offsetting the financial risk of relying solely on farming. They have also
provided an important source of capital for the respondent households, and
about one third of recipients, mostly in the wealthier groups, invest this
money in fruit farming.
While migration and remittances are an important dimension of household
income diversification and, in many cases, asset accumulation, they overlap
with growing flows of in-migrants to the three settlements. Leaders in Vinh
Kim commune estimate that 15–20 of the 100 larger fruit traders in the urban-
ising settlement of Vinh Thoi come from elsewhere, mostly from HCMC and
Da Nang city. In Hoa, 30–40 seasonal in-migrants from the rice-farming
delta communes arrive during the slack rice season between November and
February to work as labourers in the orchards, as young local men have either
migrated or moved to non-farm jobs. Sixty respondent households in Hoa,
43 in Vinh Thoi and 39 in My Thoi 1 hire outside labourers regularly or
occasionally, a significant change in traditional family farming.

From village to small market town:

traders and rural development
Many small towns grow out of a close relationship with a surrounding
rural area specialising in high value-added agricultural products. All three
settlements, but perhaps especially Vinh Thoi, are a good example of this,
and highlight the important role of local traders in this process of in situ
urbanisation and rural development.
Private traders dominate fruit marketing in the three settlements. Some of
them are long established in the business, with good market information and
extensive networks throughout the country. The larger traders (vua) located
in Vinh Thoi offer the best prices, while mobile traders collect fruit at the
farm gate and either bring it to the vua or send it to urban markets. This
network benefits small-scale producers and those with limited labour and/or
time, and in many cases is essential to connect them to markets, since large
trading companies are not usually interested in reaching small, scattered
farms. Moreover, vua do not specialise in just one kind of fruit but are able to
match a diverse supply with different demand. They are also able to match
specific supply, especially with regard to quality, to specific market segments:
86 Hoang Xuan Thanh and Cecilia Tacoli

Figure 6.2 Typical market chains of Lo Ren milk fruit, Hoa Loc mango and Nam Roi
pomelo.

best-grade fruit goes to supermarkets and is exported, the second grade is

sent to urban retailers and the lowest grade is sold in local wet markets (see
Figure 6.2).
While some farmers, especially those producing the higher-value milk
fruit, prefer to grade their fruit themselves and thus get better prices, grading
and packaging are the main value-adding activities for traders, and also those
that create much local non-farm employment. Other activities are also
directly related to the fruit trade, for example the transport of fruit to
regional and national destinations, and the production and supply of pack-
aging materials. The fruit markets themselves provide employment such as
portering, guarding vehicles, and driving motorbike taxis in and around mar-
kets (Illustration 6.2). Other services, such as hairdressing and cafés along the
main roads and small grocery shops, cater to both market visitors and local
residents. In the case of the latter, they reflect an increasingly ‘urban’, cash-
based and less subsistence-based lifestyle, while at the same time providing
additional non-farm incomes to different groups, including in-migrants from
the surrounding rural areas. Possibly because of the available opportunities
within the settlement, the thriving village of Vinh Thoi has limited out-
migration, mainly by young people who may have socio-cultural and not only
economic reasons for wanting to move. At the same time, the expansion in
trade and services in the settlement is closely interrelated with growing
demand from farmers from the surrounding rural area, which in turn reflects
their growing incomes.
 Fruit production and urbanisation in Vietnam’s Mekong Delta 87

Illustration 6.2 Agricultural markets drive the transformation of market nodes in the
Mekong Delta from large villages to centres with more urban features.
Photo by Niels Fold

Conclusions
There are three major ways in which urbanisation affects rural development
in the three study settlements. The first is through the increase in demand
from urban consumers for high-value farm produce, in this case fresh fruit.
The second is the expansion of the industrial and services sectors, mostly
urban-based, which have enabled settlement residents to diversify their income
sources, in the process decreasing their dependency on widely fluctuating
agricultural markets and investing in higher-revenue crops. In both these
cases, there is still a relatively clear distinction between what is ‘urban’ and
what is ‘rural’, both in terms of space and in terms of economic activities.
But perhaps more interesting is the fact that the extremely dynamic
trajectories of these settlements suggest that, over time and certainly with
great differences between each settlement, there is an underlying process
whereby the distinction between ‘rural’ and ‘urban’ becomes blurred. As
the complementary links between different sectors (agriculture, trade and
services) grow stronger, they also start to take place within the same
space. Under the stimulus of trade in local crops, other non-farm, typically
‘urban’ activities develop and the local economic base becomes increasingly
88 Hoang Xuan Thanh and Cecilia Tacoli
diversified. Migration does not stop but becomes ever more complex includ-
ing substantial immigration to the settlements, while remittances continue to
feed into the local economy. In situ urbanisation – the third way in which
urbanisation affects rural development – in turn has an overall positive
impact on its surrounding rural region.
The three study settlements, however, are also affected by what happens
in the wider landscape, at the regional, national and global levels. Within
the Mekong Delta region, urban expansion may increase competition over
natural resources, especially for these settlements such as My Thoi 1 which,
for the time being, are strategically located in proximity to urban centres.
This may in the future turn out to expose them to air and water pollution,
competition over land, and disputes over urban waste disposal. Indeed, as in
situ urbanisation progresses, these same issues will need to be addressed
within the settlements by local authorities with increased capacity and com-
petence. At the national and global levels, a growing emphasis on large-scale
commercial farming and the steady advance of supermarkets in low- and
middle-income nations may result in the centralisation of processing oper-
ations which would result in the marginalisation of small market towns.
At the same time, increasing competition and stricter regulations under WTO
rule may effectively push many farmers out of business.
This chapter has described how diversification – of livelihoods and of local
economic bases – is a key element of rural development, and is closely linked
to urbanisation. There are, of course, downsides to the processes documented
here: income inequalities often increase, and natural resources often come
under pressure. Overall, however, the complementary links between rural and
urban development offer a wider range of opportunities that benefit most
groups. There is no doubt that these are also transitional stages within
the context of the very fast-paced process of transformation taking place
in Vietnam. What will be the outcome for future generations will depend
very much on how these transitions are managed at the local, national and
global levels.

Notes
1 Data in this section are derived from interviews with the leaders and key inform-
ants of the study communes and settlements in February 2006.
2 Vietnam still has a household registration system, the ho khau, which until reform
in the mid 1980s was used to control rural–urban migration. Today it no longer
affects every aspect of people’s lives and the restrictions associated with the lack of
residential status are rarely implemented; however, the system has not officially
been abolished (Dang et al., 2003).
3 The full GAP certification is very expensive and would only be interesting for
export-oriented production.
4 Information gathered from interviews with leaders and key informants in the study
communes and settlements.
 Fruit production and urbanisation in Vietnam’s Mekong Delta 89
References
Central Census Steering Committee (CCSC) (2000) Population and housing census
Vietnam 1999 sample results, Hanoi: The Gioi Publishers.
Dang, Nguyen Anh, Cecilia Tacoli and Thanh Xuan Hoang (2003) Migration in
Vietnam: A review of information on current trends and patterns, and their policy
implications, London: DFID.
General Statistical Office (GSO) (2004) Statistical yearbook, Hanoi: Statistical
Publishing House.
Hoang, Xuan Thanh, Nguyen Anh Dang and Cecilia Tacoli (2005) Livelihood
diversification and rural–urban linkages in Vietnam’s Red River Delta, Rural–Urban
Working Paper Series no 11, London: IIED, available from http://www.iied.org/
pubs/pdfs/9518IIED.pdf, accessed July 2008.
Ministry of Agriculture and Rural Development (MARD) (2005) Fruit Industry
Report, December, Hanoi: MARD.
Tien Giang Province (2005) Statistical yearbook, Hanoi: Statistical Publishing House.
Viet Nam Living Standards Survey (VNLSS) (1993), available from http://www.
worldbank.org/html/prdph/lsms/country/vn93/vn93docs.html, accessed July 2008.
Vietnam Development Report (2004) ‘Joint donor report to the Vietnam Consultative
Group Meeting’, Hanoi, 2–3 December 2003.
Vietnam Household Living Standards Survey (2002) available from http://
www.gso.gov.vn/default_en.aspx?tabid=483&idmid=4&ItemID=1843, accessed
July 2008.
Vietnam Household Living Standards Survey (VHLSS) (2004) available online from
http://www.gso.gov.vn/default_en.aspx?tabid=483&idmid=4&ItemID=4343,
accessed July 2008.
Vietnam Living Standards Survey (VLSS) (1998) available from http://www.
worldbank.org/LSMS/country/vn98/vn98bif.pdf, accessed July 2008.
Vinh Long Province (2005) Statistical yearbook, Hanoi: Statistical Publishing House.
7 Agricultural frontier
settlements
Markets, livelihood
diversification and small town
development
Niels Fold and Cecilia Tacoli

Introduction
It is often claimed that small and intermediate urban centres are growing
faster than large cities. While it is true that the average annual population
growth rates of small towns and secondary cities are higher than those
of large cities, many of them stagnate or even lose population. These large
variations masked by aggregate statistics are directly linked to the huge diver-
sity of small towns, and to the importance of local factors in driving urban
change in different countries and regions (Satterthwaite and Tacoli, 2003).
The settlements described in the chapters in this section rely primarily on
agricultural production, as is often the case for small towns and large villages
in low- and middle-income countries where agriculture is an important
component of GDP. A key difference between dynamic growing settlements
and settlements that appear to be declining in both economic and demo-
graphic terms is the relative diversity of the economic base. This, in turn,
is closely linked to the nature of the crops produced in the settlement’s
surrounding area and the opportunities these provide to generate added
value through local processing. Thus the nature of agricultural production,
the value chains dynamics it creates, and the local ability to capture some
of this value and generate non-farm employment all contribute to the settle-
ments’ trajectories.
This chapter explores these issues through the discussion of a number of
interlinking themes. We start with an overview of the organisation of the
flow of agricultural products, with special attention to how the nature of
the crops and the operation of the value chains impact on the economic
dynamics of the settlements. We then examine the economic base of the
settlements, exploring how forward and backward linkages with the produc-
tion of the main crops relate to processes of local economic diversification
that include farm and non-farm activities. Finally, we describe how chang-
ing livelihood opportunities in the settlements increasingly involve some
form of mobility and in many cases result in growing social differentiation
and inequality.
 Agricultural frontier settlements 91
Agricultural markets: framing local socio-economic and
spatial trajectories
The agricultural frontiers described in the chapters in this section can be
assigned to one of two broad categories. The first is a ‘traditional’ frontier
where virgin land is still being cleared (although at a more reduced pace than
in the past) to grow perennial crops: cocoa in Ghana and coffee in Vietnam.
These frontiers are characterised by fluctuations in world market prices that
tend to follow multi-annual cycles driven by demand and supply imbalances.
The second frontier is found in well-established agricultural areas where new
crops and cultivation practices increasingly dominate smallholder farming.
An example is the recent widespread growth of production of fresh fruit and
vegetables, mostly for markets in the Global North but also for expanding
domestic urban markets. Production may also fluctuate (in this case usually
on a seasonal basis) but new forms of contractual relations between small-
holders and buyers create different dynamics compared to the spot market
transactions that take place in most traditional frontiers.
The agricultural frontiers in Ghana and Vietnam have been largely driven
by market dynamics dominated by leading agro-industrial transnational
companies (TNCs) operating in most producing countries in the Global South
and consuming countries in Global North. As a backdrop to the case studies,
this section maps out the dynamics of global markets for the two main cat-
egories of crops in the frontiers (coffee/cocoa and fresh fruit). It then exam-
ines the role of domestic markets and the factors underlying efforts to localise
processing activities.
Both coffee and cocoa have been exported in substantial volumes to mar-
kets in the Global North for more than a century. Current key drivers in this
category of value chains are international traders, primary processors and
branded manufacturers. In the global value chain for coffee, the leading
TNCs are vertically integrated and include both primary processing (roast-
ing) and branding of coffee products while international trading companies
dominate storage and purchasing operations in producer countries (Ponte,
2002). In the global value chain for cocoa, however, there is only limited
functional space for international trading companies. Most global purchase
operations are handled by primary processing TNCs (grinders) who buy
cocoa at origin and produce a highly diversified range of intermediary prod-
ucts that are sold to transnational branded chocolate manufacturers located
in the Global North (Fold, 2002). Both coffee and chocolate – the main end-
product of cocoa – are primarily sold to private consumers by supermarket
chains in the Global North despite a trend towards small retailer sales of
low-volume (but high-value) niche products such as organic or ‘fair trade’
coffee or chocolate.
The market dynamics for the second category of frontier are shaped by
more recent trends in food consumption, namely the increasing demand for
fresh fruit and vegetables. At the global level the key drivers are large retailers
92 Niels Fold and Cecilia Tacoli
(supermarket chains) mostly located in the Global North but increasingly
also present in producer countries in the South (Humphrey, 2006). Super-
markets have taken over the lion’s share of distribution and sales in the
Global North and typically have their own sourcing departments. Alter-
natively, they use specialised importers that are linked to growers/exporters in
producing countries in the Global South. Usually these growers/exporters
operate outgrower schemes supplemented by local independent smallholders
in order to secure a flexible supply capacity to cater for fluctuating demand.
Smallholders may also be organised in producer groups that grow sufficient
volumes for exports and are linked – directly or via intermediaries – to
Northern markets. In addition, smallholder produce may be purchased by
small-scale traders and shipped by local exporters and sold on a consign-
ment basis in traditional wholesale markets in the Global North; this is,
however, a highly insecure link to export markets with very unpredictable
revenues.
Whereas domestic demand for cocoa (i.e. chocolate) and coffee is very
modest in the Global South (except in a few countries), fresh fruit and
vegetables are an important and growing part of daily diets. Hence, export
and domestic markets compete for supplies. Quality products are directed
towards markets offering the highest prices if the necessary infrastructure
and logistics exist. In the case of fresh fruit for export, this includes grading
and packaging facilities, cooling systems, and means of transportation and
shipping (Friedland, 1994).
New standards for fresh fruit in the Global North including product
requirements (shape, size, colour, etc.) and process requirements (cultivation
methods, labour conditions and environmental impact) set up barriers for
exports by smallholders (Dolan and Humphrey, 2004). Compliance with
these standards is premised on sufficient competence – technological, finan-
cial, organisational, etc. It is much easier for corporate capital in agricultural
production to accommodate the standards’ requirements and remain in, or
enter the value chain. Smallholder groups need assistance from intermediaries
such as local private traders, state institutions or civil society organisations
(e.g. local and international NGOs) to set up the required reporting, monitor-
ing and sanctioning procedures. However, smallholders and local traders may
deliberately abstain from engaging with export markets if prices are too low,
risks too high or demand too volatile. An example is that of the fresh fruit
producers and traders in Vietnam’s Mekong Delta described in Chapter 6.
Standards are also becoming more stringent in the global value chains for
traditional crops although they are not as comprehensive and widespread as
in the fresh fruit and vegetable case. Industrial processors – whether coffee
roasters or cocoa grinders – have developed technologies to compensate partly
for quality differences at origin. Certain intrinsic quality parameters – for
instance caused by particular varieties or agro-environmental conditions –
are impossible to de-localise but others can be reduced or almost removed
(Fold and Ponte, 2008). In essence, these innovations reduce differences
 Agricultural frontier settlements 93
caused by country-wise variations in cultivation, harvesting and post-harvest
handling practices. For instance, a changed sequence in the grinding process
removes the adverse effects of insufficiently dried and fermented cocoa beans
and new technology in coffee roasting removes the excessively bitter flavours
in some ‘Robusta’ coffee beans (e.g. those coming from Vietnam). However,
quality still varies between different countries of origin and is reflected in
price differentials in the global commodity markets.
At the same time, these trends reinforce the global dichotomy between raw
material production in the Global South and processing in the Global North,
since they require huge investments in fixed capital equipment. Blending of
coffee or cocoa of different origins is essential for brands that need a stable
taste and flavour, and therefore it makes sense to control and carry out these
processes near the final market (Gibbon, 2001). Substantial grinding of cocoa
takes place in the Global South, but the processing technology is relatively
simple and the purpose is primarily to obtain a purified product before
shipping – and to exploit the generous tax incentives to processing industries
offered by host governments. As for coffee, only dried (‘green’), cleaned, sorted
and polished beans are shipped since roasted coffee beans are heavily exposed
to quality deterioration during transport.
Despite these structural constraints, local organisational capacity is clearly
improving in the coffee and cocoa chains where local trading companies
increase the consistency of their operations and gradually strengthen their
reputation as stable suppliers. In the Vietnamese coffee chain, a growing
number of local trading companies have expanded into value-adding activ-
ities such as cleaning, grading, packaging and shipping, gradually conquering
more functional space from international trading companies. In contrast,
exports and quality control of Ghana cocoa are completely regulated by the
state, which has guaranteed a steady supply of high-quality cocoa to the world
market. In both chains, social differentiation leading to land accumulation
and dispossession seems to be kept under a degree of control by state regula-
tory mechanisms. In Vietnam, formal restraints on the size of individual
land ownership are an effective barrier. In Ghana, the pan-seasonal and pan-
territorial purchasing price, in addition to a state-controlled company that
functions as a ‘buyer of last resort’, secures a stable market for all small-
holders, including geographically and economically marginalised groups.
Hence, traditional migratory flows of labour from the north of Ghana to the
cocoa regions in the south continue to follow the existing seasonal pattern
that adapts labour supply to the main periods of demand for labour on the
cocoa farms, i.e. the main harvest period from October to February.
As for the fruit chain, the picture is somewhat different. In the Ghana
pineapple frontier, production is primarily for export and TNCs have near
total control of the backward and forward linkages for the new pineapple
varieties that now dominate global markets. Smallholders that do not have
the capital to invest in the new varieties, and have no access to storage and
processing infrastructure, are effectively cut off from the value chain, while
94 Niels Fold and Cecilia Tacoli
migrant smallholders with only secondary rights to land have lost out to
commercial plantations. Perhaps unsurprisingly, the case study locations have
become areas of substantial out-migration and wage agricultural labour.
In contrast, domestic urban markets are the main outlet for fruit and
vegetables produced by smallholders in Vietnam’s Mekong Delta. One reason
for this is the sudden and no doubt painful contraction of the Chinese export
market which, until 2001, had absorbed about 80 per cent of Vietnamese fresh
fruit. Equally important, however, is the rapid improvement in the living
standards of both rural and urban Vietnamese households, with related
transformations in food consumption. Also important in understanding the
difference between Ghana’s pineapple frontier and fruit production in Viet-
nam’s Mekong Delta is the function of local traders. While their role in
Ghana is marginal, in Vietnam they dominate transactions and have estab-
lished most of their processing activities within local settlements, thus con-
tributing substantially to the diversification of the local economic base. The
Mekong Delta fresh fruit frontier thus suggests that domestic urban markets
are a better option for smallholder agriculture. It should be noted that this is
by no means an isolated finding: longitudinal studies in drylands Africa
(Tiffen, 2003) and in West Africa (Club du Sahel, 2000; Toulmin and Guèye,
2003) have also shown that expanding domestic urban markets are key
drivers of smallholder production even in the absence of the spectacular rates
of economic growth of countries such as Vietnam.

Settlements’ trajectories: geographical, economic and

institutional aspects
Both the economic base of the settlements and the livelihoods of their resi-
dents are closely interlinked with chain-specific dynamics. But beyond this
broad generalisation, the variations between and within the case studies’ loca-
tions point to the need to take into account a number of specific factors that
contribute to shaping these differences.
Starting from the settlements, it is clear that location is important for
economic development, but only when it is accompanied by access to good
road and transport infrastructure. Distance is often a less powerful constraint
on economic development for remote settlements than isolation. Moreover,
while most settlements manage to have links to one local urban centre, con-
nection to a network of rural and urban settlements provides a wider scope
for economic and social interactions that benefit both the settlement and its
residents (Douglass, 1998). The Mekong Delta case study shows how these
networks also shape, and are shaped by, complex trade relations that span
local, national and international markets. This is mirrored by the greater
mobility, and variety of destinations, of the residents of well-connected
settlements compared to the residents of relatively isolated settlements.
Settlements with a highly diverse range of activities generally fare better in
terms of population and economic growth. The nature of the main crop grown
 Agricultural frontier settlements 95
locally clearly has an impact on the ways in which diversification evolves.
Hence, crops that do not require local processing and can be transported
in bulk do not create related local employment. Higher-value and perishable
crops such as fresh fruit which require grading, packaging and rapid trans-
port to final markets can, in contrast, create local employment. This depends,
though, to a large extent on whether markets are controlled by local traders
based in the settlements or by global buyers such as supermarkets and
national marketing boards. Moreover, crops with pronounced seasonal pro-
duction cycles affect non-farm activities: in Ghana’s cocoa frontier, for
example, traders bring in manufactured goods and provide services at the
height of the cocoa season when indigenous as well as seasonal migrant
labourers flock to the local settlements and are more likely to have money in
their pockets. Once the season is finished, farmers, labourers and traders
move elsewhere.
The most dynamic settlements in both Ghana and Vietnam are more
‘urban’ than ‘rural’; although their economic base is closely interlinked with
agriculture, and with specific crops, their residents are more likely to be
employed in non-farm sectors and rely on purchased goods and services
rather than on subsistence. The latter is important in that it spurs the growth
of services that are not directly related to farming and depend more on
population density and income – for example, cafés and restaurants, hair-
dressers, personal transport providers. Such services are an important source
of income for the poorer groups. On the other hand, in all the settlements
described in this section the non-farm economy is closely linked to local
agricultural activities. Hence, even if the residents of the very dynamic Vinh
Thoi settlement in the Mekong Delta are less and less engaged in farming, the
economy of the settlement and the livelihoods of its residents depend heavily
on fresh fruit produced in the surrounding region. To a large extent the
fortunes of the settlement and its residents reflect those of the specific com-
modity. As the settlements in Ghana’s pineapple frontier suggest, however,
the structure of agricultural markets can be far more important than the
commodity itself.
Most of the settlements described in this section are classed as ‘rural’ and
even the most rapidly growing ones are just about to attain urban status of
the lowest order but are not quite there yet. They often provide basic services
to their inhabitants and the surrounding population – primary schools,
primary health care – but not much more. Local governments typically have
limited decision-making power and capacity, and depend on district or
provincial authorities for their revenue. Local institutions, however, can play
important roles in providing basic infrastructure such as market places, and
access to market information and regulations. They can also play a key role in
determining access to assets such as land, and thus have a major impact
on local livelihoods. In Vietnam, the redistribution of farmland to rural
households in 1998 is at the root of the relatively equitable access to land in
the country, despite the emergence of land markets. In Ghana, by contrast,
96 Niels Fold and Cecilia Tacoli
customary land tenure systems still prevail in the rural areas and traditional
chiefs’ decisions can make it difficult for migrants to obtain secure tenure.
In the case of the pineapple frontier, migrants are the first group to lose access
to land as commercial plantations enter the scene.

Livelihood transformations: income diversification and mobility

Reflecting settlements’ dynamism – or lack of it – the livelihoods of their
residents can be described as ranging from mono-activity (usually farming)
and mono-location, to multi-activity and multi-location. Households that
rely on a diverse portfolio of income-generating activities, often involving
different locations, are generally less vulnerable to shocks such as the sud-
den drop in the prices of their main crop. Moreover, households that can
rely on the safety net of non-farm incomes are usually more prepared to
invest in (and risk) farming innovation and intensification (Hoang et al.,
2008).
All the case studies suggest that access to land is an important factor
underlying income diversification. In Ghana, customary tenure systems may
force migrants into trade, wage labour and services activities, while in
Vietnam’s Mekong Delta it is more likely to be population density that
decreases the availability of farmland. The transition from farming to diversi-
fied livelihoods is a complex one, however. It can be the cause – and the result –
of increasing wealth and accumulation of assets; but it is also in many cases
the survival strategy of the poorest groups. As such, it defies generalisation
and calls for attention to local contexts and diversity.
A similar point can be made for mobility. In the stagnating and declining
settlements where local opportunities for income generation are shrinking,
out-migration is an obvious option, although in many cases not one which
is available to the poorest groups. On the other hand, the most dynamic
settlements attract in-migrants who set up enterprises or work as agricultural
labourers. This does not mean, however, that in these settlements there is
no out-migration. Quite to the contrary, there are often large flows of out-
migrants from all income groups for whom mobility implies access to better
options than those available locally – stable factory jobs, secondary or uni-
versity education, and so on. At the same time, employed migrants contribute
substantially to local household incomes and can be said to play a significant
role in the dynamism of their home settlement.

Conclusion
In all the settlements described, agricultural production is the mainstay of
the local economic base and of the majority of local livelihoods. There are
significant variations, however, in the nature of the crops, in the systems of
production and, crucially, in the structure of the agricultural markets to
which farmers have access. There are also some similarities; in the most
 Agricultural frontier settlements 97
dynamic settlements, the majority of residents are engaged in non-farm
occupations and trade is especially important, both at the settlement and at
the household levels. Trade and mobility clearly benefit from good road and
transport infrastructure providing access to the surrounding rural region
and a network of larger cities. The growing settlements in Vietnam and
Ghana described in this section thus play a key role in linking rural and
urban spaces. They are also the places where agriculture, services and manu-
facturing activities converge and interact. While these linkages and inter-
actions are shaped by the local context, however, national-level agricultural
policies and more general economic growth strategies have had an impact
on the trajectories of rural small towns in both Ghana and Vietnam.
Moreover, global-level transformations in international trade and production
systems have had – and will continue to have – a dramatic impact on these
settlements.

References
Club du Sahel (2000) Urbanization, rural–urban linkages and policy implications for
rural and agricultural development: case study from West Africa, Paris: SAH/
DLR(2000).
Dolan, C. and Humphrey, J. (2004) ‘Changing governance patterns in the trade
in fresh vegetables between Africa and the United Kingdom’, Environment and
Planning A, 36: 491–509.
Douglass, M. (1998) ‘A regional network strategy for reciprocal rural–urban linkages:
An agenda for policy research with reference to Indonesia’, Third World Planning
Review, 20(1): 1–34.
Fold, N. (2002) ‘Lead firms and competition in “bi-polar” commodity chains: Grinders
and branders in the global cocoa-chocolate industry’, Journal of Agrarian Change,
2(2): 228–47.
Fold, N. and S. Ponte (2008) ‘Are (market) stimulants injurious to quality? Liberaliza-
tion, quality changes and the reputation of African coffee and cocoa exports’,
in N. Fold and Larsen, M.N. (eds), Globalization and restructuring of African
commodity flows, Uppsala: Nordiska Afrikainstitutet, pp. 129–55.
Friedland, W. (1994) ‘The new globalization: The case of fresh produce’, in Bonanno,
A., Busch, L., Friedland, W., Gouveia, L., and Mingione, E. (eds), From Columbus
to ConAgra: The globalization of agriculture and food, Lawrence: University Press
of Kansas, pp. 210–31.
Gibbon, P. (2001) ‘Upgrading primary production: A global commodity chain
approach’, World Development 29(2): 345–63.
Hoang, X.T., Dinh, T.T.P. and Nguyen, T.H. (2008) Urbanization, fruit production and
rural livelihood transformations in the Mekong Delta, Rural–Urban Interactions
Working Paper 14, London: IIED.
Humphrey, J. (2006) ‘Policy implications of trends in agribusiness value chains’,
European Journal of Development Research, 18(4): 572–92.
Ponte, S. (2002) ‘The “Latte Revolution”? Regulation, markets and consumption in
the global coffee chain’, World Development, 30(7): 1099–122.
Satterthwaite, D. and C. Tacoli (2003) The urban part of rural development: The role of
98 Niels Fold and Cecilia Tacoli
small and intermediate urban centres in rural and regional development and poverty
reduction, Rural–Urban Interactions Working Paper Series No 9, London: IIED.
Tiffen, M. (2003) ‘Transitions in sub-Saharan Africa: Agriculture, urbanization and
income growth’, World Development, 31(8): 1343–66.
Toulmin, C. and Guèye, B. (2003) Transformations in West African agriculture and the
role of family farms, Drylands Issue Paper 123, London: IIED.
8 Handicrafts in Vietnam
Rural urbanisation in the
Red River Delta
Katherine V. Gough and
Dang Nguyen Anh

Introduction
Vietnam has been changing rapidly since the introduction of Doi Moi in the
1980s. The aim of the reforms is to create a hybrid third way, a socialist-
oriented market economy which is in between a planning system and capital-
ism (Smith and Scarpaci, 2000). The economic reform programmes have
included opening up to international capital which has contributed to annual
growth rates averaging about 9 per cent for most of the 1990s. Vietnam still
has single party rule, though, and the state continues to control a large
proportion of national production (Dixon, 2003). The development of non-
agricultural activities in the countryside and rural industrialisation are central
policies of the Vietnamese government. The aim is to boost the incomes of
rural households, stimulate demand for consumer products, and above all
create jobs and limit the rural exodus to the towns and cities (Anh, 2005). One
such policy is the promotion of handicrafts in rural areas. Handicrafts are
considered an effective measure to create in situ employment, reducing
poverty and the income gap between urban and rural regions which in turn
should help curb the flow of workers to urban areas. The Ministry of Agri-
culture and Rural Development has the target of helping to develop a par-
ticular craft in each village which should contribute a large part of the
inhabitants’ annual income. Since Doi Moi was introduced, a large number of
rural households have developed new craft and industrial activities, or have
diversified existing handicraft production, as a consequence of these policies.
In the Red River Delta, in northern Vietnam, many households have long
been engaged in handicraft production. Typically, a settlement has become
specialised in the production of one handicraft, the skills being passed down
through the generations. There is a wide range of handicraft types including
bamboo and rattan weaving, silk weaving, wood carving, conical hat making,
ceramics, pearl inlaying, embroidery and lacquerware, etc. On entering a
settlement it does not take long to discover what a village’s main handicraft
is, as people of all ages are visibly engaged in their work in open-fronted
rooms and in the open air on their plots. On entering the houses more hidden
activity is revealed, sometimes in the form of quite substantial workshops
100 Katherine V. Gough and Dang Nguyen Anh
and some of it damaging to the environment and health. In the current Doi
Moi era, what is happening to this handicraft production? How are the
products evolving over time and who is engaged in production? Do handi-
crafts provide a viable income source? How do the experiences of villagers
engaged in differing handicrafts vary? What is the nature of rural–urban
dynamics in these handicraft settlements?
These and other questions are discussed in this chapter which focuses
on the changing nature of handicraft production in the Red River Delta. It
draws on data collected in two settlements where differing handicrafts are
produced: embroidery in Quat Dong and lacquerware in Ha Thai. After a
general introduction to the Red River Delta, the cradle of craft villages in
Vietnam, the changing nature of handicraft production since the nineteenth
century is outlined. Subsequently, the implications of these changes for liveli-
hoods and settlement are discussed. It is shown how handicrafts have long
been an important aspect of livelihoods in the Red River Delta but that their
contribution has waxed and waned under differing historical conditions
illustrating an ability to adapt to the politico-economic situation. Today the
importance of handicrafts to livelihoods varies according to the nature of the
craft but clearly contributes to increasing wealth and greater inequality. This
is illustrated by the changing social and physical nature of the settlements
which are becoming as urban as they are rural.

Craft villages in the Red River Delta

Vietnam currently has more than 2000 craft villages, which are defined as
villages with over 20 per cent of the households involved in handicrafts or over
20 per cent of their income derived from handicraft production. There are an
estimated 1.4 million handicraft production households, almost 10 per cent of
the total number of households in Vietnam, who are engaged in making a wide
range of crafts. The majority of these (80 per cent) are located in the north of
Vietnam. The total production value of craft villages grew by 15 per cent per
year from 2001 to 2004 reaching around VND 9000 billion (616 million US
dollars) in 20041 (MARD, 2005). The export value of the handicrafts
increased from 235 million dollars in 2002 to 450 million dollars in 2004. The
crafts also carry important cultural and social values of the nation.
The Red River Delta is the region with the highest density of craft villages,
having about 1000, and is the most densely settled region of Vietnam. The
villages produce goods for use in the everyday life of villagers and urban
consumers, such as processed foodstuffs, religious objects, industrial products
and building materials, as well as producing handicrafts for domestic use and
export. The inhabitants of the craft villages typically enjoy income levels four
times higher than those in agricultural villages and have been able to invest in
property and improve their living conditions (Yen, 2005). DiGregorio et al.
(2003) claim that since land redistribution was completed in 1993, the Red
River Delta has experienced agricultural diversification and intensification,
 Handicrafts in Vietnam 101
economic specialisation, extensive home construction and the broadening
appeal of urban culture and consumption standards. This they refer to as
a process of ‘rural urbanisation’.
Within the delta, government plans include both maintaining and promot-
ing traditional craft villages and creating new ones. Plans for the handicraft
sector have been drawn up at province, district and commune level. The Min-
istry of Industry has programmes to boost craft villages, especially those with
famous products such as pottery, ceramics and leather goods, through con-
necting them with markets and encouraging the hiring of local labour and the
use of local materials. The aim is to develop traditional crafts to produce
goods for the domestic and export markets and develop service activities such
as transportation, power and water supply, technical services for agriculture
and livestock farming, and financial and training services. Assistance can be
obtained to facilitate access to capital, production technologies and scientific
advancement. Furthermore, every village can set up a fund to receive assist-
ance from local and international organisations to promote craft production
and marketing.
This research focused on Ha Tay Province which has one of the highest
concentrations of craft villages, reportedly 120 in total. Proximity to Hanoi
means that the villages have the possibility of being well connected to local
and international markets. Two settlements were selected for the study, both
of which have well-established handicraft industries with links to the global
market. Quat Dong village, situated in the commune of the same name, is
a traditional craft village located approximately 21 km south of Hanoi (see
Figure 8.1). Inhabitants claim that Quat Dong is the settlement in the Red
River Delta where embroidery was first practised over 400 years ago. In 2006,
Quat Dong had an official population of 4360 people (2100 males and 2260
females). Ha Thai, located in the commune of Duyen Thai about 17 km
southeast of Hanoi, is a settlement of 2950 people (1450 male and 1500
female). Lacquerware production is the main occupation of the majority of
the population of Ha Thai, who have been engaged in some form of painting
for over 200 years. Officially, the inhabitants of the settlements studied are
those who are registered as living there; this means that temporary migrants
are not included and hence did not feature in the questionnaire survey.

Changing nature of handicraft production

Quat Dong and Ha Thai have been craft villages for generations. The inhabit-
ants have worked in both agriculture, predominantly growing rice, and
making handicrafts. How they divided their time between the two activities
has varied over time. Three broad time periods can be identified: first, a
colonial period consisting of French domination from the nineteenth century
to 1945 followed by a turbulent phase (1945–1954) which included Japanese
occupation; second, a co-operative period which consisting of a phase where
war dominated (1954–1975) and a post-war phase (1975–1985); and third, a
102 Katherine V. Gough and Dang Nguyen Anh

Study area
City

Figure 8.1 Map showing location of study settlements in the Red River Delta.

Doi Moi period consisting of the initial years of Doi Moi (1987–1997) and a
later phase where Doi Moi was intensified (1997–2007). The ways in which
handicraft production has varied in relation to these time periods will be
briefly outlined below.
 Handicrafts in Vietnam 103
Embroidery boomed under the French in the nineteenth century as there
was demand for exports of embroidered bed sheets and table cloths in par-
ticular. The thread used was imported from France. When the Japanese took
control in 1940, this market disappeared and the embroidery making dimin-
ished. After the French returned and re-occupied the north five years later,
trade with France resumed and there was substantial demand from the
French colonial residents for embroidered products. Production thus
increased once again especially for cloth embroidered with dragons. During
the same time period, the primary handicraft production in Ha Thai was
painting pictures and decorations for temples and houses, for example
painted paper boxes which were used to cover tea pots and painted covers
to keep flies off food. The pictures being painted at this time were principally
of rural scenes and the products were mainly sold on the local market.
Under the communist Vietminh, co-operatives were established and
exporting was organised through officials/committees operating at various
levels. According to the villagers, the export of embroidered products fell due
to the extensive bureaucracy and because government officials were not con-
cerned about the quantity they exported as they ‘had rice to eat’ (i.e. had
salaries) regardless of the amount sold. Cloth and thread were imported from
China but the variety was greatly reduced, which jeopardised the quality of
the embroidery making it difficult to export. Lacquerware underwent a simi-
lar transformation with the establishment, then collapse, of the co-operatives
and state-run brigade in Ha Thai. The cooperative system of production
failed to take advantage of the potential benefits of working as a group,
resulting in low productivity and disregard for product quality as the profits
were shared equally. In the early 1980s, the craft cooperatives dissolved and
production was taken over by households and individuals who managed to
survive using their skills and family labour.
Since the introduction of Doi Moi in 1986, the opportunities for exporting
have increased once again and the lifting of import restrictions has improved
the range of inputs for handicraft production. Consequently, there has been a
boom in embroidery and lacquerware exports. Production of lacquerware
has increased rapidly to the extent that many of the lacquerware producers in
Ha Thai no longer farm but devote their time totally to handicraft produc-
tion. In contrast, although the time spent on embroidery work has increased
in Quat Dong, rice farming is still practised. This brief historical overview
has shown how the importance of handicrafts to livelihoods has changed
over time in relation to the macro-institutional and economic conditions. The
nature of the products and production process have also changed as the
demand and the inputs available have changed, and these are outlined below.

Embroidery
Embroidery requires few raw materials, just the thread, needle and cloth,
all of which are today produced in Vietnam. When a picture is being
104 Katherine V. Gough and Dang Nguyen Anh
embroidered the cloth is stretched over a wooden frame and the embroiderer
sits on the floor (Illustration 8.1). For smaller pieces of embroidery, such as
bag decorating, the cloth is held in the hand. The thread used to embroider is
made in Vietnam, much of it being produced locally (using chemical dyes) by
a few families. There are just two stages in the embroidery process: first the
design has to be printed on the cloth; second, the product is embroidered.
Embroidery work in Quat Dong was initially a male-dominated activity. The
skills were not passed on to women for fear that if they married outside the
settlement they would pass the closely guarded skills on to other villages.
Today, however, embroidery is predominantly done by women and there is
no longer a concern about the skills spreading. Since Doi Moi there are more
economic opportunities for men and embroidery is seen principally as an
occupation for uneducated women. Skills are passed through the generations

Illustration 8.1 An elderly woman engaged in embroidery in her home in Quat Dong.
Photo by Katherine V. Gough
 Handicrafts in Vietnam 105
in the home. Outside workers are rarely employed because of the highly
skilled nature of the work.
Much of the production in Quat Dong is done to order. The production
process is controlled by a few entrepreneurs who have managed to build up
contacts with buyers placing orders. An advance is requested, usually about a
third of the final price, which is needed both as a security and to fund the cost
of the raw materials. On obtaining a contract, the entrepreneurs outsource
the work to other households in the settlement, supplying them with the
necessary raw materials. Many of those doing the embroidery have no idea
where their products are destined and are not interested in being responsible
for the marketing as they perceive themselves as ‘simple farmers’. Even those
commissioning within the village rarely export directly but operate through
middlemen. These middlemen usually come to collect the completed work in
person. Sometimes the embroidered products are sent by post but cases were
reported of customers abroad, especially Vietnamese migrants, placing
orders which they never collected or paid for.
The local Vietnamese market for embroidered pictures and objects has
declined over the years. Most of the embroidery produced is sold to foreign-
ers, either to tourists visiting Vietnam or for export. Recent orders have come
from countries including France, Germany, Sweden, US, Japan, China and
South Korea. Some of the products are clearly designed for a foreign market,
such as small decorations, embroidered with ‘Bienvenue’ or ‘Amor’ to hang
on the wall or cover a jar. The inhabitants of Quat Dong were concerned that
if the SARS virus were to return, Vietnam might receive fewer foreign visitors
which would affect their business. Few tourists visit Quat Dong despite it
having been labelled by the province as a ‘craft tourist village’ in 2002, and
there are no organised tours or showrooms. Craft tours are especially popular
with Asian tourists who prefer to visit the pottery and silk villages close to
Hanoi with their organised workshops and stores.

Lacquerware
The production process of lacquerware has changed substantially over the
years and although still labour-intensive it increasingly relies on modern
inputs and machinery. The main inputs are bamboo, wood, lacquer paint,
mud and ground rock, and increasingly chemicals which are added to the
lacquer paint to speed up the drying process. The production process involves
a number of distinctive steps which require a range of techniques. The wood
or bamboo base of the product is made to order by local craftsmen. The
object is then sanded down before being covered in gauze. The object is then
coated with several layers of mud, which comes from the nearby river. Once
the object has dried, several layers of lacquer are applied, either by hand or
by using a spray machine. The finishing touches added to the product then
vary according to its nature. In the case of pictures, the painting is done by
hand and if pearl inlaying is required this is carried out in another village in
106 Katherine V. Gough and Dang Nguyen Anh
the Red River Delta. The final products are then polished several times using
machines and copious quantities of water.
Previously, one household would carry out all of the stages within their
home. Today many households specialise in one (or two) stage(s) of the
production process. It is common no longer to rely on only family labour but
to employ workers, especially to carry out the toxic lacquerware painting. It
takes only two to three months to become a well-trained lacquerware worker.
One third of households in the sample hired migrant labour and a few busi-
nesses have grown to employ as many as 20 workers. The workshop premises
have been expanded to provide the necessary space for the employees (Illus-
tration 8.2) who sometimes not only work in these spaces but also sleep in
them. Some of the larger enterprises outsource tasks to other households in
the village. The owners will always ensure that they put the finishing touches
to the items to control the standard of the finished product. Many house-
holds receive direct orders for work but as they operate through middlemen
they do not actually know where their goods are destined. The need for
constantly developing new designs in order to try to maintain, or even
increase, their share of the market is widely recognised by the lacquerware
producers. They are highly competitive and attempt to hide new ideas from
other producers.
Lacquerware is especially popular among tourists, who are attracted by the
colourful and interesting shapes of the products which are displayed in open-
fronted shops in the major tourist cities. Many products are exported directly;
one example of a company buying lacquerware is ‘Far Eastern’ which is based
in Hanoi and sells to France and Japan. Countries from which the lacquer-
ware producers have received orders include USA, Italy, France, Sweden,
Germany and Japan. When an order has been completed, the person who
commissioned it will usually collect it in person; if they approve of the quality
they pay, or if it is not good enough they request that the product be redone.
Ha Thai is listed on possible tourist tours by some agencies in Hanoi, though
tourists rarely take these tours. The few tourists who do visit Ha Thai do so with
their own guide and enter craft-producing homes wherever they happen to be
invited in. There are a couple of showrooms in Ha Thai but their role is to
show the range of products to potential exporters rather than display prod-
ucts for tourists to buy. The home market counts for a relatively small propor-
tion of sales but vases, which are placed on altar pieces in the home or temples
where ancestors are worshipped, are popular among the Vietnamese.

Implications of changes in handicraft production for livelihoods,

mobility and settlement
Having outlined the nature of handicraft production for embroidery and
lacquerware, we now turn to look at the implications for livelihoods and
settlement in the Red River Delta. Particular attention is paid to drawing out
the similarities and differences between the two handicrafts.
 Handicrafts in Vietnam 107

Illustration 8.2 Workers engaged in lacquerware production in a workshop in

Ha Thai.
Photo by Katherine V. Gough

Livelihoods
Handicrafts are central to livelihoods in both Quat Dong and Ha Thai but
they differ in the ways in which they are combined with other occupations. In
Quat Dong, farming is continued alongside handicraft production whereas in
Ha Thai, lacquerware production has largely replaced farming. Thus in Ha
Thai there is no seasonal nature to handicraft production whereas in Quat
Dong the production is seasonal due to farming commitments. However, in
Ha Thai the land has not been totally abandoned; many inhabitants have
rented out their land, receiving payment in exchange for the use of the land.
There are a range of reasons as to why this difference has emerged. The
production process of lacquerware requires greater capital input and gener-
ates a greater profit. The demand for lacquerware comes from both the
108 Katherine V. Gough and Dang Nguyen Anh
domestic market and for export and is considered to be relatively stable. The
embroidery, on the other hand, is mainly for export and this market is less
stable. Embroiderers continue farming rice so that if the demand for their
handicraft work falls, they can still feed their families. Embroidery is also an
activity which is easy to take up and put down again, filling in the spare hours
when there is nothing to do on the farm. Lacquerware production, on the
other hand, consists of a number of stages which have to be passed through
within a certain time-frame, making it harder to put aside if work on the
farm calls.
There is also a clear distinction between the two villages in how these
differences are played out in terms of both gender and generation. As Table 8.1
shows, embroidery is predominantly carried out by women; although some
men do engage in embroidery they are in the minority. In Ha Thai, it is most
common for men and women to engage equally in lacquerware production;
hence the work is not dominated by either gender. In both settlements,
women are predominantly responsible for farming though in almost a quar-
ter of cases men and women are deemed to be equally responsible. A notice-
able difference is that whereas in Quat Dong only 6 per cent of households are
no longer farming, in Ha Thai 27 per cent no longer farm. The differing
incomes generated by the two handicrafts are reflected in only 10 per cent of
households in Ha Thai engaging in other income-generating activities
(besides handicraft and farming) whereas almost 40 per cent in Quat Dong
do. These are mainly male household members who work as drivers, masons,
etc., at times even commuting to Hanoi.
There are also clear differences in the attractiveness of the respective
handicrafts to young people (Table 8.2). Three-quarters of respondents in

Table 8.1 Participation in income-generating activities in Quat Dong and Ha Thai

by gender

Activity Participation Quat Dong Ha Thai

Craft Predominantly male 6 19

Predominantly female 81 27
Equal male/female 13 54
Farming Predominantly male 17 4
Predominantly female 54 46
Equal male/female 23 23
No one 6 27
Other work Predominantly male 21 8
Predominantly female 12 2
Equal male/female 4 0
No one 63 90
Sample size 100 100

Source: Authors’ household survey, 2006.

 Handicrafts in Vietnam 109

Table 8.2 Attractiveness of handicraft to young people in Quat Dong and Ha Thai

Quat Dong Ha Thai

Degree of attractiveness Very attractive 32 75

Moderately attractive 54 25
Not attractive 14 0
Most attracted Male 0 4
Female 88 6
Equal 10 90
Sample size 100 100

Source: Authors’ household survey, 2006.

Ha Thai said that the youth were very attracted to working in the handicraft
compared with only a third in Quat Dong. The gender difference also
emerges very clearly with 90 per cent in Ha Thai claiming that the handicraft
is equally attractive to men and women whilst a similar number in Quat
Dong maintain that the handicraft is only attractive to women. Although
men traditionally engaged in embroidery, most young men are reluctant to
engage in an activity which they consider is traditional, back-breaking, a
strain on the eyes, and is combined with farming which the youth also find
unattractive. Despite the hopes of their parents that the embroidery tradition
of their ancestors will continue, many young women do not plan to continue
the handicraft work either. Embroidery is seen as being traditional work
using simple materials and hence as being the domain of young women who
are unable to obtain an education or alternative employment. Lacquerware
production, on the other hand, is seen as a more modern activity which uses
more modern equipment and materials. It also generates a higher and more
consistent level of income.

Mobility
Some households in both settlements have members who have migrated.
In Quat Dong this was the case for a quarter of households whereas in Ha
Thai it was reported as only 6 per cent. In most cases only one family member
had migrated though for a few (5 per cent in Quat Dong and 1 per cent in Ha
Thai) it was two or three members. This reflects the greater attractiveness of
lacquerware production (to both genders) versus embroidery. One con-
sequence of this difference is the greater emphasis placed on education in
Quat Dong. Young people in Ha Thai have access to a relatively secure and
desirable income source resulting in their having less incentive to continue in
their studies; 6 per cent of households in Quat Dong, but none in Ha Thai,
had a member living away from the household who was studying. Remit-
tances are an important source of income for only a few households: 9 per cent
110 Katherine V. Gough and Dang Nguyen Anh
in Quat Dong and 6 per cent in Ha Thai. The migrants sending money are
mainly young men and the money is predominantly used to cover daily
expenses.
In Ha Thai there is also considerable in-migration as well as out-migration
from the settlements. Migrants from other communes and districts come to
the village looking for work in the handicraft business. They are much more
likely to find work in Ha Thai as there is a greater need for hired labour and
because the skills involved in lacquerware production are more easily learnt
than embroidery. The migrants are predominantly northerners who come to
stay for a couple of years in order to save money before returning to their
hometown, often to marry. They work long hours, seven days a week for a
daily wage of 20,000 dong. They have to pay for their food out of their wage
but live in the workshops to save money, despite health hazards from the
fumes. The number of migrants coming into Ha Thai has risen in recent years
as the demand for labour has increased, both due to the industry outgrowing
the local labour supply and as the local population has been increasingly
reluctant to engage in the more hazardous parts of the production process.
In 2006 it was estimated that migrants made up about 10 per cent of the
population, i.e. about 300 people.

Settlement
The changing fortune of handicraft production has influenced the settlement
characteristics, not only in terms of the population but also in the physical
nature of the buildings. Income generation from the handicrafts has enabled
some households to invest in improving their homes resulting in many of the
traditional one-storey buildings being replaced by multi-storey houses (Illus-
tration 8.3). As Table 8.3 shows, half or more of the houses have four rooms
or more. Only one household interviewed in Ha Thai, but 11 in Quat Dong
are still living in single-room dwellings. In Quat Dong it is clearly the sub-
contractors who have built the finest houses. In Ha Thai, the symbiotic rela-
tionship between home and work (Gough and Kellet, 2001) can clearly be
seen; handicrafts being made in the home are generating income which many
households have invested in expanding their home thus creating more space
for craft work. The lacquerware products are displayed on the ground floor

Table 8.3 Size of houses in Quat Dong and Ha Thai

House size Quat Dong Ha Thai

Large (4 + rooms) 50 59
Medium (2–3 rooms) 39 40
Small (1 room) 11 1
Sample size 100 100

Source: Authors’ household survey, 2006.

 Handicrafts in Vietnam 111
of the multi-storey houses where household members can be seen putting
finishing touches to the products. The more hazardous parts of the produc-
tion process are generally carried out on the top floor where there is more
ventilation and the activity is out of sight.
The handicraft production process can have a negative impact on the well-
being of the workers and the local environment. Although embroidery work
is not polluting for the environment, it is not without health hazards;
embroiderers complain about failing eyesight from working in poor light and
of back problems from bending over the frames. Elements of lacquerware
production, however, are highly toxic providing a threat to both the local
environment and the health of the workers. Chemicals added to the lacquer
result in the painting and spraying being particularly unhealthy tasks. Typic-
ally it is migrant workers who are engaged to fulfil these tasks as the owners
do not want to expose their family members to the lung and skin problems
that are frequently reported. Due to the potential health risks, children and
the elderly rarely participate in lacquerware production, whereas they are
heavily engaged in embroidery work. Some households in Ha Thai have built
chimneys to try to extract the fumes from the lacquer and expel them into the
atmosphere. However, this is beyond the capacity of the smaller enterprises
and especially in winter, when there is a lot of fog and damp air, the pollu-
tants hang low over the village resulting in all the residents suffering from
respiratory diseases. Disposing of polluted water from washing the lacquer-
ware is also a problem. As there is no waste water treatment facility, the
polluted water enters the drains directly and flows into the natural waterways
and then onto agricultural land. The waste water pollution from lacquerware
production contains high levels of heavy metals, toxic chemicals, acids, coli-
form bacteria, etc. Consequently, it is no longer safe to drink water from the
well in Ha Thai, there are no fish left in the village pond and the pollutants
are affecting the agricultural sector (Illustration 8.3).
The proliferation of the number of enterprises in Ha Thai producing
lacquerware, and their general lack of awareness of the pollution they are
generating, has contributed to the environmental problems outlined above.
Despite water treatment projects having been considered by the provincial
and local authorities, no system is yet operational. The People’s Committees
of the commune and the province have plans to establish an industrial area
on the outskirts of Ha Thai to which the lacquerware production would be
moved. The plans have been met with a degree of resistance among the
lacquerware producers, however, who are not keen on working in the same
place as they compete over product types and orders. The producers are
interested in keeping their own ideas and contacts to themselves and most
could not afford the extra cost of paying for a workshop outside the home.
The provincial People’s Committee has taken steps to convert the status of
some agricultural land into land intended for industrial production in order
to establish lacquer mini-zones located away from the residential areas. This
is a major advance in the Red River Delta, where paddy fields are sacred and
112 Katherine V. Gough and Dang Nguyen Anh

Illustration 8.3 Wealth generated by lacquerware production has enabled many

households to build new homes in Ha Thai but the ponds and
surrounding agricultural land have been polluted.
Photo by Katherine V. Gough

the status of agricultural land is not easily changed, and should contribute to
reducing environmental pollution.
Wealth generated from the handicraft industry has resulted in social mobil-
ity for some residents. In both Quat Dong and Ha Thai, large-scale pro-
ducers/contractors have emerged and Ha Thai also has many medium-scale
enterprises. The increased wealth is reflected in the purchase of consumer
items and, as Table 8.4 shows, possession of the listed items is much more
widespread in Ha Thai than in Quat Dong. This reflects the greater income
generated from lacquerware: on average 95 million VND yearly per household
in Ha Thai compared with 21 million VND in Quat Dong. However, as well
as differences between the settlements, inequalities are also emerging
within them. Under communist rule, when the work was organised into co-
operatives, there was little income difference between the households. Since
the introduction of Doi Moi policies to promote private enterprise, house-
holds with direct links to middlemen/exporters have been able to establish
successful businesses and become employers themselves. Inequality is more
pronounced in Ha Thai than in Quat Dong, as illustrated by the income
range in the settlements being much greater in the former; in Ha Thai, the
 Handicrafts in Vietnam 113

Table 8.4 Possession of selected household goods in

Quat Dong and Ha Thai

Household good Quat Dong Ha Thai

TV 80 96
DVD/video 40 94
Display cabinet 11 68
Fridge 10 43
Telephone 18 45
Motorbike 30 79
Sample size 100 100

Source: Authors’ household survey, 2006.

highest income recorded was 700 times the lowest income compared with 300
times in Quat Dong. Furthermore, according to official statistics, 9 per cent of
households in Quat Dong are classified as being poor (living below the pov-
erty line) whereas in Ha Thai the figure is 20 per cent. Hence, although in Ha
Thai lacquerware production generates greater wealth, there is also a higher
level of poverty and greater inequality. Anh et al. (2004) documented similar
trends of increasing social polarisation in villages in the Red River Delta
specialised in rattan and bamboo crafts.

Conclusions
Vietnamese economic, cultural and social history is closely related to the
formation and development of traditional craft villages whose products are
made by skilful Vietnamese artisans. The overall picture emerging from the
Red River Delta craft villages is one of handicraft production successfully
adapting over time, but of the impacts on livelihoods varying according
to the nature of the handicraft. The development of the handicrafts has
occurred in a number of phases which go hand in hand with historical
changes in the macro-institutional economic conditions. In the case of
lacquerware, the handicraft has come to dominate livelihoods whereas in
embroidery, the handicraft is carried out side by side with farming. These
differences are due to the differing nature of the production processes, the
markets for the products, the levels of income generated by the handicrafts,
and gender and generational differences in participation in handicraft pro-
duction. Handicrafts, such as lacquerware, which are ‘modernising’ appear to
have a greater chance of surviving though they can also result in increasing
health and environmental problems and greater inequality.
Handicraft production in the Red River Delta has resulted in increased
wealth in the communities, which can be seen physically in the settlements
where many of the original single-storey houses have been replaced by
modern multi-storey houses. Some crafts, such as lacquerware, have come to
require extra labour leading to in-migration which has increased the number
114 Katherine V. Gough and Dang Nguyen Anh
and type of people living in certain craft settlements. As well as physical
characteristics, social aspects of urbanisation are also apparent with the
broadening appeal of urban culture and consumption as illustrated in the
increase in household material possessions. These processes have been
referred to as the development of a ‘new countryside’ which is culturally and
economically more urban than rural (DiGregorio et al., 2003). In the same
vein, it can be argued that there is a frontier of rural urbanisation, identifiable
both physically and socially, which is directly linked to the increased demand
for handicrafts on the global market and encouraged by government policies.
Since the introduction of Doi Moi, craft villages have grown throughout
Vietnam producing a large volume of goods, generating jobs, improving rural
incomes and contributing to reducing poverty in rural areas. However, the
development of the craft villages is still largely due to the spontaneous acts of
individual households who mainly operate on a small scale using basic
technology and often uneducated workers. This can limit the economic
development of craft villages and impact negatively on the environment. In
the coming years, the Vietnamese Communist Party plans to encourage the
development of craft villages in rural areas as they are considered to provide
a key solution to eliminating hunger and reducing poverty. As this study has
shown, the experiences and fortunes of handicraft production can vary
widely. In understanding the place and role of handicrafts in rural develop-
ment it is important to consider the nature of the handicraft as this influences
many factors such as: the type of work involved including the suitability of
combining with agricultural production; the number and type of migrants
entering and/or indigenous leaving the settlements; the level of consolidation
of housing and services in the settlements; and the nature of potential health
risks and environmental pollution. The latter has emerged as requiring urgent
attention by the state, which needs to manage effectively the environmental
pollution that can be caused by certain types of handicraft production. These
differences outlined here need to be taken into account when making policy
decisions regarding handicraft villages as their development needs will vary
according to the craft.

Note
1 In 2004, the exchange rate was 14,600 VND to one US dollar.

References
Anh, D.N., Tacoli, C. and Thanh, H.X. (2004) Stay on the farm, weave in the village,
leave the home: Livelihood diversification and rural–urban linkages in the Red River
Delta and their policy implications, Hanoi: The Gioi Publishers.
Anh, D.N. (2005) Internal migration: Opportunities and challenges for the renovations
and development in Vietnam, VAPEC, Hanoi: The Gioi Publishers.
DiGregorio, M., Leisz, S.J., and Vogler, J. (2003) ‘The invisible urban transition: Rural
 Handicrafts in Vietnam 115
urbanisation in the Red River Delta’, paper presented at the 7th International
Congress of Asian Planning Schools Association, Hanoi, 12–14 September.
Dixon, C. (2003) ‘Developmental lessons of the Vietnamese transitional economy’,
Progress in Development Studies, 3(4): 287–306.
Gough, K.V. and Kellett, P. (2001) ‘Housing consolidation and home-based income
generation: Evidence from self-help settlements in two Colombian cities’, Cities,
18(4): 235–47.
Ministry of Agricultural and Rural Development (MARD) (2005) ‘Program “one
village, one product” ’ (draft version), Hanoi: MARD.
Smith, D.W. and Scarpaci, J.L. (2000) ‘Urbanization in transitional societies: An
overview of Vietnam and Hanoi’, Urban Geography, 21(8): 745–57.
Yen, T.M. (2005) Traditional craft villages in the process of industrialization and
modernization, Hanoi: Social Sciences Publishing House.
9 A revolution in the Thai
handicraft industry?
Handicrafts, integration and
rural development in
Northern Thailand
Jonathan Rigg, Suriya Veeravongs,
Piyawadee Rohitarachoon and
Lalida Veeravongs

Introduction
From the moment that Thailand was accorded the epithet ‘miracle economy’
by the World Bank in the early 1990s (World Bank, 1993), attention has
focused on the Kingdom’s fast-track, foreign investment-driven, export-
oriented industrialisation. To be sure, structural change has been both rapid
and deep and this, in turn, has led to a fundamental restructuring of patterns
of employment and modes of living. But there are two issues that can be
placed in juxtaposition to this image of Thailand’s development practice and
experience. First of all, there remains an important rural and agricultural
context that has supported and contributed to the industrialisation process,
while also being affected by it. And second, while car parts, micro-electronics,
garments and footwear may represent the ‘new’, high-profile product lines
in Thailand’s industrialisation (see Chapter 10), there is an older industrial
sector of artisan-based activities which continues to be active, significant and
vital. It is this second, low-profile and low-impact element of the country’s
industrial sector which concerns us here.
Handicraft or artisan-based activities have a long history in Thailand. In a
real sense, every household was an artisan household. Every woman was
expected to be able to weave; most farmers would make or mend their own
farm implements; and houses and their decorative flourishes were produced
by villagers. While there were individuals with particular skills and specialisa-
tions, at a general level villagers were artisans just as they were farmers. As
the authors of Village Chiang Mai, one of the first detailed village surveys to
be carried out in the north of Thailand in 1969–1970, make clear, during the
dry season many villagers turned their hands and their time to artisanal work.
Men would repair houses, animal pens and farming tools; women would
weave; while both men and women would make household utensils such as
bamboo rice baskets (Na Ayuthaya et al., 1979). The contemporary Thai
 A revolution in the Thai handicraft industry? 117
rural handicraft sector, therefore, emerges out of a long history of handicraft
production. It is also important to note that certain parts of the sector were
monetised and commercialised at an early date, a point that Bowie (1992)
makes in her study of nineteenth-century textile production in northern
Thailand. This is relevant in so far as how we view the past shapes our view of
the historical context out of which the contemporary industry has emerged
and developed.
The view that the contemporary rural handicraft sector in Thailand is
explicitly linked – spatially, technologically and economically – to a pre-
modern ‘industry’ is, however, problematic due to the assumption that such
activities, because of their presumed and apparent associations with the
past, are necessarily community-focused, founded on traditional techno-
logies and skills, socially inclusive, environmentally benign, sustainable and
low-risk, ‘authentic’, and participatory in terms of how they are organised.
In this chapter, we argue that while the activities we analyse may be based
in rural contexts and although they may appear to draw on the past for
their artistic inspiration, they are more ‘new’ than ‘old’, more ‘modern’
than ‘traditional’, and more ‘global’ than ‘local’. These terms are, of course,
simplifications and suffer from the inadequacies and inaccuracies of all
binaries applied across varied social contexts and geographical spaces. None-
theless, they highlight the axes of concern that lie at the heart of the discus-
sion that follows. All that said, it should be emphasised that while the
contemporary handicraft industry may not be quite what it seems, or what it
presents itself to be, we do not start from the premise that this is necessarily
‘wrong’ or ‘bad’; indeed, without the adaptations that we outline and explore
in this chapter, it is arguable that artisan-based production would have
disappeared almost entirely from rural communities. As Parnwell writes
in the context of his study of ‘eco-localism’ in northeast Thailand, ‘a genu-
inely local localism is increasingly unlikely (and not altogether desirable)
within a modern reality of intense national, regional and global integration’
(2006: 193).
Not only is the notion of continuity between the present and the past
problematic on historical grounds, but this is of practical and policy import-
ance because of the perceived role that such ‘traditional’ activities can play
in rural development. Rural poverty-alleviation programmes in Thailand
are partially based on re-energising traditional craft-based skills and activ-
ities (ALRO-MOAC, 2006). Viewed particularly in the light of the Asian
economic crisis and in the context of Thailand’s continuing flirtation with
‘localism’ (Hewison, 1999, 2001; Parnwell, 2005, 2006), rural industries are
seen as a key means of ensuring that development is pro-poor, biased in
favour of rural areas, community-based, participatory, and sustainable.1
Rural industrialisation is often held in favourable contradistinction to state
industrialisation policies to date which are seen to have created an economy
that is vulnerable, dependent, environmentally destructive and unequal.
118 J. Rigg, S. Veeravongs, P. Rohitarachoon, L. Veeravongs
Research in context: Ban Mor and Ban Pasakluang
The fieldwork on which this chapter is based was carried out between late
2005 and mid 2006 in two villages in Chiang Mai province in Northern
Thailand: Ban Pasakluang in Doi Saket District, and Ban Mor in Sankam-
phaeng District (Figure 9.1). The villages were selected because they repre-
sent different handicraft development ‘trajectories’. Ban Mor specialises in a
‘new’ handicraft – mango wood crafts – while Ban Pasakluang reveals the
revitalisation of an old craft – saa (‘mulberry’) paper.
Until the 1970s, Ban Mor and Ban Pasakluang were farming-focused
settlements. The production of wet rice for household consumption was the
central and guiding activity of their populations, and the focus of their atten-
tion, efforts, knowledge and expertise. Government policy from the 1980s was
to encourage farmers to grow high-value cash crops and invest in new tech-
nologies and methods of cultivation. Local people observed, however, that
just as they embraced these new capital-intensive systems, so both production
and returns to farming were becoming more unstable. In particular, the price
of rice was falling while the costs of chemical fertilisers, pesticides and

Figure 9.1 Map showing location of study settlements in Northern Thailand.

 A revolution in the Thai handicraft industry? 119
herbicides – all encouraged by the government – were increasing (see UNDP,
2007). The effects of this economic squeeze, moreover, were compounded by
growing environmental challenges that were limiting farmers’ ability to inten-
sify production. While the rain-fed rice farmers of northern Thailand have
always had to juggle risk in the context of a capricious climate, population
growth and the expansion of cultivation onto more marginal lands has
accentuated these risks.
The 1980s and 1990s also saw considerable change in agricultural tech-
nologies and farming methods. Tractors were introduced to mechanise land
preparation and, more recently, harvesting machines have been introduced.
Traditional forms of labour exchange were replaced by wage labouring as
village economies became increasingly monetised and, with growing mechan-
isation, wage labouring itself has been squeezed. During the 1990s, the trends
of the 1980s continued and deepened. More villagers found employment off-
farm in the service and public sectors. With weak farm-gate prices, rising
costs of inputs, and the difficulty of finding labour to work the land, many
farmers decided that it was no longer worthwhile investing either cash or time
in rice farming. Rice cultivation became focused on production for family
consumption. Some households even gave up their role as own-account
producers, and rented out their fields. Others sold their rice land as a final
indication of their lost commitment to farming.
Even in the heyday of farming, Ban Mor’s population did more than just
farm. In particular, during the dry season, women would make earthenware
pots or mor, after which the village is named. Men were not involved in the
production of pots; their role was connected with the selling of the products
in neighbouring villages and districts, often bartering pots for paddy, salt and
chillies. But just as agriculture was in decline for the reasons noted above, the
other ‘leg’ of Ban Mor’s traditional economy was also being compromised as
demand for earthenware pots dwindled in the face of the wide availability of
alternative aluminium and plastic substitutes. These were often cheaper, usu-
ally more long-lasting, and were also seen in local terms as emblematic of
modernity.
To combat this contraction in traditional occupations, households in
Ban Mor began to experiment with new activities, and particularly with
household-based non-farm activities which drew on villagers’ expertise in
small-scale industrial enterprise. In 1973, villagers began to use local raw
materials to produce a range of craft products including various bamboo-
based articles. By 1977 this had become a significant alternative income-
generating and livelihood-sustaining element in the village economy. Later, a
range of crafts based on the lathing of teak was developed. Souvenirs such as
miniature traditional farming tools, water dippers and toothpick holders
were developed for sale. These were followed by models of birds made from
apple wood (mai kra torn) and wooden dolls carved from the wood of the
kapok tree (mai noon). More recently, a range of decorative articles lathed
from mango wood have come to dominate village handicrafts, from vases to
120 J. Rigg, S. Veeravongs, P. Rohitarachoon, L. Veeravongs
table lamps. This process of diversification within the village handicraft
sector was given a fillip by the promotion of the OTOP – One Tambon, One
Product – policy by the Thai government from 1999. Particularly significant
in the development of Ban Mor’s wood-based handicraft industry was the
shift from teak to mango wood (see next section).
At the time of the survey in 2005, fewer than 5 per cent of households stated
that farming was their main source of income, although a further 22 per cent
reported farming as a secondary or tertiary activity. In total, then, over one
quarter of households retained a significant or at least a vestigial hand in
agriculture, and juggled farming and nonfarming activities.2 It is also true
that Ban Mor remains embedded in an agricultural landscape (Illustration
9.1). In 2005, 1484 rai (237 ha) of land was cultivated and of this, 1202 rai
(192 ha) was used for rice farming. Even so, the direction of livelihood
change in Ban Mor over the last two to three decades is clear (Figure 9.2).
In 1995 – based on villagers’ recall – 16 per cent of households recorded
agriculture (including livestock-raising) as their main economic activity, and
31 per cent some form of handicraft activity; ten years later, in 2005, the
respective figures were 5 per cent and 67 per cent.
Ban Pasakluang, like Ban Mor, is a long-established settlement with an
agricultural past (Figure 9.1). Also like Ban Mor, home-based manufacturing
has become increasingly important in the village economy and household

Illustration 9.1 Surrounding agricultural landscape of Ban Mor.

Photo by Jonathan Rigg
 A revolution in the Thai handicraft industry? 121

Figure 9.2 Sum of primary, secondary and tertiary activities, by selected occupation,
Ban Mor, 1995 and 2005.
Source: Authors’ household survey, 2005.
Note: Households usually engage in more than one activity or occupation.

livelihoods. The diversification and delocalisation of the economy began

rather earlier in Ban Pasakluang than it did in Ban Mor, during the 1960s,
in response to the evolution of the wider regional economy of Northern
Thailand. Wood carving emerged as a supplementary activity in the 1970s
while pillow embroidery was also undertaken, mainly by women and chil-
dren. Not only was work coming to the village, but some young women began
to leave Ban Pasakluang to find jobs in the service sector (in hotels) and in
manufacturing (in textile factories). From 1997, another phase in the evolu-
tion of Ban Pasakluang’s nonfarm economy took hold: the expansion of saa
paper production.
The evolving balance of farm and nonfarm occupations among sampled
households in Ban Pasakluang was not dissimilar to Ban Mor: in 2005, farm-
ing was a primary, secondary or tertiary activity among 30 per cent of the
households interviewed, while 51 per cent were engaged in some form of craft-
making activity. Ten years earlier, in 1995, the respective figures were
44 per cent and 39 per cent (Figure 9.3). Like Ban Mor, few households – just
6 per cent – relied on farming in 2005 as their main source of income. Today
some 70 households work in the handicraft industry and, on average, each
122 J. Rigg, S. Veeravongs, P. Rohitarachoon, L. Veeravongs

Figure 9.3 Sum of primary, secondary and tertiary activities, by selected occupation,
Ban Pasakluang, 1995 and 2005.
Source: Authors’ household survey, 2005.
Note: Households usually engage in more than one activity or occupation.

Table 9.1 Summary statistics: Ban Mor and Ban Pasakluang

Ban Mor Ban Pasakluang

Population (2005) 1083 690

Households (2005) 385 210
Households sampled, 2005 130 88
Agricultural landholdings (ha) 237 131
Rice land (ha) 192 105
Land holding per household (ha) 0.62 0.62
Riceland holding per household (ha) 0.50 0.50
Proportion of households engaged in handicrafts 49 38
(% total households, 2003)
Number of productive activities per household (2003) 3.7 1.7

Sources: Ban Mor Kor Chor Chor 2 Khor 2003: 2–4; Ban Pasakluang Kor Chor Chor 2 Khor
2001 and 2003; Tambon Health Station and District Census Registration Bureau statistics.

household farms half a hectare of rice land – a sub-livelihood holding

(Table 9.1).

Mapping and tracking craft transitions in northern Thailand:

from craft to industry
Notwithstanding important differences between Ban Mor and Ban Pasak-
luang, both – on first impressions at least – fit with the current fashion to laud
and promote village-based development initiatives that build on local skills,
expertise and knowledge and which stress the small scale over the large scale,
 A revolution in the Thai handicraft industry? 123
the traditional over the modern, and ‘appropriate’ technology over high
technology.

Ban Mor and the mango wood industry: creating a handicraft

Ban Mor has a long history of specialised craft-making, but today the
most celebrated of Ban Mor’s products are lathed mango wood products.
Some villagers involved in production can be regarded as household-based
artisans, but around 20 households run large workshops or ‘factories’
(Illustration 9.2). While mango wood craft-making draws on a long artisan
tradition in Ban Mor, the technology of production is modern. Powered
lathing machines have replaced hand operated lathes, modern kilns have
supplanted other means of drying, chemical dyes and paints are used in place
of natural methods of staining the products, and hand painting has been
superseded by spraying.
The use of mango wood dates from the late 1980s when the introduction of
a logging ban in Thailand caused the supply of teak wood to dry up. Mango
wood presented an attractive alternative; there was an abundance of mango
trees, the wood carves and lathes well and, at the same time, it is not greatly
valued for other uses – the wood burns poorly and produces low-quality
charcoal. The introduction of powered lathes made it possible to produce

Illustration 9.2 Sanding mango wood in a workshop in Ban Mor.

Photo by Jonathan Rigg
124 J. Rigg, S. Veeravongs, P. Rohitarachoon, L. Veeravongs
larger articles and meant that workers could acquire the necessary skills more
quickly and easily, leading to a change in the make-up of the workforce. In
addition, with increasing demand for mango wood, local artisans had to
range further afield to source the wood. From a position in the early 1990s
when it was locally abundant and could be acquired virtually for free, by 2005
producers were having to source wood from provinces in the Northeastern
region, from Nakorn Si Thammarat in the South, from Uttaradit and
Sukhothai in the Lower North, and from Uthai Thani in the upper Central
Plains.
There is a market hierarchy which varies according to product but which is
not dissimilar to that for saa paper. Sub-contractors receive their orders from
entrepreneurs who, in turn, obtain their orders directly from customers or
from dealers. At the same time, these sub-contractors distribute the orders to
their groups or work teams or to their individual employees. The products are
transferred from one worker to the next as each stage in the production
process is completed. Sometimes this involves the transfer of the unfinished
product between independent workers; in other instances it is contained
within a single factory or production unit. They always end, however,
with the highest-level entrepreneur who delivers the finished product to the
customer or dealer.
Mango wood-craft production, unlike saa paper, shows stronger segmenta-
tion by task, gender and generation. Men tend to peel the bark from the
mango logs and work the powered lathes. Scrubbing by machine and the
manual smoothing-off of the wood, however, is mostly done by women
because, it was said, this requires care, attention to detail and a delicate touch.
Age also plays a role. Older men, for example, only occasionally work the
lathes or remove the bark from the raw mango logs. Villagers involved in
mango wood-craft production usually work full time and just a few workers
juggle farm work with mango wood work.
The mango wood handicraft industry in Ban Mor relies, to a significant
extent, on migrant workers. Some commute from neighbouring settlements,
but there are also migrants from other Northern provinces and from Burma.
In total there are between 600 and 1200 workers employed in the mango
wood industry in Ban Mor, dwarfing the number who work on the land.
Dormitory accommodation for the long-distance domestic and international
migrants is provided by the mango wood entrepreneurs. The expansion of the
mango wood industry has played a central role in re-working the social fabric
of Ban Mor through creating the conditions whereby migrant labourers are
attracted to work and live in the village.3 Entrepreneurs, intent on expanding
their businesses but cognisant of the limited local pool of labour, build dor-
mitories and other social infrastructure to support their business plans but,
in so doing, alter the village as a social entity as well as an economic unit.
Because Ban Mor has such a vibrant economy, there is little out-migration.
There are cases where villagers commute to work in the tambon, the local
district or elsewhere in the province. Students may also choose to continue
 A revolution in the Thai handicraft industry? 125
their studies in Chiang Mai city or Bangkok. Generally, however, Ban
Mor is a village of attraction and in-migration rather than a village of out-
migration.

Saa paper and saa paper products: revitalising a traditional industry

Saa paper has been a traditional craft of Sankamphaeng district for centur-
ies. It was used in candle-making and tailoring (as pattern paper), to make
Northern-style pennants and lanterns, to cover bai laan Buddhist scriptures
and, most famously, in the production of saa paper umbrellas. The expansion
and commercialisation of saa paper production began in the 1960s, accelerat-
ing and deepening in the 1970s, and involved a three-fold shift in the basis and
operation of the ‘industry’. To begin with, the product itself began to be
developed and diversified in new directions, from simple sheets of saa paper
to more elaborate papers and saa paper products. Second, demand for saa
paper in established, traditional markets began to decline. But just as trad-
itional uses of saa paper were being squeezed out, so new possibilities were
appearing. This third shift really gained momentum from 1975. In 1981 saa
paper became an export product when the first shipment of paper was sent
overseas, to Japan. As demand increased so the perimeter of production
spilled out from Ban Tonpao – the traditional heart of the industry – to other
villages and districts, including Ban Pasakluang, the study site.
The raw material needed for the production of saa paper is the fibre
stripped from the inner bark of por saa, the paper mulberry tree (Broussonetia
papyrifera). Por saa flourished wild in the provinces of Chiang Mai and Chiang
Rai, growing along the banks of rivers and streams and, until the 1970s, was
abundant and collected as an open-access resource. As saa paper production
expanded, however, so local supplies of por saa became increasingly scarce
and the plant had to be sourced from other areas of Thailand and, more
recently, from neighbouring Laos and Burma. Por saa today is grown on
privately owned plantations, some in Laos and Burma, harvested by wage
labourers, and sold in bulk to large users. Not only has the cultivation of the
key raw material been outsourced (and commoditised), but there has also
occurred an out-sourcing and de-centring of some aspects of production. In
particular, the basic sheets of saa paper are now often made in districts and
provinces beyond Sankamphaeng and Chiang Mai.4 Local entrepreneurs
concentrate their energies on transforming the paper into decorative products
with greater added value, signing contracts with saa paper makers elsewhere
to supply good-quality saa paper to their workshops.
As demand for saa paper increased, so various innovations were introduced
to speed up the production process. In 1987, the Department of Industrial
Promotion introduced a powered churning machine to do away with time-
consuming manual pulverisation. In 1992, overseas buyers in Japan and the
USA recommended that chemical dyes be used to standardise products, and
advisors came to train producer families in their use. Later, the use of sodium
126 J. Rigg, S. Veeravongs, P. Rohitarachoon, L. Veeravongs
hydroxide in the boiling stage was promoted, reducing the time to process and
prepare the inner bark from six hours to just one.
While such modern technologies and techniques may have increased
productivity and standardised quality, there has been a recent resurgence in
interest in ‘traditional’ methods of production, largely driven by demand
from some quarters for items produced using traditional methods. As a
result, some producers have divided their production processes into a ‘nat-
ural’ or ‘traditional’ approach, and a ‘chemical’ or ‘modern’ approach, to
serve the demands of different markets. Manual pounding of the fibre makes
for a rougher, more authentic product, while the churning machine produces
a finer fibre and end product. Chemical dyes create standardised, vibrant
colours for customers; natural dyes are not controlled for consistency, and
the resulting product tends to be less vibrant but more ‘natural’.
Broadly, there are three categories of producer. The first consists of large-
scale ‘factories’ where wage labourers are employed as workers (Illustration
9.3). These factories usually have product showrooms attached and are suf-
ficiently large to manage their own marketing networks. The second type of
operation consists of medium-scale producers who often have teams of free-
lance sub-contractors working for them, often based in the same village. Raw
materials are provided to these sub-contractors who then produce an agreed
product, for a set rate, at a pre-arranged date. These medium-scale producers

Illustration 9.3 Saa paper ‘factory’.

Photo by Jonathan Rigg
 A revolution in the Thai handicraft industry? 127
do not have showrooms and are not in a position to market their products; in
the main, they sell to the larger-scale factories. The last group are the piece-
work producers. They have no contact with customers but work for and
through the medium- and larger-scale entrepreneurs.
Producers build up skills and expertise in particular areas, ranging from
sheets of decorative paper to various saa paper products, such as envelopes,
boxes or notebooks. These producers are small-scale and organised at the
individual or household levels; there is usually no division of labour. Generally,
orders are received by the large-scale factories or entrepreneurs, both from
abroad and from Thai-based buyers. Medium-scale sub-contractors take
orders from the large-scale factories and then return to their village or
sub-district to allocate or sub-contract this order to smaller groups or to
individual households. The same network of links operates in the reverse
direction, with the finished product being passed back up the production
chain and with, in some cases, further stages in the production process being
undertaken en route. Production networks can involve 300–400 households
scattered across several districts. There is no gender division of labour in the
main production activities and in some cases entire families are employed.

Reflecting on transformations in the craft industry:

rural development and rural development paradigms
Like many developing countries, there has been considerable debate in
Thailand about the benefits of ‘appropriate’ development, most obviously
reflected in discussions about the ‘sufficiency economy’. This has been closely
associated with the King of Thailand and was included as the guiding phil-
osophy for the 9th Economic and Social Development Plan (2002–2006)
(NESDB, n.d.) and also provided the inspiration for the 2007 Thailand
Human Development Report (UNDP, 2007).
In the popular imagination, the Thai handicraft industry would seem to
tick many of the boxes of the sufficiency economy. It is spatially rooted in
rural villages. It draws on traditional skills that link with Thailand’s craft
traditions. Raw materials are locally sourced. The units of production are
generally small and based on the household. There is a high degree of local
ownership and control. And patterns of employment are flexible, permitting
inter-locking livelihoods to emerge that productively link farm and non-farm
activities. In this way, handicrafts can be regarded as representing a more
locally rooted, relevant and appropriate form of development, and one that
can be favourably contrasted with large-scale, fast-track, foreign-invested,
export-oriented industrialisation. Even a brief reflection on the descriptive
account above would lead one to question the veracity of such an inter-
pretation of craft-making in Ban Mor and Ban Pasakluang.
The first area of debate relates to the spatialities of production: the
‘where’ of craft-making. At first glance, it does seem that production remains
rooted in the village context. But this operates in a networked manner where
128 J. Rigg, S. Veeravongs, P. Rohitarachoon, L. Veeravongs
different stages in the production process are disaggregated between villages
and over space. In the case of saa paper, the raw materials are often sourced
from abroad and the paper itself is produced in other provinces in the north
of Thailand. Mango wood is no longer readily available locally and is trucked
in from distant provinces. The workforce in both villages, but particularly
in Ban Mor, is largely non-local, while particular stages in the production
process are divided between households within communities, and between
villages and districts in the locale. While authenticity relies on a claim that
products are based on traditional skills and technologies, the two crafts
described here increasingly draw on and use non-traditional and non-local
knowledge, equipment and skills.5 Equally, the products themselves, even in
the case of a traditional craft like saa paper, have metamorphosed and
developed in line with patterns of demand and inspiration that lie beyond the
local context. In both research sites, we discern a progressive separation
in terms of knowledge, space and activity.
The handicraft industry in Ban Mor and Ban Pasakluang, therefore,
reveals a mixed bag of spatial processes. We see a marked degree of concen-
tration as products are made in particular villages or districts, so much so that
these have acquired a national and even international reputation. At the
same time, however, within the villages and districts production has become
increasingly spatially disaggregated, with households and individuals spe-
cialising in certain stages in the production cycle. This spatial disaggrega-
tion is facilitated by the social networks that link households and villages,
which provide a supporting base of social capital on which economic com-
plementarities between settlements and households can be developed and
sustained. That said, a de-skilling through mechanisation of some stages in
the process (for example, mango wood lathing) has permitted non-specialist
outsiders to be drawn into the social network. This spatial outsourcing is even
more dramatic in terms of the raw materials of production. The sustain-
ability of the localised handicraft industry in Chiang Mai is, therefore,
achieved through strategic non-local links that tie village-based producers
into non-local networks. Most dramatic of all, of course, is the geographical
dispersal of sales. In the past sales were locally concentrated; today the
vibrancy of the industry depends on its ability to entice foreign tourists and
enter international markets. Paper produced in Ban Pasakluang is exported,
for example, to Japan, China, the USA, Canada, Spain, France and Saudi
Arabia while the whole area is marketed as a tourist destination to which
international visitors are directed and channelled.6
The spatial restructuring of work in Ban Mor and Ban Pasakluang has
also restructured the basis of inequality in the villages. Traditionally, the poor
were the landless and land-poor, those unable either to meet their subsistence
needs or generate a reasonable surplus from own-account farming. The
changes set in motion by the expansion of the handicraft industry have cre-
ated two new – and overlapping – classes of ‘poor’: the village industry wage
labourer, and the migrant worker. Something akin to the old divide between
 A revolution in the Thai handicraft industry? 129
the landed and the landless has come to characterise the operation of the
craft industry, with a poverty/prosperity separation between those families
who have become handicraft entrepreneurs, and those who provide the
labour for these businesses. The family members of the entrepreneurial class
no longer engage with production per se but concentrate their time in mar-
keting, design, accounts and management, and acquire the skills through
education to pursue these occupations. Wage labourers and pieceworkers,
some of them migrants, fill the production void, much as wage labourers used
to work on the fields of the land-rich. The key divide in the villages, therefore,
is no longer between the landed and the landless, but between employers
and workers.
It is clear that in various ways, craft production in Ban Mor and Ban
Pasakluang is being stretched out of its formerly rural location. But it none-
theless occurs in particular rural spaces, and it not only shapes those spaces
but is, in turn, shaped by them. Table 9.2 reveals one aspect of this with
regard to occupation, by age category, in the two villages. To begin with, it is
striking how farming has become the preserve of older villagers. Of the 92
individuals whose primary activity was identified as farming, 67 (73 per cent)
were aged 51 years or older. On the basis of this, we can say that a process of
geriatrification of farming is well advanced. At the same time, handicraft-
based employment was primarily the preserve of the middle cohort, aged
between 31 and 50 years old, with some 57 per cent of handicraft workers
falling into this category. Finally, a significant proportion (49 per cent) of
those aged 17–30 years old were students. We suspect, based on our inter-
views, that comparatively few of these younger individuals will take up
handicraft work and even fewer will consider farming as an occupation. In
essence, what we see in Table 9.2 is an insight, at a particular point in

Table 9.2 Age and occupational status, by proportion of respondents in each age
category, 2005

Ban Pasakluang Ban Mor

17–30 31–50 51+ 17–30 31–50 51+

years years years years years years

Farming 0 15 40 0 10 27
Employee 11 27 4 23 23 11
Casual labourer 0 18 14 2 23 19
Student 36 0 0 25 0 0
Handicraft work 3 55 20 24 87 62
Trading and shop – – – 1 5 17
keeping
Total 50 115 78 75 148 136
Sample size 243 359

Source: Authors’ household survey, 2005.

130 J. Rigg, S. Veeravongs, P. Rohitarachoon, L. Veeravongs
time, into the livelihood transformations under way in Ban Mor and Ban
Pasakluang. We see the tail end of smallholder ‘peasant’ farming, concen-
trated among a handful of ageing villagers. We identify handicrafts as a
significant occupation for those in their middle years who have not benefited
from Thailand’s educational revolution but who have chosen to pursue a
non-farm activity, probably for aspirational as well as pecuniary reasons. And
third, we also discern, in the degree to which younger women and men are
continuing in education after secondary level, a likely future shift out of
handicrafts, at least among ‘core’ village households.7 This shift will further
de-localise livelihoods as villagers engage with work in other spatial arenas
and sectoral contexts.8 Of course handicraft production may persist, but
village labour will increasingly be supplanted – as is already happening – with
non-local (and perhaps non-Thai) labour, while most core villagers who
remain in the industry will increasingly take on managerial and marketing
roles. The restructuring of the poverty/prosperity nexus discussed above will,
we suspect, thereby become further accentuated as the separation between
the ‘village’ as a community of people and the village as a site of production,
widens. What Table 9.2 does not reveal is the degree of demographic turbu-
lence in the villages. It does not pick up the many migrant workers who live in
the villages, but are not part of the village community (a point also explored
in the next chapter); not does it fully reflect the departures from the settle-
ments as young people move elsewhere and establish their own households,
separate from their natal villages.

Conclusion: continual revolution in a craft industry?

Taken together, the changes sketched out above mean that the bases of eco-
nomic and livelihood sustainability in Ban Mor and Ban Pasakluang have
altered. The reliance on the wider national and international marketplace, on
non-local raw materials, on higher levels of capitalisation, on migrant labour,
on non-traditional skills and technologies, and on marketing and production
systems that link villages into wider networks, have altered the vulnerabilities
and dependencies that characterise work and life. The usual gloss given to
handicrafts tends to underplay the degree to which such activities are under-
going continual revolution. The need to sustain supplies of raw materials for
production (mango wood and por saa); the difficulty of maintaining a stable
workforce; the requirement to invest in new technologies; the ease with which
competitors can copy designs; the fickleness of the international market;
and the more fundamental risk that some families might simply opt out as
generational change and livelihood preferences squeeze handicrafts out of
the equation – all these are at play. We do not, therefore, see in Ban Mor and
Ban Pasakluang two villages partially shielded from the perturbations and
demands of engagement with global production by their reliance on local raw
materials, household labour, traditional knowledge and technologies, further
insulated by the cushioning effect of semi-subsistence farming. Indeed, we
 A revolution in the Thai handicraft industry? 131
are left with the question as to what, in essence, comprises the difference
between, say, the foreign-invested factories of the industrial estates of nearby
Lamphun around 25 km away – to which many young people in the region
are drawn – and the village workshops of Ban Mor and Ban Pasakluang.
Certainly there are important distinctions that we can draw: in terms of
location, size and capitalisation, levels of technology that are brought to bear,
and ownership. But nonetheless there has been a narrowing of these differ-
ences so that village-based craft activities such as those described here are,
perhaps, better viewed as a form of manufacturing and part of the main-
stream industrial sector. The folksy image and their official presentation (in
terms of how such activities are recorded in the official statistics) as somehow
outside the mainstream, serve to obscure a ‘revolution’ in the handicraft
industry.
All this challenges – but does not necessarily irredeemably undermine – one
of the abiding characteristics of handicrafts: the place-specific and place-rich
character of production. We suggest that commitment to place has been
eroded and is on the verge of being compromised by a combination of the
changing nature of handicraft work in the study villages, greater ease of
mobility, and the changing priorities of villagers. The nature of the work also
means that it is increasingly difficult to dovetail craft-work with agricul-
tural work and, therefore, one of the key attractions of such work in the
wider discourse of rural development is undermined, namely that farm
and nonfarm should be integrated and combined in productive and comple-
mentary ways.9
The tendency to place craft production in an industrial and development
category separate from the mainstream, linked often to a localist agenda and
community development paradigm, plays down the development dynamic
involved. What may begin as ‘local’ becomes increasingly non-local; what
starts as traditional becomes modern; and what, at the outset, may be
described as simple becomes more stratified, complex and hierarchical. This
dynamic tends to take such local initiatives away from their founding char-
acteristics. Moreover, and importantly, they are an outcome of success and
vitality, rather than failure and stagnation. And yet, from the purist point of
view this dynamic undermines many of the key attractions of such craft-based,
village-centred, rural development.

Notes
1 The ‘localism’ debate and agenda emerged during the 1990s as Thailand pro-
gressively integrated into the international economy, and was given a fillip
during Thailand’s economic crisis of 1997–9. The ‘localists’ are a broad and
unstructured association of NGOs, new social movements and intellectuals who
have an agenda – quasi-nationalist in tenor – that emphasises self-reliance and
self-sufficiency, local knowledge and community development and which cri-
tiques the country’s integration into the world economy, and globalisation more
widely.
132 J. Rigg, S. Veeravongs, P. Rohitarachoon, L. Veeravongs
2 It is striking the degree to which households in Ban Mor embrace multiple
activities to meet their needs, with an average of 3.7 income-generating activities
per household (see Table 9.1).
3 It is hard to estimate accurately the number of migrant workers in Ban Mor.
According to the former president of the TAO Council, approximately one hun-
dred migrants were living and working in Ban Mor at the time of the survey in
2005. In addition, however, there were workers who based themselves in other,
surrounding villages but came to Ban Mor daily for employment.
4 The provinces of Phrae and Naan are now both important producers of saa paper,
and paper is also supplied to producers in the study sites from Lampang, Mae
Hong Son, Sukhothai and Chiang Rai.
5 See Wherry (2006) for a paper on authenticity in Northern Thai crafts.
6 Tours are operated to Ban Tonpao and in 2005 the provincial government
announced the promotion of a ‘craft village tourist network’, with the labelling of
some villages in the network as ‘model craft communities’.
7 By ‘core’ village households we mean those individuals and households who are
not migrants to the village.
8 By ‘delocalisation’ we mean the spatial disembedding of livelihood activities from
the spatial context of the village and its surrounding lands.
9 This broader debate is explored in Chapter 11.

References
ALRO-MOAC (2006) ‘National report on agrarian reform and rural development
in Thailand’, prepared by the Agricultural Land Reform Office (ALRO) and Min-
istry of Agriculture and Cooperative (MOAC), Bangkok, Thailand and submitted
to the International Conference on Agrarian Reform and Rural Development
(ICARRD), March 6–10, Brazil. Download from: http://www.icarrd.org/en/
icard_doc_down/national_Thailand.pdf.
Bowie, K.A. (1992) ‘Unraveling the myth of the subsistence economy: Textile pro-
duction in nineteenth century Northern Thailand’, Journal of Asian Studies, 51(4):
797–823.
Hewison, K. (1999) ‘Localism in Thailand: A study of globalisation and its dis-
contents’, CSGR working paper no. 39/99, Centre for the Study of Globalisation
and Regionalisation, University of Warwick. Download from: http://www
2.warwick.ac.uk/fac/soc/csgr/research/workingpapers/1999/wp3999.pdf.
Hewison, K. (2001) ‘Nationalism, populism, dependency: Southeast Asia and res-
ponses to the Asian crisis’, Singapore Journal of Tropical Geography, 22(3): 219–36.
Na Ayuthaya, Abha Sirivongs, F.W. Fuhs and Suthep Soonthornpasuch (1979)
Village Chiang Mai: Social and economic conditions of a rural population in
Northern Thailand, Bangkok: Social Research Institute, Chulalongkorn University.
NESDB (n.d.) The Ninth National Economic and Social Development Plan (2002–
2006), Bangkok: National Economic and Social Development Board.
Parnwell, M.J.G. (2005) ‘The power to change: Rebuilding sustainable livelihoods in
North-East Thailand’, The Journal of Transdisciplinary Environmental Studies,
4(2): 1–21.
Parnwell, M.J.G (2006) ‘Eco-localism and the shaping of sustainable social and nat-
ural environments in North-East Thailand’, Land Degradation and Development,
17: 183–95.
UNDP (2007) Thailand Human Development Report 2007: Sufficiency Economy and
 A revolution in the Thai handicraft industry? 133
Human Development, Bangkok: UNDP. Download from: http://www.undp.or.th/
publications/documents/NHDRBookEnghi_final.pdf.
Wherry, F.F. (2006) ‘The social sources of authenticity in global handicraft markets:
Evidence from Northern Thailand’, Journal of Consumer Culture, 6(1): 5–32.
World Bank (1993) The East Asian Miracle: Economic growth and public policy,
Oxford: Oxford University Press.
10 From farm to factory
Village change in a
rice-growing region
Jonathan Rigg, Suriya Veeravongs,
Piyawadee Rohitarachoon and
Lalida Veeravongs

Introduction
In Asia, there has been a partial separation of factories from their assumed
and traditional urban and peri-urban locations. In Indonesia, Malaysia,
Vietnam, the Philippines and – in this instance – Thailand, there are large,
export-oriented, often foreign-invested factories in areas that have tradition-
ally been regarded as resolutely ‘rural’. The reasons for this are not hard to
discern: primarily, cheap land and the availability of low-cost labour. With
improving communications and government policies that have promoted the
dispersal of industry to poorer (rural) regions, some of the impediments and
disincentives for rural industrialisation have also been removed or amelior-
ated. The result, in Thailand, is that the industrial investment frontier has
moved outwards from the metropolitan core (Bangkok), to the wider Central
Plains region and Eastern Seaboard, in some instances leap-frogging to even
more distant areas in the Northern (Lamphun), Northeastern (Korat) and
Southern (Songkhla) regions.1
The novelty of the factory frontier extends beyond the driving forces behind
the process – the investment decisions of commercial, often foreign firms,
supported by national planning agencies. It is novel for two other reasons.
First, because it is not physical space being colonised by people, but human
(labour) and physical (land) space being colonised by capital. Second, the
frontier is both a space of habitation by a core population (i.e. village residents)
and, as we will show, a space of attraction and residence by a population of
migrant sojourners from other areas of the country. Because of this, we have
a series of parallel, but only partially intersecting transformations under way:
spatial changes linked to the interplay of farming with non-farming activities
and processes; agricultural changes associated with the economic, social
(aspirational) and environmental trade-offs between farming and other activ-
ities; livelihood changes that arise from the widening opportunities available
to both the core, resident population and the in-coming migrant sojourners;
and social changes linked to transformations in inter-generational and inter-
community relations. Some of these transformations have been explored and
 From farm to factory 135
elucidated in a separate paper (Rigg et al., 2008); here we wish to focus parti-
cularly on the ways in which the village, as a settlement and a community, has
been ‘deconstructed’.
The Central region of Thailand, including Bangkok and Bangkok’s
extended metropolitan region, supports the bulk of Thailand’s export-
oriented manufacturing enterprises. The Central Plains has also historically
been the centre of Thailand’s export-oriented rice industry. We have in this
region, therefore, a rich and complex spatial and economic mosaic of factor-
ies and farms. Ayuthaya province, 75 km north of Bangkok by road, was
selected as the study site for the manufacturing frontier component of the
project because of this duality in the province’s character: a past history
framed in terms of its place at the core of Thailand’s ‘rice bowl’ (Molle and
Thippawal Srijantr, 2003); and future prospects seemingly dictated by its
position in Thailand’s industrialisation project. Ayuthaya is the site of
several industrial parks and estates, and the province was also highlighted as
a target province for industrial development in the Sixth National Social and
Economic Development Plan (1987–1991). In mid 2005 there were 1425
officially registered factories in the province employing 192,584 workers.2
This does not include many unregistered – and therefore unrecorded – workers
employed in smaller units of production.3

The research sites

The research was undertaken in two villages in the sub-district (tambon) of
Khan Haam, Uthai District, situated around 15 km east of Ayuthaya city
(Figure 10.1 and Table 10.1): Ban Khan Haam and Ban Khokmayom. The
district is low-lying and much of the agricultural land well irrigated by the
Nakornluang Irrigation Project. Some 87 per cent of land in the district is still
classified as agricultural, and of this the great bulk is paddy land. The district
is also the site of an industrial park and, in this way, has been drawn into
Thailand’s export-oriented, foreign direct investment-driven, industrialisation
drive. Since the mid 1980s there has been a radical reorientation of land-use
within Tambon Khan Haam. Large areas of former agricultural land have

Table 10.1 Ban Khan Haam and Ban Khokmayom, background statistics

Tambon Ban Khan Ban

Khan Haam Haam Khokmayom
(village no. 1) (village no. 9)

Households (2005) 2603 130 –

Population (2005) 5495 494 162
Primary household Farming – 20 38
occupation Non-farming – 110 124

Source: Tambon Administrative Organisation documents.

136 J. Rigg, S. Veeravongs, P. Rohitarachoon and L. Veeravongs

Figure 10.1 Map showing location of study settlements in Central Thailand.

been turned over to industrial uses, some 1200 rai (192 hectares) in total, with
the result that agricultural land in the tambon is now a scarce commodity.
The Rojana Industrial Park is centred on Tambon Khan Haam, although
its land area extends into neighbouring sub-districts. The park was estab-
lished in May 1988 and, with successive waves of expansion, now covers an
area of 4200 rai (672 ha). In 2005 there were 134 factories operating in the
park. The bulk of these – 93 or 69 per cent – were Japanese-owned, and a
further 21 (16 per cent) domestically (i.e. Thai) owned. These factories are
engaged in many of the activities that are emblematic of the country’s indus-
trialisation drive since the mid 1980s: electronics, automobile parts, car
assembly, textiles and apparel, footwear, and food and food processing. In
 From farm to factory 137
2005 the park employed 43,410 workers. The majority of these workers –
70 per cent according to the park management4 – were non-local, mostly from
the poor Northeastern region of Thailand.
Ban Khan Haam and Ban Khokmayom (Table 10.1) were settled more
than a century ago when they were rice-growing communities. Until the
1960s, economic and social life revolved around agriculture and, more par-
ticularly, rice cultivation. At that time only a cart track linked the villages
with the main road to Ayuthaya, and during the rainy season, with extensive
flooding, the settlements were often only accessible by boat. Surpluses of
rice were sold to generate a modest cash income, but Ban Khan Haam and
Ban Khokmayom were not too far from the subsistence, self-reliant ideal
popularised in the work of Chatthip Nartsupha (1986; 1996; 1999).
Since the 1960s, and like other agricultural communities in Thailand, agri-
culture in Tambon Khan Haam has been squeezed by a range of environ-
mental, economic and socio-cultural factors: fluctuating agricultural prices,
especially rice prices; unfavourable input/output ratios; declining terms of
trade; poor water management and continuing flooding; the seasonality of
work; the declining fertility and productivity of the land; and the emergence
of farming as a generally low-status activity. Their location close to an indus-
trial park has provided the inhabitants of the villages, directly (through
employment) and indirectly (through the other demands that the factories
have generated), with a range of alternative occupations and activities. Ban
Khan Haam and Ban Khokmayom were not selected for study just because
we were interested in exploring how rural settlements become implicated in
Thailand’s regional industrialisation strategy, however. We were also intent
on investigating how they become magnets for migrants from other places.
Ban Khokmayom, particularly, has seen the construction of large-scale
dormitories (hor pak) for factory workers, in so doing altering the essence of
the village community. Ban Khan Haam, by contrast, does not support such
dormitories. This is because the settlement is physically (spatially) con-
strained, limiting the scope for substantial building expansion – although it
has, even so, been profoundly influenced by its location at the edge of the
industrial park. While the current (and likely future) history of Tambon
Khan Haam and the two study villages is in the process of being structured
and mapped out by wider national and international policy and investment
decisions, the past history of the sub-district and study villages is better
illuminated by reference to the area’s agricultural and farming roots.
The beginnings of the thorough-going incorporation of the study villages
into the mainstream can be linked to the construction of a laterite feeder road,
which passed close to the gates of the village monastery, Wat Khokmayom, in
the mid 1970s. At that time, rice farming remained the main occupation of
villagers and the non-farm jobs that did exist (mainly in construction) were
undertaken post harvest, and arranged around the demands of agriculture.
The limitations of farming, however, were becoming evident as households
began to find it increasingly difficult to meet their escalating needs from just a
138 J. Rigg, S. Veeravongs, P. Rohitarachoon and L. Veeravongs
single crop of rice, and securing supplementary work during the dry season
became increasingly important, even a necessity. The 1980s saw the emergence
of greater employment opportunities for women as pioneer textile factories in
and around Ayuthaya city opened up for business. The 1990s saw these nas-
cent changes – which can be traced back to the 1970s – accelerate and deepen
in terms of their impact, influence and ubiquity, propelled in no small way by
the opening of the Rojana Industrial Park in 1988.
This event caused land prices to soar, from 3000–4000 baht per rai in the
early 1980s, to 60,000–70,000 in the mid 1980s, rising to 2 million baht by the
end of the decade, and 5–6 million baht/rai today (2006).5 As the temptation
to sell land which had remained in family hands for a century increased,
so its potential for agriculture and to meet household needs was in decline.
Managing water levels in paddy fields became difficult as drainage channels
were blocked and carefully honed systems of water management comprom-
ised. Insects became more of a problem for farmers as expanding areas of
idle land provided a breeding ground for pests. Local people also highlighted
discharges from surrounding factories contaminating the water and land.
Today, farming is the preserve of the very few.6 Instead villagers work in the
industrial park, in the service sector, and as housekeepers, government
officers, teachers, and nurses.
In the discussion that follows we shift our locus of discussion between the
sub-district (the tambon) and the two study villages (Ban Khan Haam and
Ban Khokmayom), according to the topic being addressed. The village
surveys provide data which are relevant to wider-scale transformations and
processes and we use their experiences to reflect on this broader, sub-district,
context. On occasion we combine the survey data from both villages to expli-
cate, for example, what we see to be more generic patterns of livelihood change.
At other times, however, and particularly when the discussion turns to the
place and role of the migrant workers who live in Ban Khokmayom (but not
in Ban Khan Haam), we more specifically highlight certain village-focused
developments and concerns.

Spatial, sectoral and livelihood turbulence on the

factory frontier
Ban Khan Haam and Ban Khokmayom, and Tambon Khan Haam more
widely, present a number of practical, methodological and conceptual chal-
lenges. These challenges arise from the nature of spatial, sectoral and liveli-
hood change in the villages and the sub-district. To begin with, the past
history of the area is agriculture-based and farming-centred. Its present and
likely future are industry- and services-focused and non-farming oriented. In
terms of practical development interventions, therefore, will the promotion
of ‘rural’ development, in the traditional sense, have any traction and effi-
cacy? Second, the core registered populations of the villages (and particularly
Ban Khokmayom) are stable but the de facto population is significantly
 From farm to factory 139
higher and characteristically fluctuating. Methodologically, are we focusing
on the social actors that comprise the resident population of the area, or on
all those who happen to be residing in the geographical space of Tambon
Khan Haam, irrespective of their origins and allegiances? And third, while
the core population does not engage to a great extent in migration it is,
nonetheless – and counter-intuitively – highly mobile. Conceptually, therefore,
how do we think about populations which are, at the same time, becoming
less likely to be migrants, but more likely to be mobile?
Until the 1970s, space, economy, society, identity and livelihoods – broadly
speaking – mapped onto each other. As a researcher it would have been
possible to ‘enter’ the village and, in so doing, to enter the social as well as the
economic universe of the villager. The settlement (the spatial unit), the com-
munity (the social entity), the (village) economy, and (household) livelihoods
would have been revealed as intersecting and overlapping in intimate and
manifold ways.7 Of course this can be stretched too far. The village was not,
in a complete sense, a ‘world unto itself’ (Elson, 1997: 33), but the degree of
interaction and interdependency with entities beyond the locale was distinctly
limited, compared with today.
The progressive dissociation of the village in spatial, social, economic
and livelihood terms since the 1970s is not just important because of its
implications for how we understand the development process and appro-
priate development interventions, but also because of the implications it
has for research methods and our conceptualisation of ‘the village study’.
Focusing on the latter, it means that the entry point for the traditional village
study – namely, ‘the village’ – is no longer quite so neat and unproblematic.
The village is clearly still there, in the rural landscape, but as a container for
study it is shot through. The population is in flux; the space economy con-
sists of elements (such as the industrial estate) which are largely rooted
and driven extra-locally; village livelihoods are dependent on activities in
other spatial and sectoral arenas; even village identities have shifted from
their prior self-identification as chao naa – rice farmers – to something
more ambiguous.
The degree of turbulence is most strikingly illustrated when it comes to
pinning down the population of Ban Khomayom. The Uthai district census
bureau provides a figure for 2005 of 378. The tambon health station, for the
same year, has records indicating that the population of the village is 1257.
Based on our village survey and the number and size of dormitories in the
village, however, we would put the de facto population of Ban Khokmayom
at around 3000. This marked discrepancy is due to the fact that the migrants
who throng the hor pak are not registered as de jure residents of the village,
and so do not appear in the official figures. This presents both a practical,
methodological challenge and a more conceptual one: rhetorically, how do
we treat this shifting, migrant population when it comes to interpreting
village change?
140 J. Rigg, S. Veeravongs, P. Rohitarachoon and L. Veeravongs
Interpreting turbulence on the factory frontier
The dissociation of work from the study villages began, as noted above, in the
early 1970s when a number of textile factories were established in Ayuthaya
and a small number of local women took up jobs to supplement their
incomes. These original textile factories have since closed in response to com-
petition from other, lower-wage locations in Asia. Some more adventurous –
or more desperate – women migrated temporarily to work in textile factories
in Rangsit District, Pathumthani Province, while men left to work on the
construction sites of Bangkok. But it was the 1980s that saw really marked
structural and livelihood change in the tambon, associated with the opening
of the Rojana Industrial Park.
The occupations of adults included in the survey clearly reveal how far
farming has become not just a subsidiary occupation, but a relict one under-
taken by a few intransigent die-hards: in the two villages, of the core popula-
tion aged over 16 years old, just three individuals out of 508 reported that
farming was their main occupation. Moreover, it was clear that this trans-
formation from farm to non-farm was well advanced in the mid 1990s; some
30 individuals reported that farming was their main occupation ten years prior
to the survey. Skilled, semi-skilled and unskilled (casual) work, very largely
located outside the boundaries of the village, provided the main occupations
for over half of adults in the two villages. Given that another fifth to a quarter
remained in education, this means that of economically active adults the
significant majority build their livelihoods largely beyond the settlement.
Furthermore, the trading, shop-keeping, restaurant and ‘private’ (mostly
dormitory) businesses of around another 10 per cent of adults – which are
largely situated within the confines of the village – are sustained by the
purchasing power of migrants residing in the tambon but who work in the
industrial estate and other factories in the vicinity.
The factories of the Rojana Industrial Park not only provide employment
for local people, but also act as magnets for workers from less prosperous
areas of the country. While some factories provide on-site accommodation,
most do not and the villagers of Ban Khokmayom particularly have exploited
the demand for accommodation by building dormitories or hor pak.8 In
2006 there were around 30 dormitories within Ban Khokmayom, ranging
from small, single-storey affairs with just a handful of rooms, to grandiose,
multi-storey structures with 50 to 100 rooms (Illustration 10.1). Many of
these have attached toilet facilities and rental rates of 1500–2000 baht a month
(US$45–60). The tambon as a whole supports approximately 300 dormitories
and Tambon Thanu – the neighbouring tambon – a further 70 (interview with
Uthai assistant district officer, 5 June 2006).
The migrants who live in the physical space – but not the social space – of
Ban Khokmayom are mostly from the Northeast of Thailand. Their clear and
driving rationale for coming to Ayuthaya is to find work. This, in itself, is
not new. There has long been a flow of migrants from the environmentally
 From farm to factory 141

Illustration 10.1 Dormitory block in Ban Khokmayom.

Photo by Jonathan Rigg

marginal and economically impoverished Northeast to the richer Central

Plains. Traditionally, however, these migrants worked in the rice fields of the
region; now they migrate, and in much larger numbers, to work in the factor-
ies of the area. We found no migrant sojourners in Tambom Khan Haam who
were employed in the farm sector, although in other areas of the province this
is likely to be the case.
Traditionally, in rural village studies, land is the key production resource,
the central marker of wealth and, by association, the key indicator – through
its lack – of poverty. The centrality accorded to land remains a guiding
principle in many studies, as Borras et al. (2007: 1) make clear in the introduc-
tion to their book on land, poverty and livelihoods: ‘in many agrarian settings
a significant proportion of the income of the rural poor still comes from
farming . . . [and] hence lack of access to land is strongly related to poverty
and inequality’. What we identify in this area of Ayuthaya, however, is a
three-fold de-linking of land from its formerly central role: a de-linking of the
village from its land resource so that much village land is no longer owned
and operated by villagers; a delinking of livelihoods from land; and a delink-
ing of inequality and poverty from land. Across the two villages, of the 133
‘core’ households surveyed (i.e. excluding migrants to the villages), not one
household head reported farming as their main occupation, and just three
142 J. Rigg, S. Veeravongs, P. Rohitarachoon and L. Veeravongs
out of 508 aged over 16 years were said to be engaged, primarily, in farming
activities. In total, 39 households or 29 per cent owned land, but much of this
was not located in the village, and an even smaller proportion – less than a
fifth – was farmed by its owners. It was in the early 1980s, informants told us,
that villagers began to sell land to outsiders, tempted by the simply enormous
sums being offered. The two study villages may still be situated in a rural
setting (Illustration 10.2), but much of the surrounding land is not owned
by villagers, and still less is it instrumental in structuring livelihoods and
explaining patterns of well-being.
It has been normal to understand the engagement with activities outside
farming as linked to life-course transitions. Young people leave home to earn
income and see the world as part of a rite of passage into adulthood. In time,
it is assumed, many will return to farming and, if they have left home, to the
village. It is certainly true that there is a generational pattern to employment
in the study villages (Figure 10.2). But, we argue, this is indicative of era-level
transformations rather than generational changes, with the non-farm/farm
transition occupying a past transitional stage. Instead, we see working through
the population and, we surmise, in quite a permanent manner, a shift from
unskilled and casual employment to skilled and semi-skilled. But unlike the
‘return to farming’ argument, we do not expect there to be a similar ‘return to
casual work’. This is a permanent, two-stage change in the complexion of the

Illustration 10.2 Ban Khokmayom and Rojana Industrial Park.

Photo by Jonathan Rigg
 From farm to factory 143

Figure 10.2 Age and occupational status of adult household members: Ban Khan
Haam and Ban Khokmayom, 2005.
Source: Authors’ household survey, 2005.

Figure 10.3 Formal/skilled and casual/unskilled occupations among working-age

household members: Ban Khan Haam and Ban Khokmayom, 2005.
Source: Authors’ household survey, 2005.

rural economy and rural livelihoods: First, from farm to non-farm (which
occurred from the mid 1980s); and second, from un-skilled/casual to skilled/
semi-skilled (which occurred from the mid 1990s) (Figure 10.3). The first of
these may, to begin with, have reflected generational changes associated with
the life-course. The second, though (and, we suspect, increasingly the first) are
transformations which will resonate through the generations.
Of course, there is always the possibility that transitions/transformations
may be thrown into reverse, particularly when national economic progress is
replaced by stagnation or retrogression. This was anticipated, for example,
with the onset of the Thai economic crisis of 1997 (Parnwell, 2002; Rigg,
2002). The economic attractiveness of farming can also shift, and the recent
144 J. Rigg, S. Veeravongs, P. Rohitarachoon and L. Veeravongs
(2008) surge in agricultural commodity prices may lead some people to return
to farming or to realign the balance of their work between farm and non-
farm. Thus the current trend of de-agrarianisation may shade into a degree
of re-peasantisation given changing circumstances. Our view, however, is that
such a shift ‘back’ to farming will not occur readily or easily, and only in
markedly changed circumstances.
These changes in the complexion of livelihoods and the village economy
have a bearing on the sustainability of the village and the nature and inter-
play between production and consumption. Formerly it was not just produc-
tion that was centred – spatially – on the village, but so too were patterns of
consumption. Through to the 1970s, the village was largely self-sufficient in
food. The main source of protein was fish caught in local waterways, most
households kept poultry, grew fruit and vegetables in their house compounds
and rice, of course, was cultivated by almost everyone. Little was purchased.9
From the 1970s, and progressively since, the villages have become dependent
on external sources and supplies of food, and much else beside. Today, vil-
lagers buy goods from the local tambon market and from ‘mobile markets’
(rod kab khao), which are supplied from Bangkok’s wholesale markets, or
they travel to supermarkets on the outskirts of Ayuthaya, such as Big-C,
Tesco-Lotus and Carrefour. There has also been a proliferation of shops in
the tambon, but these are convenience stores and air-conditioned minimarts
selling pre-packed, and often pre-cooked foods, serving the needs of migrant
workers and villagers who have disposable income but little disposable time.
In addition to food, other aspects of village life have been de-localised, such
as schooling and entertainment. Better-off villagers send their children to
schools in Uthai district or Ayuthaya town, which are thought to provide a
better education than the local school. Privately operated school buses pick
children up each morning, depositing them back in the village at the end of
the school day.
As noted above, Tambon Khan Haam – and particularly Ban Khokmayom
– has become a node of attraction for migrants from other regions of Thailand.
It is not, therefore, just that the original settlements are being prised apart by
the movement of their core population into activities which are distant geo-
graphically and historically from the original community. It is also that the
settlements themselves are being colonised by migrants from other places who
hold no great allegiance to the village. Ironically, while these migrants may be
playing a role in dissolving the village as a community, they are also, by their
presence, helping to sustain the village as an economy and as a settlement.
The demand for accommodation, not to mention other amenities and ser-
vices from beauty salons and laundry shops to minimarts and food stalls,
sustains the livelihoods of the core population, keeping them in the village
even while – some villagers thought – their community was fragmenting.
Perhaps it is for this reason that while we identify a dissociation of the
traditional village community, we do not see a dissolution of the village.
Indeed the core village, on paper, seems surprisingly resilient. The majority of
 From farm to factory 145
core villagers (i.e. excluding migrant sojourners) were born in their village of
residence or in the neighbouring district: 77 per cent and 66 per cent in Ban
Khan Haam and Ban Khokmayom respectively. In addition, surprisingly
few of the surveyed household members – less than a fifth – had engaged in
migration involving an absence of three months or longer. These data would,
at first sight, indicate that the villages are socially and economically quite
‘rooted’. Moreover, this is surprising insofar as Thai society has become
increasingly mobile.
On closer inspection, however, the data reveal that mobility is taking on a
new guise in the study sites. Rather than long-distance and long-term patterns
of movement (i.e. migration), it is quite intense patterns of daily mobility that
characterise the villages. Ease of travel coupled with a proliferation of local
non-farm employment opportunities have permitted villagers to leave farm-
ing, without leaving the village. In the day before the survey, around one half of
the 669 household members surveyed had left the village. Of these daily move-
ments, more than 60 per cent were for work or business reasons (Table 10.2).
The ease with which people can move relatively long distances, quite quickly,
is significant, particularly when compared with the African cases in this
volume. Factories in the Rojana Industrial Park provide buses to pick up
workers daily up to a radius of 100 km; furthermore, many workers have their

Table 10.2 Mobility in Ban Khokmayom and Ban Khan Haam

Ban Khan Ban

Haam Khokmayom
(core villagers)

Household members who have engaged in long term 14 21

migration (> 3 months’ absence) (%)
Household members who travelled outside the village in 43 58
the day before the interview (%)
Purpose of daily mobility Work, business 79 148
(number of household
members) Social 17 46
Education 30 52
Total 126 246
Mode of daily travel (number Private (motorbike, car, 81 133
of household members) pick-up)
Public or firm-provided 31 76
transport
Other 14 37
Total 126 246
Household ownership of Motorbike 51 72
motorised transport
(number of households) Motor vehicle (car, 20 49
pick-up)

Source: Authors’ household survey, 2005.

146 J. Rigg, S. Veeravongs, P. Rohitarachoon and L. Veeravongs
own, private means of transport (Table 10.2). Of the households interviewed,
over 80 per cent owned a motorcycle, and 45 per cent a motor vehicle. There
is, as one might expect, a correlation between daily mobility and age
(Figure 10.4). The members of the cohort aged 0–16 years are primarily
leaving the villages for purposes of education; those in the working age
cohorts (17–30 years and 31–50 years) depart mainly for work-related
reasons; while private errands are concentrated among those villagers aged
over 30 years. It is the members of the oldest cohort (51+ years) who are least
mobile.
It has been noted elsewhere that development studies and development
economics have overlooked the role of mobility in re-working the nature and
basis of rural development. As Dercon writes, ‘understanding rural poverty
changes cannot naively focus only on what happens in the rural sector, or in
agriculture . . . if development has to do with people moving between sectors,
then this will often imply spatial movement’ (2006: 5 and 6). This experience
from Ayuthaya takes the argument one step further in its reconsideration of
the bases of rural vitality and development. In Ban Khan Haam and Ban
Khokmayom there has, indeed, been a very pronounced shift from farm to
non-farm in people’s livelihoods. This has been accompanied by marked
structural changes in the village economy. There is also a high intensity of
mobility. But, at least for the core households in the villages (i.e. excluding
migrants), this has not involved leaving the village. The colonisation of rural
space by industrial activities, coupled with an efficient transport infrastructure

Figure 10.4 Daily mobility and purpose on the day prior to the survey: Ban Khan
Haam and Ban Khokmayom, 2005.
Source: Authors’ household survey, 2005.
 From farm to factory 147
and high levels of ownership of private means of transport, has enabled the
village economy and rural livelihoods to be fundamentally re-worked without
the spatial movements of people that Dercon highlights.

Conclusion: re-making the Thai rural economy and

rural livelihoods
The implications of structural and spatial changes in the Thai economy for
rural livelihoods have been noted elsewhere (see Molle, 2003; Molle and
Thippawal Srijantr, 1999; Rigg, 2001, 2006; Rigg and Sakunee Nattapoolwat,
2001). In large part, this work has focused on the way in which people have
been attracted (or propelled) out of the village and farming, by opportunities
in other spheres and/or by a squeezing of traditional, rural livelihoods. This
has necessitated, in many instances, long-term migration, denuding the
village of its human resource base and leaving it ‘bereft’ of young people
of working age. Outstanding questions that have not yet been adequately
explored include first, whether these trends indicate a permanent trans-
formation in the character of the rural economy and rural livelihoods;
and second, how these changes are restructuring the bases of poverty and
prosperity in the countryside.
The discussion in this chapter has contributed a slightly different view
to the debate over de-agrarianisation in Thailand and, more broadly, in Asia.
But before generalising from the experience of Ban Khan Haam and Ban
Khokmayom it needs to be emphasised that the villages are atypical in
their sheer proximity to the Rojana Industrial Park and, in the case of Ban
Khokmayom, in its emergence as a ‘dormitory village’. Nonetheless, we
highlight as a conclusion two key issues that, we believe, have wider relevance.
To begin with, there is the issue of the broad-level and multi-faceted
‘dissociation’ of the study sites. Thompson has discussed ‘dissociation’ with
reference to the Malay-peasant complex in his study of Sungai Siputeh
(Thompson, 2002, 2003, 2004, 2007). His main concern is to explore dissoci-
ation from the village of young men (particularly) who have been drawn into
the modern economy, creating a degree of separation from their traditional
identity and identification as orang kampung (people of the village).10 In the
cases of Ban Khan Haam and Ban Khokmayom, we are interested as much
in the dissociation of the village as in dissociation from the village. There is,
quite clearly, a relationship between the two insofar as the village is not just a
settlement – a place – but a community which is socially and culturally
‘made’. That said, while the social and cultural may frame the meanings that
villagers attach to ‘their’ village, there are also framing structures that are
economic and material: patterns of land ownership, the presence of hor pak,
idle land, the shift from water to road transport, polluted waterways, the
gates and barred windows that protect villagers and their new-found wealth
from the strangers who throng the village, beauty salons and minimarts,
motorcycles and, of course, the factories that lie just across the fields. We have
148 J. Rigg, S. Veeravongs, P. Rohitarachoon and L. Veeravongs
sought to describe this multi-faceted dissociation to highlight the degree to
which a formerly quite tight association, or nesting, of space, economy and
identity has become fragmented by the intertwined processes of social, eco-
nomic and spatial change.
The second theme that the chapter has explored is that of mobility.
As noted above, there is a relative abundance of studies of migration in
Thailand.11 Most of these emphasise migration and the associated spatial and
temporal dislocation of people from the village. In this chapter, however, we
note the degree to which daily mobilities rather than migration characterise
the study sites, as villagers, of all ages but particularly those of working age,
move rapidly and frequently across a vital space of economic and social
interaction.12 The Central Plains region of Thailand, the so-styled ‘rice bowl’
of the Kingdom, has become functionally part of the extended Bangkok
metropolitan region. Villages such as Ban Khan Haam and Ban Khokmayom
have been drawn directly into this process, not as reservoirs of cheap labour,
but as sites of industrialisation and metropolitanisation. Seductively, the
village is still to be seen and studied; it exists as an administrative unit and
as a physical entity. But in many other respects the village has become an
historical artefact, replaced by a settlement with a population who owe little
allegiance to the land and who have become dependent on work ultimately
shaped by forces that lie beyond the area.

Notes
1 For Thai Board of Investment (BoI)-sponsored industrial estates, see http://
www.boi.go.th/english/how/industrial_estates.asp.
2 Source: www.industry.go.th/min/intro/province/Ayuthaya/web/factorydata.com.
3 The total population of the province at the end of 2004 was 740,000.
4 Interview with the Rojana Industrial Park officer, 3 November 2005 and the
General Manager of the Rojana Industrial Park Company, 16 February 2006.
5 The baht:US$ exchange rate for much of the 1980s and 1990s was 25 baht =
US$1. The Thai economic crisis in 1997–98 led to a sharp decline in the value of
the baht, bottoming out at 56 baht = US$1 in January 1998. In June 2008 the
exchange rate was US$1 = 33 baht.
6 Note that the fieldwork and survey were undertaken before the recent steep
increase in the value of rice (and other agricultural commodities).
7 Because of the nature of the landscape in the Central Plain, villages tend to be
strung out along waterways rather than nucleated in the way that they are in the
Northeastern region and much of the North.
8 We use the word ‘dormitory’ because these rooms and buildings are generally
known as hor pak – usually translated as dormitory. However, they are not dormi-
tories in the sense that several people live together in the same room. The rented
rooms are for single migrants or couples and newer hor pak have attached toilet
facilities.
9 Amyot, in his study of village Ayuthaya undertaken in 1969–1970, writes: ‘Food
consumed in the household is usually that which is found locally such as fish,
vegetables from the family garden, and of course rice. Little money is spent on
food so that meat and other vegetables not produced by the household are bought
only in small quantities’ (Amyot, 1976: 80).
 From farm to factory 149
10 ‘Dissociation, in my view, however, does not signal an end to social and cultural
structure or a moment in which “anything goes.” Rather, it is a moment in which
the prevailing social forces and subjects acting within the constraints of those forces
rework the landscape of identity, doing so at points of structural dissociation and
indeterminate articulation, such as the points at which Malayness, masculinity,
migrancy, and class intersect in Malaysia’ (Thompson 2003: 431).
11 See, for example, Esara (2004), Jones and Sirinan Kittisuksathit (2003), Jones and
Tieng Pardthaisong (1999), Mills (1997;1999), Rigg (1989), Singhanetra-Renard
(1999), Tomosugi (1995).
12 In fact, the villages reveal two forms of movement: the migration of people to the
area from other parts of Thailand; and the mobilities that take both these
migrants and the core population out of the village on a daily basis.

References
Amyot, J. (1976) Village Ayutthaya: Social and economic conditions of a rural popu-
lation in Central Thailand, Bangkok: Chulalongkorn University Social Research
Institute.
Borras, S.M. Jr, Akram-Lodhi, A.H. and Kay, C. (2007) ‘Agrarian reform and rural
development: Historical overview and current issues’, in A.H. Akram-Lodhi, S.M.
Borras Jr and C. Kay (eds), Land, poverty and livelihoods in an era of globalization:
Perspectives from developing and transition countries, London: Routledge, pp. 1–40.
Chatthip Nartsupha (1986) ‘The village economy in pre-capitalist Thailand’, in Seri
Phongphit (ed.), Back to the roots: Village and self-reliance in a Thai context,
Bangkok: Rural Development Documentation Centre.
Chatthip Nartsupha (1996) ‘The village economy in pre-capitalist Thailand’, in M.C.
Hoadley and C. Gunnarsson (eds.), The village concept in the transformation of
rural Southeast Asia, studies from Indonesia, Malaysia and Thailand, Richmond,
Surrey: Curzon.
Chatthip Nartsupha (1999) The Thai village economy in the past, Chiang Mai,
Thailand: Silkworm Books (transl. Chris Baker and Pasuk Phongpaichit).
Dercon, S. (2006) Rural poverty: Old challenges in new contexts, GPRG-WPS-072,
Global Poverty Research Group, downloaded from: http://www.gprg.org/pubs/
workingpapers/pdfs/gprg-wps-072.pdf, accessed 15 December.
Elson, R.E. (1997) The end of the peasantry in Southeast Asia: A social and economic
history of peasant livelihood, 1800–1990s, Basingstoke: Macmillan.
Esara P. (2004) ‘ “Women will keep the household”: The mediation of work and
family by female labor migrants in Bangkok’, Critical Asian Studies, 36: 199–216.
Jones, H. and Sirinan Kittisuksathit (2003) ‘International labour migration and qual-
ity of life: Findings from rural Thailand’, International Journal of Population
Geography, 9: 517–30.
Jones, H. and Tieng Pardthaisong (1999) ‘The impact of overseas labour migration on
rural Thailand: Regional, community and individual dimensions’, Journal of Rural
Studies, 15(1): 35–47.
Mills, M.B. (1997) ‘Contesting the margins of modernity: Women, migration, and
consumption in Thailand’, American Ethnologist, 24: 37–61.
Mills, M.B. (1999) Thai women in the global labor force: Consumed desires, contested
selves, New Brunswick, NJ: Rutgers University Press.
Molle, F. (2003) ‘Knowledge in the making: A brief retrospective of village-level
studies in the Chao Phraya Delta during the 20th century’, in F. Molle and
150 J. Rigg, S. Veeravongs, P. Rohitarachoon and L. Veeravongs
Thippawal Srijantr (eds), Thailand’s Rice Bowl: Perspectives on agricultural and
social change in the Chao Phraya Delta, Studies in Contemporary Thailand 12,
Bangkok: White Lotus, pp. 11–35.
Molle, F. and Thippawal Srijantr (eds.) (2003) Thailand’s Rice Bowl: Perspectives on
Agricultural and Social Change in the Chao Phraya Delta, Studies in Contemporary
Thailand 12, Bangkok: White Lotus.
Molle, F. and Thippawal Srijantr (1999) Agrarian change and the land system in the
Chao Phraya Delta, DORAS-DELTA research report no. 6, ORSTOM, Bangkok:
Kasetsart University.
Parnwell, M.J.G. (2002) ‘Coping with crisis and migration “reversal” in Thailand’, in
P.P. Masina (ed.), Rethinking development in East Asia: From illusory miracle to
economic crisis, Copenhagen: NIAS, pp. 261–82.
Rigg, J. (1989) International contract labor migration and the village economy: The case
of Tambon Don Han, Northeastern Thailand, papers of the East-West Population
Institute, East-West Center, Hawaii, No. 112 (July).
Rigg, J. (2001) More than the soil: Rural change in Southeast Asia, Harlow, Essex:
Prentice Hall.
Rigg, J. (2002) ‘Rural areas, rural people and the Asian crisis: Ordinary people in a
globalising world’, in P.P. Masina (ed.), Rethinking development in East Asia: From
illusory miracle to economic crisis, Copenhagen: NIAS, pp. 241–60.
Rigg, J. (2006) ‘Land, farming, livelihoods and poverty: Rethinking the links in the
Rural South’, World Development, 34(1): 180–202.
Rigg, J. and Sakunee Nattapoolwat (2001) ‘Embracing the global in Thailand:
Activism and pragmatism in an era of de-agrarianisation’, World Development,
29(6): 945–60.
Rigg, J., Suriya Veeravongs, Lalida Veeravongs and Piyawadee Rohitarachoon (2008)
‘Re-configuring rural spaces and re-making rural lives in Central Thailand’,
Journal of Southeast Asian Studies, 39(3): 355–81.
Singhanetra-Renard, A. (1999) ‘Population mobility and the transformation of the
village community in Northern Thailand’, Asia Pacific Viewpoint, 40(1): 69–87.
Thompson, E.C. (2002) ‘Migrant subjectivities and narratives of the kampung in
Malaysia’, Sojourn, 17(1): 52–75.
Thompson, E.C. (2003) ‘Malay male migrants: Negotiating contested identities in
Malaysia’, American Ethnologist, 30(3): 418–38.
Thompson, E.C. (2004) ‘Rural villages and socially urban spaces in Malaysia’, Urban
Studies, 41(12): 2357–76.
Thompson, E.C. (2007) Unsettling absences: Urbanism in rural Malaysia, Singapore:
NUS Press.
Tomosugi, T. (1995) Changing features of a rice-growing village in Central Thailand: A
fixed-point study from 1967 to 1993, Tokyo: Centre for East Asian Cultural Studies.
11 Handicraft and
manufacturing frontier
Transforming livelihoods,
re-shaping settlements
Katherine V. Gough and
Jonathan Rigg

Introduction
Artisanal craft production has long been an important activity for many
rural households, often complementing but at times taking precedence over
agricultural activities. While this production was initially for home consump-
tion or sale locally, increasingly handicrafts produced in rural areas have
found their way to national and international markets. The potential benefits
of rural industrialisation have been recognised by many national govern-
ments, especially in Asia, who see rural industry as having the potential to
absorb surplus labour while reducing migratory pressure on cities (Tacoli,
2008). Thus handicraft production is being supported by national govern-
ments and manufacturing industries are being encouraged to relocate to
rural areas.
The three core research chapters comprising this section of the book have
analysed rural industrialisation in Thailand and Vietnam with a particular
focus on changes in handicraft production. Each chapter emphasised, in dif-
ferent ways, the degree to which ‘rural’ and ‘urban’ and the activities, lives
and lifestyles that characterise these spaces are being unsettled by the trans-
formations under way in the study sites. This leads to some quite fundamental
questions about the virtuous cycle of farm–nonfarm relations and how
development interventions are structured and targeted in such development
spaces. It also leads us into a consideration of the constitution or make up –
both material and non-material – of the spaces that we label ‘rural’ and
‘urban’. In this short chapter, therefore, we reflect on these wider issues in
the light of the experience of the study villages in Thailand and Vietnam.

Virtuous cycle of farm–nonfarm relations

A long-standing debate among scholars of rural development has been
whether there is a virtuous cycle of farm–nonfarm relations, or whether the
relationship is insignificant or, even, negative. For proponents of rural indus-
trialisation, one of its key attractions is the notion that it can create the
152 Katherine V. Gough and Jonathan Rigg
momentum for a self-sustaining process of rural development, helping to
maintain the integrity of rural communities, boost agricultural output,
ameliorate rural poverty, and sustain rural livelihoods.
In this virtuous cycle of farm–nonfarm relations, rising agricultural
incomes create a demand for consumer goods and services, helping to drive
rural development by stimulating the rural non-tradable sector (including
handicrafts), which helps to absorb un- and underemployed workers (surplus
farm labour) in the countryside. This, in turn, further boosts demand for
farm output while contributing cash for investment in agriculture, thus
propelling yet additional increases in agricultural production (Timmer, 2007;
Briones, 2006; Evans, 1992; Evans and Ngau, 1991; Grabowski 1995; and
Haggblade et al., 1989). This model has two particular attractions for rural
development practitioners. First, it generates pro-poor growth by expanding
employment and incomes among the rural poor(er); and second, it is geo-
graphically focused in the countryside, narrowing rural–urban inequality,
thereby acting against the forces promoting urban bias, and helping to pre-
vent the dissolution of the village as a social and economic unit.
The notion, however, that there is a synergistic relationship between farm
and nonfarm has been vigorously challenged. Hart, for example, regards such
a view as ‘deeply suspect’ (1996: 246). In particular, she disputes the basis on
which the farm–nonfarm multipliers are calculated, contesting the assump-
tion that agricultural growth will lead to the expansion of the local nonfarm
economy. Instead she postulates that much of the increase in income generated
by agricultural growth will leak out of the immediate vicinity into the wider
national economy (Hart, 1996: 251–2). Furthermore, there are cases where
rural industrialisation has challenged the agricultural sector, and rather than
stimulating increases in productivity is causing just the reverse effect.
How does the evidence from the craft villages in Thailand and Vietnam
contribute to this debate? Certainly, rural incomes have been boosted,
although largely through nonfarm employment rather than increases in agri-
cultural productivity. Indeed, agricultural productivity has been compromised
by nonfarm work both directly through extracting the rural population from
farming and, indirectly, by contributing to the declining status of farming as
an activity and a way of life. Only in the Vietnamese embroidery village do we
see any real evidence of inter-locking livelihoods emerging, where farming
remains an important component in most households’ livelihoods and where
handicraft production is shaped around the demands of farming. In the
Vietnamese lacquerware village and in the saa paper and mango wood-craft
villages in Thailand, craft work has become full-time, with a degree of
specialisation, separation of tasks, investment in technology, and use of
hired – sometimes migrant – labour.
Why have these differences among craft villages emerged? The most obvi-
ous explanation lies in the nature of the handicraft itself. Embroidery is a
craft which has changed little over the centuries. Although the markets for the
final products have changed over time, the nature of the work remains the
 Handicraft and manufacturing frontier 153
same. The embroidery is still done by hand, using skills that take considerable
time to learn and are passed down through the generations, usually from
mother to daughter although previously men also engaged in embroidery
work. Some younger women are continuing the tradition but the craft is
increasingly seen as being a last resort for uneducated people who cannot
access any alternative income-generating activity. The nature of embroidery
work means that it can easily be combined with agriculture – the handicraft
will not spoil if it is left while the crops are tended. For the other handicrafts
studied, the production process has become ‘modernised’ – and in some cases
partially mechanised. The subsequent division of labour means that the
owners can increasingly specialise in putting the final touches to the handi-
crafts and in the development of new products and marketing leaving the, at
times dangerous, production to paid workers. As the profits which can be
made from handicraft production rise, and as the nature of the production
process becomes more inflexible making it harder to stop and start, so less
time is spent on agricultural work. Craft work of this type appeals to young
people and many of the craft producers no longer see themselves as farmers
but as entrepreneurs running a business.
Much the same sequence of farm–nonfarm interactions can be seen oper-
ating in the factory villages of Ayuthaya in Thailand. The demands of such
work entail working a full day, leaving little or no time to engage in agri-
cultural work. Young people generally view factory employment, or service-
related work supporting factory workers, as more attractive and amenable
than agricultural endeavours. In addition, the environmental consequences of
factories locating in such a rural context have knock-on effects for farming.
In such a context, it is hard to see agriculture as a motor for rural development,
and any synergies between farm and nonfarm are more imagined than real.
A central assumption of the virtuous cycle of farm–nonfarm relations is
that the two facets of the rural economy are co-situated. Our studies, however,
highlight the manner in which craft and factory production relies on – and is
sustained by – wider links and associations. Handicraft production can no
longer rely on local sources of raw materials as the supplies run out (such
as local wood sources) or as cheaper/better alternatives which are produced
elsewhere can be bought (such as chemical dyes). Increasing demand for the
products requires the use of waged, often migrant, labour who tend to do
the most menial tasks which can sometimes be dangerous to their health.
Handicrafts have found their traditional markets eroded through competition
from non-local substitutes; plastic and aluminium containers have replaced
terracotta pots, manufactured paper has replaced saa paper, and posters have
replaced embroidered pictures. To survive and expand, it has been necessary
to connect handicrafts with non-local demand and adapt the products them-
selves to new markets. Saa paper is used to make stationery attractive for
gifts, colourful lacquerware adorns many a mantelpiece worldwide, and small
pieces of embroidery embellish trinkets and garments. What may have started
life as rural activities, using rural inputs, populated by (local) rural people,
154 Katherine V. Gough and Jonathan Rigg
and serving a largely rural market, has been transformed. Today, the links
between locality and activity are fractured. This argument about the partial
disconnection of rural industries from their rural location is, of course, even
more pronounced when it comes to Ayuthaya’s factory villages. Such factor-
ies may be situated in the countryside, but they are not of that space in any
profound sense.
On balance, and based on the Thai and Vietnamese case studies in this
section of the book, we are sceptical about the assumption in some quarters
that there is a virtuous cycle of farm–nonfarm relations. This view is per-
vasive. The 2009 World Development Report, for example, argues that the
evidence suggests that ‘growth in the nonfarm sector will stimulate growth in
agriculture as inputs become cheaper, profits are reinvested in agriculture,
and technological change allows better farming methods’ (World Bank, 2009:
200). While there may be cases where such a set of links and interactions do
reveal themselves, the chapters here indicate that demand from rising incomes
in the nonfarm sector did not spill over into agriculture, and that farming
was therefore not invigorated by the process, indeed often the reverse.

Forces shaping rural and urban spaces

As highlighted above, the associations and networks that sustain ‘craft’
activities in northern Thailand and northern Vietnam, and which have
initiated factory-based rural industrialisation in central Thailand, cannot be
understood without viewing their current vitality and future sustainability in
wider regional, national and international terms. Hence, the forces that are
shaping rural and urban spaces operate at a range of spatial scales.
In the case of the Ayuthaya villages in central Thailand, it is their links
with foreign-owned factories in the vicinity and the associated ties with
migrant sojourners from other, poorer rural areas of Thailand which are
critical in understanding the processes under way. These, in turn, need to be
understood in the context of Thai regional development policies which have
promoted and supported the spill-over of industries from the core Bangkok
metropolitan region to surrounding provinces (of which Ayuthaya is one),
and further afield. The character of the handicraft settlements in northern
Thailand, although they may lie within the ambit of the local provincial
capital of Chiang Mai, is being shaped by global networks of buyers rather
than by dynamics which are rooted in the immediate provincial context. In
addition, the emergence of Chiang Mai as an important node in Thailand’s
international tourist industry further links the growth and sustainability of
the industry with non-local networks and flows. In Vietnam, the introduction
of economic policies under Doi Moi provided the impetus for many handi-
craft producers to expand their businesses and increase their export trade.
But while national and provincial bodies have introduced policies that are
supportive of craft production, at the local level (commune and district) there
is more reluctance to implement appropriate policies. Local policies tend to
 Handicraft and manufacturing frontier 155
be piece-meal in fashion often aiming to levy the producers, which can then
restrict their development. Much needed environmental control and labour
protection policies for some types of craft production have been lacking,
severely impacting on the living and working conditions in the settlements.
The fortune of the handicraft producers is thus dependent on local and
global demand, national policies, and the particularities of the handicraft
in question.
If we shift the focus of our discussion from the activities or industries
themselves to the social contexts in which they occur, we identify an equally
profound set of transformations operating at the settlement level. The
increasing need for waged labour has resulted in migrants moving into the
settlements, often housed in rudimentary conditions and sometimes sleeping
in the workshops themselves. These migrants often owe little allegiance to
the places where they sojourn, leading to a dissociation of the settlement as
a community. More tangibly, profits from the craft production have been
invested in improved homes. In Vietnam this has resulted in multi-storey
dwellings built side by side, resembling those found in nearby Hanoi. In
Thailand, large villas have been built which would not be out of place in the
suburbs of Chiang Mai. Services have been brought into the settlements
which are now reached by paved roads and supplied with electricity, water
and drainage. The inhabitants are less engaged in farming and increasingly
in specialised handicraft production tasks either as employers or employees.
In the handicraft settlements in Thailand and Vietnam we are seeing, as
Thompson proposed in the context of Malaysia, rural villages becoming
de facto ‘urban’ spaces in terms of production, consumption and social
interaction (Thompson, 2004; 2007). Bramall, writing of rural industrialisa-
tion in China, highlights another aspect of this process of urban ‘becoming’:

Any rural area which is successful is re-classified as urban, and its urban
status determines how it is viewed for analytical purposes. In consequence
the rural–urban income divide never closes: how could it when any area
which successfully industrialises is automatically re-classified as urban?
The reality is that ‘urban’ and ‘rural’ are not used to denote location,
but rather as pseudonyms for ‘successful’ and ‘failed’.
(Bramall, 2007: 323)

Essentially, we are suggesting that the transformations under way in the

study settlements in Thailand and Vietnam are not manifestations of the
incorporation of the villages into an extending urban fabric, nor are they
reflections of the power and influence of urban areas over rural spaces and
lives. Rather they are indicative of a diverse and diffuse set of shaping pro-
cesses that operate at varying scales. In the same way that Shen and Ma
(2005) have called our attention to the varying drivers of urbanisation in
China, in which they distinguish between urbanisation from above and
urbanisation from below (or in situ urbanisation), so we would highlight the
156 Katherine V. Gough and Jonathan Rigg
need to see rural change being shaped by an array of forces with multiple
locations and sometimes conflicting or divergent outcomes. The discourse
of ‘rural’ development is seductive but, in our view, often tends to divert
attention from a set of important dynamics associated with societal change,
settlement dynamism and sustainability, population turbulence, and evolving
cultural preferences.

Conclusion
A strong case can be presented that none of the study villages in the three
core chapters that comprise this section of the book is rural, in the classical
sense. They may have histories that are rooted in the rural context, they may
even resonate with a particular vision of rurality, but in terms of livelihoods,
occupations, aspirations, social mores, consumer preferences and modes of
living they have a character which is more urban than it is rural.
While there are clearly links between the villages studied in the three
foregoing chapters and their respective urban centres, we do not see this
association as being the critical one in shaping the transformations we have
identified. They are associated – in the main – neither with peri-urbanisation,
nor with rural–urban linkages/interactions, but rather with links, networks
and associations that span other geographical scales. While we would like to
avoid simply relegating all this to ‘globalisation’, it is clear that the regional
and national contexts are not sufficient to adequately explain the transform-
ations under way. Just as we have suggested that the study villages are not
rural settlements, notwithstanding their location in a rural context, so we also
propose that the processes that are shaping these transformations are situated
in a range of settings stretching across the national space economy into the
international context.
The nature of the transformations under way in the settlements studied in
Thailand and Vietnam highlights the need for development interventions to
be carefully targeted. While handicraft production can boost rural incomes,
it is not without social and environmental costs. The nature of the handi-
craft being produced will partially determine whether farm and nonfarm
activities will continue side by side or whether the nonfarm activities will
come to dominate livelihoods. The latter is not necessarily negative but must
be considered as a possible outcome of the processes that have been outlined
in the three chapters. More than anything, what the chapters demonstrate is
the need to be aware of the entry points that we privilege when it comes to
viewing, interpreting, analysing and assessing the processes under way.
Specifically, we can view rural development through the lens of an indi-
vidual’s career, a household’s livelihood, a settlement’s sustainability, an
industry or sector’s vitality, or through the lens of regional development.
The position we adopt – essentially, the viewpoint we take – often provides a
very different assessment of the processes under way and the values that we
assign to them.
 Handicraft and manufacturing frontier 157
References
Bramall, C. (2007) The industrialization of rural China, Oxford: Oxford University
Press.
Briones, R. M. (2006) ‘Employment generation for the rural poor in Asia: Perspectives,
patterns, and policies’, Asian Development Review, 23(1): 87–116.
Evans, H. E. (1992) ‘A virtuous circle model of rural–urban development: Evidence
from a Kenyan small town and its hinterland’, Journal of Development Studies,
28(4): 640–67.
Evans, H. E. and Ngau, P. (1991) ‘Rural–urban relations, household income diversifi-
cation, and agricultural productivity’, Development and Change, 22: 519–45.
Grabowsky, R. (1995) ‘Commercialization, nonagricultural production, agricultural
innovation, and economic development’, The Journal of Developing Areas,
30: 41–62.
Haggblade, S., Hazell, P. and Brown, J. (1989) ‘Farm–nonfarm linkages in rural
sub-Saharan Africa’, World Development, 17(8): 1173–201.
Hart, G. (1996) ‘The agrarian question and industrial dispersal in South Africa:
Agro-industrial linkages through Asian lenses’, Journal of Peasant Studies,
23(2–3): 245–77.
Shen, Xiaoping and Ma, L.J.C. (2005) ‘Privatization of rural industry and de facto
urbanization from below in southern Jiangsu, China’, Geoforum, 36: 761–77.
Tacoli, C. (2008) Links between rural and urban development in Africa and Asia,
UN/POP/EGM-URB/2008/09, New York.
Thompson, E C. (2004) ‘Rural villages as socially urban spaces in Malaysia’, Urban
Studies, 41(12): 2357–76.
Thompson, E C. (2007) Unsettling absences: Place, migration, and urbanism in rural
Malaysia, Singapore: NUS Press.
Timmer, C. P. (2007) ‘How Indonesia connected the poor to rapid economic growth’,
in Timothy Besley and Louise Cord (eds), Delivering on the promise of pro-poor
growth: Insights and lessons from country experiences, Washington, DC: World
Bank and Basingstoke, UK: Palgrave Macmillan, pp. 29–57, available to download
from: http://siteresources.worldbank.org/INTPGI/Resources/PPG_eBook.pdf,
accessed 15 December 2008.
World Bank (2009) World development report 2009: Reshaping economic geography,
Washington, DC: World Bank.
12 Rural–urban transitions in
Tanzania’s northwest
mining frontier
Deborah Fahy Bryceson and
Rosemarie Mwaipopo 1

Introduction
The recent discovery of minerals and precious stones, particularly gold
and diamonds, has catalysed extraordinary growth spurts in previously
unpopulated or sparsely populated rural areas of Tanzania. Small-scale
miners have migrated in large numbers to the mineral strike sites to try their
luck. Their early makeshift settlements have expanded into jumbled concen-
trations of rural and urban modes of livelihood, metamorphosing from
remote ‘deep rural’ to pulsating urban centres of goods and service provision-
ing. In many cases, these population growth spurts, however dramatic, have
been short-lived and have dissipated as the extractable ores become less
accessible to the small-scale miners. This inevitably leads to the question of
how sustainable mining-led urbanisation is over time.
Tanzania, which has for decades been seen as an impoverished agrarian
country, is metamorphosing into a significant diamond and gold producer on
the African continent over the last two decades. Small-scale miners have
played a central role in this transformation (Chachage, 1995; Mwaipopo
et al., 2004; Fisher, 2007). This chapter traces mining as a ‘frontier livelihood’
and ‘mining-led urbanisation’ in two northwestern Tanzanian settlements:
Maganzo, a long-established diamond-mining settlement in Shinyanga
region and Nyarugusu, a gold-mining site in Mwanza region which experi-
enced its first significant rush in the 1980s. Our interest focuses on the unfold-
ing phases of livelihood and settlement transformation following the original
population surge connected with mineral discovery.
The first section of this chapter briefly reviews Tanzania’s post-inde-
pendence urbanisation trends prior to the influence of mining, before focus-
ing on the background history of mining and the character of the study sites.
After outlining the growth and migration patterns of the two settlements, the
agricultural foundations of the settlements, the demographic and economic
trends associated with mine-led urbanisation and the sustainability of urban
settlements subject to mineral booms and busts are considered in turn.
 Rural–urban transitions in Tanzania 159
Tanzanian urbanisation trends
Colonial and post-colonial government policies during the twentieth century
officially discouraged urban growth for the most part. The rigid colonial
racial three-tiered pyramid, in which Africans were farmers, Asians were
traders and Europeans were administrators was dismantled after independ-
ence. Thereafter the country’s formal employment structure expanded to
absorb educated and skilled Africans as well as many uneducated migrants.
Most of these jobs were concentrated in the national capital, Dar es Salaam.
In 1971, the Tanzanian government embarked on a decentralised govern-
ment administration system, which entailed sending government staff ‘to the
regions’, twenty in total. The transfer of trained staff and government fund-
ing for the expansion of regional service infrastructure headquartered in
the small up-country towns designated as regional capitals triggered rapid
secondary town growth which exceeded Dar es Salaam’s rate of expansion.2
During the economic crisis-ridden years of the 1980s, these towns continued
to grow twice as fast as Dar es Salaam (Potts, 2006). Their intermediate size
between rural village and capital city offered the advantage of combining both
rural and urban lifestyles (Bryceson, 1993).
Meanwhile, mining-led settlements have sporadically appeared since the
colonial period when artisanal and small-scale miners converged in the several
mineral-rich parts of the country, such as Lupa in Chunya, despite mining
being seen as an illegal activity. This trend continued after independence,
even though small-scale mining was still prohibited by law. As the country’s
economic liberalisation policies unfolded, the controls were eased. By the mid
1990s, mineral fever was in the air. Villagers adjacent to valuable mining
deposits, who had engaged in mining for some time, were joined by sizeable
numbers of migrants eager to get rich. Rising mobile phone usage in the new
millennium accelerated the rate and numbers rushing to newly discovered
mineral-bearing sites. They arrived within days rather than weeks or months
after the mineral’s discovery. Small villages of a few thousand people could
quickly balloon towards settlements of over 20,000 people. The northwestern
part of the country, notably Mwanza and Shinyanga regions, led the nation
in population movement of this nature.

Tanzanian mining and settlements of rapid population

growth in the northwest mining frontier
In this study, we deliberately chose urbanising mining sites that are ‘middle-
aged’ with respect to the timing of their migration rush in order to trace the
course of the settlements as their mining potential declines. The following
outlines their origins and population growth.
160 Deborah Fahy Bryceson and Rosemarie Mwaipopo
Nyarugusu gold site
The history of Nyarugusu’s gold wealth began when a pastoralist herding
his cows saw gold flecks attached to the roots of a fallen tree uprooted by a
strong wind. Mining was prohibited under President Nyerere’s agrarian
socialist policies. Nonetheless, some village men surreptitiously mined on a
limited scale. During the economic crisis conditions that followed Tanzania’s
1979 war with Idi Amin’s troops in nearby Uganda, increasing numbers
of artisanal miners began drifting towards Nyarugusu. The area was still
relatively forested, and the migrant miners lived in a village, some three
kilometres from the mining site where housing was available. As more miners
gathered, relations with the villagers became tense. Local residents accused
the miners of stealing food crops from their fields and endeavoured to expel
them. Pushed closer towards their mining sites in Nyarugusu, the miners
constructed temporary shelters in what is now Nyarugusu’s core settlement
area. Thereafter in 1983, the Tanzanian government became increasingly
lenient towards small-scale mining.3 Word spread about Nyarugusu’s gold
and young men flocked to the site from all over Tanzania and neighbouring
Kenya. Pits were only three metres deep at the time and gold was relatively
plentiful and pure. For security purposes, a local militia group (sungusungu)
based on traditional age-grade segmentation was established with elected
leaders.
By the mid 2000s, the number of claim owners was over 100. There is
a division of labour in gold mining in which claim owners usually do not get
directly involved in production. A wide range of capital and labour arrange-
ments entwine the interaction between ‘claim owners’ who have obtained
Primary Mining Licences (PMLs)4 and take roughly 30 per cent of the mined
ore, the ‘pit holders’ who lease the site from the claim owner and organise
production taking about 40 per cent of the mined ore after paying for the
production costs like timber, explosives, food and medicine, and the manual
labourers who dig the gold and receive the remaining 30 per cent of the mined
ore to divide amongst themselves.
Gold mining is generally considered the work of young men (kazi ya vijana
wa kiume) because it requires the exertion of a great deal of physical strength.
By the early to mid 1990s Nyarugusu’s surface deposits were gone. Miners
had to build shafts with wooden supports increasingly more deeply (40–60
metres).5 It is at this stage that the small-scale artisanal miners began facing
considerable technological difficulties. Ordinary agricultural irrigation
pumps are not up to the task of pumping out the water that is found at these
depths. The necessary equipment for deep-shaft mining is beyond the finan-
cial means of most small-scale miners, preventing further safe exploitation of
the mine.
During the past decade, Nyarugusu’s dwindling accessible gold deposits
prompted many miners to move on to gold sites in neighbouring regions
and beyond. Small-scale mining, however, continues at Nyarugusu and
 Rural–urban transitions in Tanzania 161
occasionally there are localised gold strikes, which reward the efforts of shaft-
digging miners (Illustration 12.1). Nyarugusu’s present population is esti-
mated by the settlement leadership to number 3625 households comprising
about 20,000 people. There is no way of knowing how many people were
resident in Nyarugusu at its peak in the early 1990s. Informants claimed that
Nyarugusu’s population rocketed to 35,000 whereas the 1988 population cen-
sus registers only 5968. Mwaipopo et al. (2004) estimated the current popula-
tion of Nyarugusu ward at 27,000 people. Given the magnetic force of gold,
the ethnic composition of Nyarugusu’s population is now very diverse, repre-
senting over 25 different ethnic groups. Throughout Nyarugusu’s gold fever, a
large segment of the population has farmed, producing staple foodstuffs for
local consumption.

Maganzo: small-scale diamond prospecting in the shadow of a

large-scale mine
Maganzo6 began as a camp for people looking for employment in the large-
scale foreign-owned7 Williamson diamond mine at Mwadui which started
operation in 1940. For two decades people were forced to shift their settle-
ment here and there around the perimeter of the large mine because of the

Illustration 12.1 Gold production in Nyarugusu.

Photo by Deborah Fahy Bryceson
162 Deborah Fahy Bryceson and Rosemarie Mwaipopo

Figure 12.1 Map showing location of study settlements in northwestern Tanzania.

suspicion that their proximity to the mine facilitated diamond smuggling.

Eventually in the late 1950s, a permanent settlement at Maganzo was estab-
lished under the leadership of the local chief. In effect, Maganzo functioned
as a labour market for those seeking work in the Williamson diamond mine
and as a service centre for the African mine employees.
During the national villagisation campaign of 1974, Maganzo residents
faced renewed pressures to move from their settlement. Rumours about
diamond smuggling abounded. Blame was focused on ‘wabeshi’, stealthy
intruders linked to the disappearance of Williamson’s sand tailings.8 These
charges were used as justification for evicting people from their homes. Resi-
dents sent an appeal to President Nyerere who came to their rescue and
ordered that Maganzo could remain intact as a settlement.
Maganzo continues to exist in the shadow of Mwadui mine. At the time
of the last census in 2002, Maganzo had 2256 households and a population
of 12,180 people, which is likely to have been a gross undercount in view of
the miners’ attempt to maintain a low profile given the suspicion with which
they are regarded by the authorities. In fact, Maganzo is the more outwardly
urban of the two settlements. There is a central market surrounded by a
cluster of shops. In 2005 the trunk road to Shinyanga town bordering the
settlement was completed. A petrol station is strategically sited on the main
road. Furthermore, Maganzo is on the national electricity grid, enabling
 Rural–urban transitions in Tanzania 163
wealthier residents to have access to relatively cheap electricity. The settle-
ment’s core has four well-laid out streets and several long-established houses.
There are clear signs of wealth differentiation reflected in the housing.
Located in a semi-arid area, the acute shortage of water in Maganzo causes
residents to harbour resentment against Mwadui mine’s piped water supply,
which they have been requesting permission to tap into without success.
Previously, Maganzo’s commercial role in the shadow of Mwadui attracted
some Tanzanian Arab settlement. With the government’s privatisation pol-
icies from 2000 onwards, some of the land that residents had been using for
diamond mining was allocated to large-scale Tanzanian commercial mining
interests, significantly reducing the area in which local people could mine.
Residents of the settlement have diversified occupations spanning trade,
diamond mining and agriculture. The latter two now co-exist incompatibly
amidst a growing shortage of land. People are torn between using their land
for farming to provision their household’s basic food needs as opposed to
digging for diamonds on their land.
Small-scale diamond mining entails digging a pit about two to three metres
in diameter and from one to two metres deep during the dry season. The
potentially diamond-bearing sand from the hole is placed to one side of the
hole until the rainy season comes and the hole fills with water (Illustration
12.2). The miner can then ‘wash the sand’ with a sieve pan which sifts the
worthless grains of sand away from the diamonds that are caught in the
sieve’s mesh. Not only does diamond mining interfere with the time a miner

Illustration 12.2 Diamond production in Manganzo.

Photo by Deborah Fahy Bryceson
164 Deborah Fahy Bryceson and Rosemarie Mwaipopo
might otherwise be engaged in planting, cultivating and weeding crops, but
also the pit digging destroys the fertile topsoil needed for agriculture. In many
areas, people have dug holes only two to three metres apart. This creates a
landscape with a peculiarly barren moon surface appearance that requires
careful circumnavigation to traverse the area. In this way Maganzo has
increasingly lost its agricultural potential, forcing residents to become pro-
gressively more reliant on food purchase.

Migrant households and livelihood change

In both surveyed settlements, the overwhelming majority of household heads
were migrants. Table 12.1 summarises the demographic characteristics of
our survey households showing that 26 per cent of Maganzo’s household
heads (HHH) were native-born whereas in Nyarugusu the number was only
7 per cent. Figure 12.2 shows the year of arrival of household heads in the
settlements.
The motivation for moving was primarily economic. Mining in Nyarugusu
(46 per cent) figured prominently but the ‘search for better opportunities’ held

Table 12.1 Comparative household demographic statistics by settlement

Household Characteristics Nyarugusu Maganzo

Number of households sampled 73 72

Average age of HHH 45 51
Mean year of migrant household head’s move to 1982 1983
settlement
Migrant households (%) 93 74
Female-headed households (%) 5.5 12.5
Household size 7.1 6.7
Household sex ratio 114 113

Source: Tanzania mining, Author’s household survey, 2005.

Figure 12.2 Migrant household heads’ year of arrival.

Source: Authors’ household survey, 2005.
 Rural–urban transitions in Tanzania 165
sway in Maganzo (34 per cent). Significantly mining ranked only fourth in
Maganzo preceded by trade. In this locality, the pursuit of diamond mining
does not have the same stampede pattern as gold strikes.
The average household size was 6.9 members, which is above the national
and regional averages (Tanzania, 2002). A surplus of males was reported
as numerous sons compared with daughters, especially in Maganzo. While
it is not impossible that households attracted to mining are fortuitously
better-endowed with male offspring, it is more likely that survey respondents
reporting household composition tended to exaggerate familial ties of
dependency with the term ‘sons’ for young males in the family. Traditionally,
as well as during the socialist period of the country’s development, the hiring
of wage labour by households was not sanctioned, so the unusual number of
sons is likely to be masking types of tied labour relationships other than that
of offspring.
Migration to Maganzo has been fairly steady. The presence of diamonds in
the area has been known since the 1940s. Hence, diamond mining has been
a livelihood activity there for a long time. Located in a semi-arid area, settle-
ment in Maganzo for agricultural purposes has never been very attractive.
Rather the trade and service nature of the settlement in the shadow of
Williamson mine has been significant. Maganzo represents the intermediate
case of mixed trade and mining that contrasts with the extreme of Nyarugusu’s
gold rush character. Furthermore, the data indicate that migration to
Maganzo has levelled off during the 1990s as the shortage of land impinges
on people’s way of life. Certainly this was confirmed by open-ended discus-
sions with Maganzo residents who expressed the view that a deepening land
constraint was worsening their livelihood prospects.

Proliferating divisions of labour catalysed by mining activities

Studies of livelihood in rural Africa to date have stressed the highly fluid and
diversifying nature of work over the past two decades (Ellis, 2000; Bryceson
and Bank, 2001). Processes of de-agrarianisation and de-peasantisation
underline these tendencies (Bryceson, 2002a; 2000b). Tanzanian rural liveli-
hood diversification has been widely documented (Bagachwa, 1997; Madulu,
1998; Bryceson, 1999). However, during the process of livelihood change,
field researchers often encounter difficulties in gaining an accurate report of
its occupational composition. This is because rural informants’ definitions of
their work are intricately related to their cultural identity and way of life
rather than being simply income and subsistence-begetting activities.
In Tanzania, and especially in Sukumaland, rural people have traditionally
seen their agro-pastoralist activities as fundamental to their ethnicity as
Sukuma. Thus, answers to questions about one’s occupation have to be
treated with caution since the knee-jerk reaction of both rural men and
women is to state that they are farmers. Therefore, it is meaningful when
people report non-farming primary occupations. This indicates that not only
166 Deborah Fahy Bryceson and Rosemarie Mwaipopo
are they spending a large segment of their time in non-agricultural activities
but they are also not feeling socially constrained to admit it. Table 12.2 shows
that approximately 30 per cent of the surveyed Tanzanian heads of house-
holds identified as non-farmers. Nyarugusu had the lowest incidence of
non-farmers, most of whom were miners. Non-farming in Maganzo was
mainly split between miners and traders, whereas in Maganzo trading was
most prominent. Added together, we find that those who state farming as a
primary or secondary occupation account for 87 per cent of all household
heads: in other words, only 12 per cent of household heads report being
entirely distanced from farming.
Agriculture remains the most important activity accounting for over
50 per cent of reported activities in Nyarugusu. In Maganzo, declining land
availability restricts household heads’ reliance on agriculture to less than
40 per cent, although it is nonetheless highly valued for its contribution to
household food needs. Mining, however, is the strongest secondary activity in
both sites indicating that the gains that can be had from mining are well
appreciated. Figure 12.3 shows the weighting of household heads’ work activ-
ities in terms of time spent. One sees that the mining settlements’ occu-
pational structure is relatively restricted, although Maganzo has some trade
and skilled work given its more strategic location on a main road relative
to Nyarugusu.
The unpredictability of mining returns dictates that miners generally
actively engage in other livelihood activities, particularly agricultural produc-
tion, or rely on the cultivation of other family members be it their wives,
children or extended family members. Even migrant miners who would have
no prior claim on land try to devise ways and means of accessing land
through mine pit holders or local villagers in order to obtain a reliable daily
food supply regardless of the vagaries of their cash flow for food purchases.
In Nyarugusu, roughly 50 per cent of staple foods, usually maize, cassava and

Table 12.2 Stated primary and secondary occupations of household heads

(% total HHH)

Nyarugusu Maganzo Both

Occupation First Second First Second First Second

Farmer 76 16 64 17 70 17
Miner 17 41 9 34 13 38
Trader 1 14 10 10 6 12
Unskilled worker – – – 1 0 1
Skilled worker 6 6 16 6 11 6
Formal employee – – – – 0 0
None – 23 1 31 1 27
Total 100 100 100 100 100 100

Source: Authors’ household survey, August 2005.

 Rural–urban transitions in Tanzania 167

Figure 12.3 Household heads’ livelihood activities.

Source: Authors’ household survey, 2005.
Notes: Agriculture/Livestock – plant and animal husbandry and cattle-keeping
Mining – diggers using their manual labour to extract and process minerals and mine pit holders.
Trade – small-scale traders dealing in commodities of an agricultural, mining or general nature
or traders with premises.
Unskilled worker – self-employed involved in trades requiring no special training including
security guards, brickmakers, masons, water carriers, etc.
Skilled worker – self-employed engaged in activities that require some instruction or education,
notably drivers, mechanics, carpenters, plumbers, medics, tailors, etc.

to a lesser extent rice, are supplied by own farm production. Fruit and espe-
cially meat are mainly purchased. Maganzo, on the other hand, provisions
less than a quarter of its staple food needs. As an old mining settlement
hemmed in by large mining claims, almost half the residents are landless
forcing them to become highly dependent on purchased food (Figure 12.4).
Many have decided to dig up their farmland to search for diamonds.
Environmentally, valuable topsoil is destroyed, undermining the productive
capacity of the land.

In situ urban transformation

Until recently, rural migrants moved to urban areas to gain the perceived
advantages of urban livelihoods and lifestyles. Our case study sites represent
settlements where rural residents have experienced in situ urban transform-
ation. The frontier livelihood pursuits of mining and trade have triggered
extremely rapid population growth and a transitional process in which both
original and migrant residents have recourse to rural and urban livelihood
activities as well as a spectrum of rural to urban styles of living and modes of
168 Deborah Fahy Bryceson and Rosemarie Mwaipopo

Figure 12.4 Size of household land holdings (acres).

Source: Authors’ household survey, 2005.

consumption. As centres of rapid in-migration, these settlements have served

as ‘destinations of desire’. The vast majority of migrants anticipated
enhancement of their material lives through migration to these localities and
in the process their locational identity altered increasingly towards an urban
outlook.
When asked in the survey whether they lived in a rural or urban settlement,
65 per cent saw themselves as urban dwellers. Interestingly Nyarugusu had the
highest per centage (84 per cent). This may be because there is a strong desire
on the part of the leadership and population to keep the momentum of urban
development going despite the out-migration of gold miners. They are lobby-
ing heavily for Nyarugusu’s role as a service and business sector and are
demanding official urban status and a good road. In Maganzo, 65 per cent of
the population saw themselves as urban which is not surprising given its long
standing as a centre of in-migration and its urban infrastructure and wealth
differentiation.
In East Africa, it has been widely assumed that early stage urbanis-
ation is primarily characterised by male migration. This was no doubt the
case in the first waves of migration to Nyarugusu and Maganzo. However,
after two decades both of these settlements’ sex ratios became more
balanced.
The 15–34 years of age cohort comprises those who are not only the most
economically active but also the most likely to migrate and engage in new
‘frontier’ livelihoods, notably mining and trading. They are, in effect, the
dynamo age group which reflects the economic viability or otherwise of the
settlement. Quite different patterns emerge in our two settlements with
respect to this age group (Figure 12.5). Nyarugusu has had higher propor-
tions of its population in the 15–34 age group: 34 per cent in 1978, 40 per cent
in 1988 but then as the mining fortunes of Nyarugusu have declined this age
 Rural–urban transitions in Tanzania 169

Figure 12.5 Percentage of total population aged 15–34 in 1978, 1988 and 2002.
Source: Tanzania Population Census, 1978, 1988 and 2002.
170 Deborah Fahy Bryceson and Rosemarie Mwaipopo
group has contracted to only 27 per cent. In all three periods, males have
superseded females, implying that mining remains the central livelihood
strategy for the settlement.
Maganzo has had a more peculiar pattern, displaying a high degree of
volatility and a reversal of gender dominance over time. In 1978 the per cent-
age of population in the 15–34 cohort was 37, plunging to 24 in 1988 and
rising to 54 in 2002. In 1978 women superseded men, which could tally with
the settlement acting as a recreation and service sector in the shadow of the
Mwadui mines. In 1988, the male and female population ratio of the 15–34
age group was balanced but had declined as a per centage of the total popula-
tion whereas in 2002, females and especially males in this age group had risen
as a proportion of the settlement’s total population.
Although both settlements are expanding in size, only Nyarugusu is edging
towards a balanced gender ratio in the dynamo age group. In the process,
Nyarugusu is ‘ageing’. Maganzo achieved a gender balance in 1988 as its
dynamo generation shrank in size, then developments took a surprising twist.
The dynamo age group rapidly expanded with a heavy male bias. Based on
our survey results and interviews, this can be explained as a result of distress
rather than urban take-off. Maganzo experienced increasingly acute land and
water constraints, which led it to take on more of the character of a diamond-
mining work camp rather than an urban residential location. Mining began
encroaching on residents’ land allocation for subsistence agriculture and
housing. Male migrants were being recruited into residential households
signalling the intensification of mining and a proportional decline in the
female and child population normally expected in the composition of rural
households. The large number of landless households in our survey cannot be
interpreted as a sign of proletarianisation along the classic Western model of
industrialisation and urbanisation. Instead it indicates that Maganzo’s
depleting resource base, notably its lack of water, is unlikely to allow it to
follow an urban growth trajectory for much longer, in the absence of piped
water supplies from De Beers or Shinyanga town.

Urban momentum of frontier settlement growth:

spark versus sustainability
How durable are Nyarugusu and Maganzo’s urban economies as small-scale
mining reaches technological investment barriers? The answer to this ques-
tion is difficult given the enormous volatility of population movement to and
from the settlements and the lack of reliable inter-censal population data. We
do not know what the two settlements’ peak populations were. It is clear that
both settlements are still growing at rates that supersede the Tanzanian
national average,9 even though residents report that the settlements are not as
vibrant as before, and, in the case of Nyarugusu, heavy male out-migration to
other ‘fresher’ mining settlements has become a salient tendency over the past
decade. Nonetheless, both Maganzo and Nyarugusu continue to be centres of
 Rural–urban transitions in Tanzania 171
net in-migration. In view of the dwindling returns to mining, there are clearly
other factors attracting migrants.
The available evidence suggests that, in addition to better health and edu-
cation services compared with rural villages, both of these settlements offer
their populations other livelihood opportunities, notably trade in Maganzo
and farming in Nyarugusu. The farming prospects of Nyarugusu might be
seen as negating rather than maintaining an urban mode of existence, but this
is debatable. Farming in Nyarugusu not only offers a vital subsistence fall-
back and cash income from crop sale, but it also represents speculative land
investment. Large-scale international investors are interested in the gold
deposits that may lie deeper than those currently being worked by small-scale
miners. Many Nyarugusu households stand to gain by staying put rather than
pulling up stakes and moving on to new gold rush settlements, exercising
shrewdness about property values. In so doing, they are exhibiting an urban
commercial outlook rather than tribal attitudes towards land linked to ethnic
identity and community solidarity. This would suggest that Nyarugusu resi-
dents are more urban than rural in their outlook. Generally, people in mining
areas of the country are increasingly applying for claims in areas where there
may be gold, even though they do not have funds to mine the area. They are
speculatively waiting for a mining company to offer to buy their claim (or for
pit holders to start mining), knowing that claim owners can sell their mining
titles to large-scale commercial mining companies, whereas the legally stipu-
lated compensation to farmers with land titles is very limited.10
Maganzo has the smallest population but is visually more urban with its
prominent permanent housing, superior physical infrastructure, broad array
of livelihood activities and residents’ heavier reliance on purchased food and
petrol, electricity, kerosene and charcoal as opposed to firewood. However,
Maganzo has reached a critical population density vis-à-vis its land and water
resources. It now experiences acute shortages of both which impinge on the
livelihood viability of its residents, particularly that of the poor. Residents’
reliance on purchased food is regretted rather than boasted about.
Even more than Nyarugusu, the continuation of Maganzo’s urbanisation
process depends on positive external stimuli, namely the settlement’s func-
tional relationship to Williamson mine, and its access to services such as food
supplies and water. Currently Maganzo lies along a water pipeline con-
structed from Mwanza to Shinyanga town, which is projected to serve the
adjacent Williamson mine compound and Mwadui township. However, even
if it gains access to this piped water, Maganzo still faces a shortage of land for
its mining and agricultural activities. The diamond pit digging that has
already taken place is very extensive and destructive to the topsoil. Some
households are experiencing a livelihood crisis. Maganzo’s female and child
population is shrinking relative to the male adult population, as evidenced in
the changing age–sex distribution of Maganzo’s inter-censal population.
Family units are finding it difficult to fulfil their basic needs for food.
The livelihood viability of the settlement revolves around its commercial
172 Deborah Fahy Bryceson and Rosemarie Mwaipopo
trade potential if and when artisanal diamond mining in Maganzo becomes
untenable. Its location on a major highway and proximity to Mwadui mines
and Shinyanga town 16 km away should stand it in good stead in terms of
being close and readily accessible to a large consumer population. Shinyanga
has a far larger urban population with superior services and shops, which
may compromise Maganzo’s commercial potential. Thus, although out-
wardly Maganzo is more urban in appearance than Nyarugusu, its urban
foundations are similarly fragile.

Conclusion
Where does rapid mine-led urbanisation lead? Clearly small-scale mine-led
urbanisation is not an urban developmental path that can be replicated
everywhere nor does it have a very long-term trajectory where it does occur.
Over time the issue of the exhaustible supply of this catalysing resource
arises. Its non-renewable nature causes slowdowns or even contraction of
the resident population, but the process of urbanisation that it has set in
train can and often does continue. Small-scale mining populations have an
erratic boom-and-bust character. They, however, catalyse internal growth
processes based on service provisioning and trade and generally maintain the
settlement’s agrarian activities as long as they can. Small-scale mining gener-
ates a variant urbanisation tendency – reliance on immediately accessible
mineral wealth that is extracted, processed and exchanged for means of
subsistence.
In summary, there are three main tendencies in mine-led urbanisation that
are decisive for a settlement’s long-term urbanisation. First, small-scale min-
ing is generally dependent on local farming of staple food supplies which
nonetheless over the medium- and longer-term mining competes with agri-
culture for land, eroding the rural agrarian base and initiating urban pur-
chased consumption patterns as demonstrated in Maganzo. Second, mining
settlements generate multiplier effects in the economy through their purchas-
ing power. Commercial trade and recreational services like bars and prostitu-
tion bolster local employment. Third, the population expansion related to
mining catalyses public service provisioning of schools, healthcare, post
offices, churches, police, etc. Such services, which are generally superior to
those in the surrounding countryside, attract migrants, even if mining activ-
ities are declining. By contrast, recreational services may serve as population
deterrents. Prostitution and high levels of drinking are associated with a
higher incidence of HIV/AIDS or household instability that can subtract
from rather than add to population totals.
The gold- and diamond-mining rushes of the past two decades mark a
profound economic as well as cultural change for Tanzania’s northwest
region. Labour absorption is the motor force of livelihood diversification and
urbanisation. Despite its capricious nature, small-scale mining can ultimately
lead towards sustainable urban settlement but time will tell if Nyarugusu and
 Rural–urban transitions in Tanzania 173
Maganzo specifically can get beyond the temporal delimitations of small-
scale frontier mining.

Notes
1 We wish to pay tribute to the intellectual contribution and inspiration of
Professor Ndalawa Faustin Madulu. He coordinated the fieldwork upon which
this chapter is based and was involved in the data analysis up to his untimely death
in May 2007. We are grateful to Jesper Bosse Jønsson for his criticisms and advice
on an earlier draft of this chapter.
2 It should be noted that initial urban growth rates tend to be exceptionally high
because they are starting from a low population base. See Potts (2006) for a discus-
sion of this effect.
3 Chachage (2006) notes that the most significant policy change paving the way for
small-scale mining appeared in the Mining Act of 1979 which vested ownership of
all mineral rights in the state but gave local miners the scope to peg claims without
direct state involvement. This was followed by the Small Scale Mining Policy Paper
of 1983, which encouraged Tanzanians to engage in small-scale mining alongside
other income-earning activities. Surprisingly, this policy preceded government
economic liberalisation measures by several years. Chachage interprets it as the
government’s timely reaction to the rising price of gold in the world market.
4 Under the Mining Act of 1998 (United Republic of Tanzania, Ministry of
Energy and Minerals), a PML gives the claim owner the right to mine an area for
five years, which can be renewed or transferred to another named holder.
5 The small-scale mining process involves the blasting of the underground rock,
which is then hauled up and first crushed by hand and then machine-crushed.
Once the rocks are pulverised, the sand is washed and sieved to separate the gold
particles from the sand. The tailings are piled to one side to be sieved again at a
later date. Mercury is used to amalgamate the gold dust.
6 Located in the Kishapu district of Shinyanga region.
7 The Tanzanian government now has a 25 per cent share and De Beers 75 per cent.
8 Sand tailings are waste soil and sand remaining after panning or other forms of
diamond extraction. They have potential value given that some diamonds may
have evaded extraction.
9 The national annual inter-censal growth was 2.9 per cent with high regional rates:
3.3 per cent in Shinyanga and 3.2 per cent in Mwanza.
10 Personal communication from Jesper Bosse Jønsson, April 2007.

References
Bagachwa, M.S.D. (1997) ‘The rural informal sector in Tanzania’, in Bryceson, D.F.
and V. Jamal (eds), Farewell to farms: De-agrarianisation and employment in Africa,
Aldershot, UK: Ashgate Publishing Ltd: 137–54.
Bryceson, D.F. (1993) Liberalizing Tanzania’s food trade: Public and private faces of
urban marketing policy 1939–1988, London: James Currey.
Bryceson, D.F. (1999) ‘African rural labour, income diversification and livelihood
approaches: A long-term development perspective’, Review of African Political
Economy, 80: 171–89.
Bryceson, D.F. (2002a) ‘Multiplex livelihoods in rural Africa: Recasting the terms
and conditions of gainful employment’, Journal of Modern African Studies, 40(1):
1–28.
174 Deborah Fahy Bryceson and Rosemarie Mwaipopo
Bryceson, D.F. (2002b) ‘The scramble in Africa: Reorienting rural livelihoods’, World
Development, 30 (5): 725–39.
Bryceson, D.F. and L. Bank (2001) ‘End of an era: The development policy parallax’,
in Bank, L. and D.F. Bryceson (eds), Livelihoods, Linkages and Policy Paradoxes,
Special Issue of the Journal of Contemporary African Studies, 19(1): 1–19.
Chachage, C.S.L. (1995) ‘The meek shall inherit the earth but not the mining rights:
The mining industry and accumulation in Tanzania’, in Gibbon, P. (ed.), Liberal-
ised development in Tanzania, Uppsala: Nordiska Afrikainstitutet, pp. 37–108.
Chachage, C.S.L. (2006) ‘Can Africa’s poor inherit the Earth and all its mineral
rights?’, Dakar: Codesria, available online at www.codesria.org/Links/conferences/
general_assembly11/papers/chachage.pdf.
Ellis, F. (2000) Rural livelihoods and diversity in developing countries, Oxford: Oxford
University Press.
Fisher, E. (2007) ‘Occupying the margins: Labour integration and social exclusion in
artisanal mining in Tanzania’, Development and Change, 38 (4): 735–60.
Madulu, N.F. (1998) Changing lifestyles in farming societies of Sukumaland: Kwimba
District, Tanzania, Institute of Resource Assessment & Leiden. African Studies
Centre Joint Working Paper vol. 27.
Mwaipopo, R., W. Mutagwaba, D. Nyange and E. Fisher (2004) ‘Increasing the con-
tribution of artisanal and small-scale mining to poverty reduction in Tanzania:
Based on an analysis of mining livelihoods in Misungwi and Geita Districts,
Mwanza Region’, Report prepared for the Department for International Devel-
opment (UK).
Potts, D. (2006) ‘Urban growth and urban economies in Eastern and Southern Africa:
Trends and prospects’, in Bryceson, D.F. and D. Potts (eds), African urban
economies, London: Palgrave Macmillan, pp. 67–104.
Tanzania, United Republic of (1982) 1978 Population Census, Dar es Salaam:
Population Division, Bureau of Statistics.
Tanzania, United Republic of (1991) 1988 Population Census, Dar es Salaam: National
Bureau of Statistics.
Tanzania, United Republic of (2002) Population Census 2002, Dar es Salaam:
Government Printers.
Tanzania, United Republic of (2003) Tanzania Population Census 2002, Dar es
Salaam: Government Printers.
13 The dynamics of the gold
mining industry and its effects
on settlements and livelihoods
in Wassa West District, Ghana
Paul W.K. Yankson

Introduction
According to conventional wisdom, countries that possess rich mineral
deposits are fortunate. Such deposits are assets which form part of a country’s
natural capital. Mining is the key that converts dormant mineral wealth into
various forms of capital that directly contribute to a nation’s economic
development. Despite the intuitive appeal of conventional wisdom, a new
view of mining has emerged over the past two decades that questions the
positive relationship between mineral extraction and economic development
(Davis and Tilton, 2002). The issue, therefore, is how to ensure that mining
contributes as far as possible to economic development and poverty reduc-
tion not only for a country as a whole, but also for the localities which are
directly affected by the mining. The negative socio-economic and environ-
mental effects of mining, particularly surface gold mining, which can coun-
teract much of the positive effects of mining, need to be considered (Sarin
et al., 2006).
The mineral sector has played and continues to play an important role in
mineral-rich countries worldwide. Ghana currently is among a few mineral-
rich African countries with the most attractive geological and investment
environments (Akabzaa and Darimani, 2001). Since the mid 1990s this has
attracted a number of large- and medium-scale gold-producing companies
which are based in the United States of America, Australia and South Africa
and are involved essentially in surface-mining operations in the country. A
number of small-scale nationally owned and registered mining companies
and a large number of gold-mining ‘gangs’ operating illegally are also practis-
ing surface mining in Wassa West District. This has led to loss of farm lands
and is partly responsible for the increased level of migration, especially by
young people, to the main urban centres in search of waged employment,
particularly in the mining industry.
This chapter examines the effects of the boom in gold mining on Wassa
West District in terms of settlement growth, economic activities, livelihoods
and related issues such as unemployment and poverty. Trends in the gold-
mining industry in Ghana are presented in the next section, followed by a
176 Paul W.K. Yankson
brief description of Wassa West District and the rationale for the choice of
study sites. The sections that follow examine the key issues that this chapter
addresses.

Trends in the mining industry

Ghana’s mining industry is dominated by gold production. After a long
period of decline in the minerals and mining sector, the country has witnessed
massive investment in the sector, particularly in gold mining. Since 1986 the
country has been able to attract over five billion US dollars of direct invest-
ment into mineral exploration and the establishment of new mines, as well as
the expansion and the rehabilitation of existing mines (Barning, n.d.). Output
and exports of mineral products have also increased considerably. The share
of gold in total export earnings from mineral exports stood at about
89 per cent in 1980 and rose to 96 per cent in 1998 (Twerefou et al., 2007). In
2000, mining was the major export earner, as it contributed about 5.5 per cent
to gross domestic product (GDP) and provided about 6 per cent of total
national employment (Aryee, 2002, quoted in Twerefou et al., 2007: 2). Many
of the gross export earnings for gold have come from the large-scale foreign
mining companies.
Wassa West District, which occupies about 10 per cent of the Western
Region, plays a key role in the production of gold in Ghana and of virtually
all the known minerals in the country with the exception of bauxite. All the
major companies in gold, manganese and diamond production are located in
this district. Of the 18 large-scale foreign mining companies operating in
Ghana, at least 7 are located in Wassa West District (Agbesinyale, 2007). The
district also has over 200 registered medium- and small-scale gold- and
diamond-mining companies. Wassa West District also harbours a very large
number of artisanal or unregistered small-scale mining groups, locally
referred to as galamsey. In terms of employment, it is estimated that the
large-scale gold-mining sector provides direct employment for approximately
15,000 workers, while about 300,000 people are estimated to be directly
engaged in small-scale mining (Hilson and Potter, 2005). There are, however,
no meaningful operational and other forms of linkages between the two
scales of gold-mining operations in Ghana (Larsen et al., 2009).
Gold mining has been carried out for several centuries in Ghana, long
before modern gold mining in Wassa West District (Dickson, 1969; Hilson,
2002). Modern gold mining started in the Tarkwa area in the 1880s (Dickson,
1969). There has been a boom in gold mining in the district since the
implementation of the Economic Recovery Programme and the Structural
Adjustment Programme in the early to mid 1980s. The rapid rise in explor-
ation and excavation activities by large-scale mining companies has resulted
in the displacement of thousands of the previously undisturbed subsistence
small-scale or artisanal gold miners (Akabzaa and Darimani, 2001; Hilson,
2001, 2004; Hilson and Potter, 2005) and increased the incidence of conflict
 Dynamics of the gold mining industry in Ghana 177
and of contests over land between the small- and large-scale mining oper-
ations (Aubynn, 2009). As a result, artisanal miners have had several confron-
tations with the state security agencies, and at several times in the past that
activity had been proscribed, throwing many of the illegal operators out of
employment, particularly in Wassa West District.

Study sites in Wassa West District

The Wassa West District case study focused on three settlements: Tarkwa, the
administrative capital of the district and the oldest centre of formal gold
mining, not only in the district but also in the country (see Figure 13.1); and
two of its surrounding communities namely, Teberebie and Adieye. These two
villages are in the mining concession of Anglogold-Ashanti, one of the
multinational mining companies operating in Wassa West District. These
villages are fairly close to each other and within the daily commuting zone of
Tarkwa. Adieye is an old village that received some households that had to be
resettled from other villages in the active zone of the company’s mining
operations. Teberebie, one the other hand, is a completely new settlement
created to settle communities displaced by active surface mining within the
company’s mining concession.
Tarkwa, as an old mining settlement, is undergoing rapid urbanisation,
partly as a result of the revival of the mining economy. On the other hand,
Adieye and Teberebie are rural settlements under the shadow of Tarkwa.
They represent communities whose livelihood patterns have changed as a
result of the relocation and loss of farm lands. Tarkwa’s development was
initially induced by gold mining. Other factors responsible for the rapid
growth of Tarkwa, up to the end of the nineteenth century, included Tarkwa’s
accessibility from the coast. Improvements to the many routes to Tarkwa
accentuated the value of Tarkwa’s nodal position and its role as a trading and
distribution centre for its wider region. In addition, it was made an adminis-
trative centre and hence performed an administrative function in addition to
its mining and commercial activities. The railway from the port of Sekondi,
which reached Tarkwa in 1901 (Dickson, 1969), enhanced the role Tarkwa
played in its region. The functions it performed induced rapid population
growth, due largely to the in-migration of people into its mining industry
(Huq, 1989: 154, quoted in Songsore et al., 1994: 27). For instance, by 1948
the population of Tarkwa stood at 7805. This increased to 13,545, 14,708 and
22,107 in 1960, 1970 and 1984 respectively. In 2000, without its adjoining
villages, which had grown to become part of Tarkwa, the town had a popula-
tion of 30,360. Together with its surrounding villages, the population of
Tarkwa in 2005 was estimated by the Planning Office of the Wassa West
District Assembly to be 53,323.
Adieye was founded in 1946 by a farmer from old Teberebie. Later on a
number of migrants, looking for land to cultivate cash crops, especially cocoa
and palm oil, were given land to farm around Adieye. Being a very old
178 Paul W.K. Yankson

Figure 13.1 Location of Teberebie gold mine in Wassa West District.

 Dynamics of the gold mining industry in Ghana 179
abandoned mining area, mining activities were restarted around there in
the mid 1980s by the Ghana-Australian Goldfields (GAG) company, which
operated surface mining in the area. In the course of Ghana-Australian
Goldfields’ operations the company had to resettle a community of 12
households, by name Iduaprim, at Adieye. Farming continues to be the main
activity at Adieye, and this has attracted a lot of migrants, involved in both
food and cash crops, to the area. In addition, a number of tenant households
who work with the various mining concerns in the area have settled in Adieye.
The population of Adieye was 1524 in 2004 (Opportunities Industrialisation
Centres International (OICI), 2004)).
Teberebie came into being in 1986. The resettlement was undertaken
originally by Teberebie Goldfields (TG) Ltd, an American mining company,
and part of the concession was acquired by Anglogold-Ashanti when TG
folded. Teberebie has grown in population and physical size in recent years.
Its’ population was 1137 in 2003 (OICI, 2004). Though seriously constrained
by shortages of farmland, it has attracted people, mainly workers of mining
companies and also workers from Tarkwa, as a result of the relatively low
rental units available in the village. People have built houses there mainly for
rental purposes.
The question remains, what have been the effects of the revival of gold
mining on the growth of Tarkwa, Adieye and Teberebie and on the
livelihoods of the people?

Settlement growth
The effects of the revival of gold mining in Wassa West District are more
marked on Tarkwa than on Adieye and Teberebie. A noticeable feature
of Tarkwa’s growth in recent years has been in its physical size. It has
experienced an appreciable level of physical expansion during the last two
decades, partly as a result of the revival of the formal gold-mining industry.
Unfortunately, for decades its physical development has not benefited very
much from physical planning. With the exception of a few new residential
settlements, which conform to the norms of physical planning, the majority
of the buildings do not adhere to planning rules, with the result that there has
been a serious amount of unplanned and unregulated development. A strik-
ing feature of the occupied housing units in Wassa West District as a whole is
overcrowding, as a larger proportion of households use small units with only
one sleeping room each. This is most extreme in Tarkwa (Wassa West District
Assembly, 2002). The continuous influx of migrants into the district, and into
Tarkwa in particular, in search of jobs has created a major housing problem
and placed a lot of strain on the urban infrastructure and service provision in
Tarkwa. Besides those forced to live in severely overcrowded conditions,
many young migrants have turned the railway station and lorry parks into
places of abode (Akabzaa and Darimani, 2001). The spill-over effects are felt
by Adieye and Teberebie, as described above. The Wassa West District
180 Paul W.K. Yankson
Assembly, the local government body, has not been able to respond
adequately to the challenges of environmental management, the provision of
social services and economic infrastructure to provide a well-functioning
urban economy.
The boom in mining has also stimulated development in the service and
commerce sectors in particular. New hotels have come into service in the last
ten to fifteen years. Related to the boom in gold mining has been an increased
level of social vices such as prostitution and drug abuse. Both mobile and
resident sex workers operate in Tarkwa. Drug abuse is especially prevalent
among operators of illegal small-scale mining operations, who use drugs
to fortify themselves for their risky and dangerous activities (Akabzaa and
Darimani, 2001).

Changing livelihood opportunities

The study ascertained the key activities that people were engaged in and their
changes over time, the assets they had available for organising their activities,
and the effects of the mining industry on their livelihood activities. Until its
decline in the 1970s and early 1980s, the mining industry could be considered
the growth industry in Tarkwa. Its revival since the late 1980s has had effects
on other sectors, such as commerce, construction, administration, social
services and transportation. Since the early 1990s, the mining industry has
brought considerable investments into Wassa West District generally and
Tarkwa and its surrounding areas in particular. However, the overall occu-
pational structure of Tarkwa has remained virtually unchanged from that in
the 1960s. An examination of the occupational structure of Tarkwa in 1960,
1970 and 2000 as provided in the population census reports shows that in
1960 mining and quarrying, though the most important growth industry,
accounted for only 12.3 per cent of total employed labour within Tarkwa
town. Women were virtually not represented at all in this sector of Tarkwa’s
formal economy. By far the most important activity in terms of numbers
employed was commerce, which engaged about 31 per cent of the working
population and was dominated by women. Commerce was followed by
services, with 19 per cent, dominated by men, and manufacturing with
14 per cent. The overall structure of the economy of Tarkwa remained largely
unchanged in 1970.
The dominance of males in mining and related activities, in construction,
transport, storage and communications, as well as in community, social and
personal services, did not change between 1960 and 1970. On the other hand,
the ratio of males to females in commerce increased during the period 1960–
1970 (1:6.2).1 The employment structure of Tarkwa in 2000 was also not very
different from the 1970 situation. Mining and quarrying employed about
14 per cent of the labour force, a slightly higher proportion than the situation
in 1970. This certainly is an indication of the revival of the mining economy
but employment generation from the large-scale mining companies has
 Dynamics of the gold mining industry in Ghana 181
been limited. Commerce is the single most important sector in terms of
employment.
The results of our baseline survey confirm what the census reports show,
namely that commerce was the main activity of households in terms of the
amount of time devoted to this activity and also the main source of income
for households in 2005 in Tarkwa, as shown in Tables 13.1 and 13.2.
In the case of Adieye and Teberebie, it was farming that was the single
most important activity of respondents in 1995 (17 per cent), but trading had
taken over as the most important single activity of respondents in 2005
(22 per cent). In terms of income, Table 13.3 shows that farming was overtaken

Table 13.1 Major economic activities of households in

Tarkwa in 1995 and 2005 (%)

Economic Activity 1995 2005

Trading 14 23
Skilled work 9 16
Farming 7 7
Mining 3 4
Formal employment 3 3
Unskilled work 4 3
Landlords – 4
Students 6 8
Others 3 –
None 52 32
Sample size 832 832

Source: Author’s household survey of Tarkwa, 2005

Table 13.2 Proportion of income from household

economic activities generating most income in Tarkwa in
1995 and 2005 (%)

Economic Activity 1995 2005

Trading 29 35
Skilled work 17 21
Farming 15 14
Mining 10 12
Unskilled work 8 8
Formal employment 2 3
Landlords 6 6
Others 2 1
None 11 1
Sample size 282 282

Source: Sub-sample of the Author’s household survey of

Tarkwa, 2005
182 Paul W.K. Yankson

Table 13.3 Proportion of income from household

economic activities generating most income in Teberebie
and Adieye in 1995 and 2005 (%)

Economic Activity 1995 2005

Trading 9 9
Skilled work 9 9
Farming 33 31
Mining 27 38
Unskilled work – –
Formal employment – 2
Landlords – –
Others 2 11
None 20 –
Total 55 55

Source: Sub-sample of the Author’s household survey of

Teberebie and Adieye, 2005.

by mining as the activity that generated the most income for the households
of respondents between 1995 and 2005. This came from wages and salaries
not only from large-scale mining, but also from small-scale mining.
Overall, self-employment is the dominant form of employment in Tarkwa,
Teberebie and Adieye, as the commerce/retail, manufacturing and service
sectors employ the largest proportion of the labour force. These economic
activities have to a large extent developed around the mining industry. When
the mining economy experiences a downward trend, the whole economy
suffers. This is what happened with the closure of the Teberebie mines in the
late 1990s as a result of the slump in global gold prices. Currently gold prices
are soaring, and the mining economy is strong. The vibrancy of the revived
mining economy has provided a favourable environment for business in
Tarkwa in particular. The study sought the views of some self-employed
people in small-scale manufacturing and related services, retailing and per-
sonal services in the study sites (161 respondents in Tarkwa, 13 in Adieye and
11 in Teberebie). About 81 per cent of the sampled operators indicated that
over the years their enterprises had generally performed well. Thirty-six
per cent of operators (in Tarkwa 50 per cent) would like to expand their busi-
nesses. Sixty-six per cent of respondents saw the future prospects of their
businesses as being good, and in Tarkwa, 51 per cent and 42 per cent respect-
ively saw the future prospects of their businesses as good and very good.
A number of businesses and services have also been built around the
mining industry, including industrial laboratories, mining catering-support
services, private security services and ancillary mining services like providers
of servicing and parts for special mining equipment. In Tarkwa and its
surrounding areas there are two industrial laboratories, 22 ancillary firms, five
mining catering-support services, three private security services, two sanitary
 Dynamics of the gold mining industry in Ghana 183
companies, transport services and haulage companies, and contract mining
companies. The mining companies, however, import many of the inputs
for their operations (crushers, millers, steel, etc.). Local contracting firms,
from which some inputs are procured, also import some of their inputs (e.g.
chemicals). Working closely with local companies providing inputs, the large-
scale mining companies have assisted them to achieve the required quality
of inputs (e.g. products from the Western Casting and Tema Steel Company).
Special mention should also be made of artisanal or small-scale gold
mining in Wassa West District, particularly in communities around Tarkwa.
Though artisanal gold mining has been carried out in Ghana since time
immemorial, increased open-pit gold-mining operations by large-scale mining
companies have since the 1990s resulted in an increased contestation for
mining lands in the country. Trespassing by small-scale mining operators on
concessions acquired by large-scale mining companies has become common.
This has resulted in confrontations between the large-scale mining companies
and small-scale mining operators, who believe the acquisition of large tracts
of mining lands by the large-scale mining companies is an encroachment on
their sources of livelihood. The small-scale miners consider mining to be a
major source of employment and income, without which most of them would
have been unemployed and without a means of livelihood. Ten of the owners
of 26 small-scale mining concerns covered in the study used to work with
large-scale mining companies but had been retrenched. Having acquired the
skills and experience of mining, they were then applying these by working for
themselves. A few of them were farmers and artisans who had shifted to
working as small-scale miners. For 19 of them, mining was their previous
main occupation. This phenomenon was not noticed in Adieye and Teberebie,
though it is known that artisanal mining occurs throughout Wassa West
District.
An issue related to small-scale or artisanal mining is the number of small-
scale gold-buying firms in operation in Tarkwa. Eleven small-scale licensed
gold-buying firms were covered in the study in Tarkwa. These firms had been
set up since the early 1990s. They were owned mainly by individuals, only
three of them being partnership companies. The owners were relatively
young, between the ages of 39 and 48, with little schooling. All of them,
except two with secondary levels of education, had just basic levels of educa-
tion. They were all married, and they were mainly migrants from within the
Western Region and outside the region, but most of them had lived in
Tarkwa for between 15 and 39 years. They were previously engaged in all
kinds of employment – teaching, farming, hawking and driving – and only
one of them used to be a small-scale mining operator (galamsey). The firms
obtained much of the gold directly from small-scale mining concerns, both
legal and illegal, and from agents whom the licensed buyers themselves
pre-financed. They in turn sponsored gold producers scattered throughout
the district or region and beyond. Tarkwa offers the right environment for
their businesses. A considerable number of small-scale miners operate in the
184 Paul W.K. Yankson
district to supply them with gold. Many of the small-scale gold-buying firms
sell their gold through the Precious Minerals Marketing Company (PMMC),
a state agency which has an office in Tarkwa.

Unemployment and poverty

Two notable features of the livelihood dynamics of the study sites were
increasing unemployment and deepening poverty. These were recorded
through the household questionnaire surveys, as well as focus-group discus-
sions held in the study sites. Tarkwa, being a district capital and centre of
mining activities in the Western Region, tends to attract people looking for
waged employment, especially in mining. Most of those seeking employment
register with the District Labour Office at Tarkwa. From the records of the
Labour Office in 2005, there were 305 registered job-seekers on the active
register as of the end of January 2005. Of this number, 81 were registered in
January 2005 alone. Most of those on the active register were local applicants,
while others were from other districts. The number of job-seekers on the
active register had increased to 421 (415 males and 6 females) by the end of
August 2005. Despite the number of job-seekers, there have not been adequate
job opportunities. The large-scale gold-mining sector’s heavy reliance on
local labour in the past has been substantially eroded, as current production
has become more and more capital- and skills-intensive due to technological
developments (D’Souza, 2005). Moreover, the new mining economy provides
very few opportunities for unskilled labour. The types of job openings
favoured those with skills such as drillers, excavators, heavy-duty operators,
welders, multi-skilled operators, security personnel and artisans generally.
Unfortunately, most of those on the active register were unskilled labour.
Some were skilled but could neither read nor write and therefore had difficulty
securing job placements in the mining sector. The mining companies had
also been sponsoring some of the local population in higher institutions of
learning so they could become employable in the mining industry. Those
without skills were also being trained to take advantage of alternative liveli-
hood programmes being implemented by the mining companies.
In our household survey, respondents were asked to describe the employ-
ment situation in Tarkwa in the distant past (between 10 and 20 years ago)
and in the recent past (five to 10 years ago). Fifty-three per cent of them said
that, when small-scale ‘illegal’ mining was a thriving activity in the distant
past, there was very little unemployment in Tarkwa. In the recent past, as
many as 60 per cent of respondents said there were very limited employment
opportunities in Tarkwa, some attributing this situation to the prohibition of
‘illegal’ small-scale mining in the district by the government. Respondents in
Teberebie, and to a lesser extent those in Adieye, complained about a lack of
land for farming and alternative livelihood opportunities. The prospects for
youth employment were considered to be a major problem, as 70 per cent of
all respondent household heads indicated that both young men and women
 Dynamics of the gold mining industry in Ghana 185
would have virtually no chance of securing employment in the district in the
foreseeable future. The alternative was for young people to migrate to other
areas in search of employment. Indeed, migration out of the study sites was
one of the livelihood strategies adopted by households. About 46 per cent of
respondent households received remittances from household members and
other relatives living outside the settlements covered in the survey. The remit-
tances were mainly in the form of cash (69 per cent), foodstuffs (17 per cent)
and other forms (14 per cent) such as drugs, provisions and clothing. About
86 per cent of those who remitted them did so monthly. The main benefici-
aries of the remittances were the spouses of respondents and their children
(53 per cent). An interesting result of the study was the incidence of reverse
remittances, though on a limited scale (26 per cent), to those who had
migrated from respondents’ households. This was in the form of cash
(15 per cent), cash and food items (14 per cent) and herbal medicine (3 per cent).
Other households (9 per cent) received occasional gifts and income derived
from renting out rooms in their homes (3 per cent).
While recognising that the direct employment effects of large-scale mining
operations had been limited, respondents in our household survey pointed
to the benefits that communities had also received through companies’
investments in the area of their corporate social responsibility (CSR). A
significant proportion (34 per cent) of respondents, particularly in Teberebie
and Adieye, pointed to the provision of social services – health, education,
provision of potable water and sanitation facilities – by the mining companies.
Others (20 per cent) pointed to the provision of infrastructure such as road
construction, while roughly another 29 per cent pointed to other benefits,
including the implementation of alternative livelihood programmes in the
communities directly affected by mining. The views of respondents were also
sought regarding the effects of small-scale or artisanal mining on the liveli-
hoods of the households in their communities. A majority of respondents
(70 per cent) indicated that small-scale mining helped to retain young people
in the settlements covered in the study and so helped to stem the tide of social
vices in their communities. Others (23 per cent) pointed to income generated
in small-scale mining which supported economic activities, such as com-
merce, in the settlements. About 34 per cent of respondents in Tarkwa thought
that small-scale mining supported the local economy in terms of the demand
for goods and services. However, some negative impacts of mining operations
on livelihoods were also mentioned. Loss of farmland and environmental
degradation were given as the two major problems caused by mining in the
study settlements but social problems were also mentioned, especially by
respondents in Tarkwa (32 per cent).
Despite a boom in gold mining, poverty has worsened over the years due to
the loss of land as well as restricted wage employment opportunities in Wassa
West District. As the following extracts from focus-group discussions carried
out in various neighbourhoods in Tarkwa indicate, the impact of poverty is
widespread and households are adopting a range of coping strategies:
186 Paul W.K. Yankson
Here at Effuanta, Tarkwa, there is high level of unemployment. There
is no money, families cannot cater for their needs such as feeding
properly, paying school fees of children and other wards. Poverty is on
the increase as unemployment deepens and galamsey activities have been
proscribed by government so that many youth are unemployed and life
is hard here. People cannot access health care financially. They cannot
pay for the National Health Insurance Scheme (NHIS), school levies
for children and wards. There is financial stress on families leading to
waywardness of children, increased street children phenomenon, teenage
pregnancy and prostitution, disruption of marriages, vices, infidelity of
life partners, wage employees taking away wives of unemployed persons,
people selling-off personal effects to survive and to pay children’s
school fees.
(focus-group discussion, Effuanta, Tarkwa, 19 January 2007)

At New Atuabo, the unemployed persons find it difficult to care for their
households, difficult to feed them and cannot cater for the cost of their
children’s education. Their children are not healthy. Poverty is very much
widespread in this community, and poverty is intensifying as the
unemployment situation deepens. Those able to manage are few, and
they, generally, are those working with the mining companies operating
in the area. Some heads of household have rented some of the rooms
used by their households and are crowded into very limited spaces as a
coping strategy. Some heads of household have sent their children and
other wards to live with relatives elsewhere.
(focus-group discussion, New Atuabo, Tarkwa, 17 January 2007)

Concluding remarks
Despite the revived gold-mining economy in Wassa West District, there have
been very limited opportunities for wage employment. The increasing
unemployment levels, particularly among young people, and the deepening
poverty situation have had serious implications for sustainable livelihoods in
the district. Although Tarkwa and its environs are experiencing some out-
migration, the area is likely to continue to be a popular destination for
migrants looking for livelihood opportunities in the mining economy. Tarkwa
is growing in population size, and Adieye and Teberebie continue to be ‘dor-
mitory’ communities for those working or seeking employment in or near
Tarkwa. The population increase in the settlements is likely to compound
the problem of unemployment and all its attendant socio-economic effects.
Local government has been weak in responding to the increased levels of
urbanisation, as indicated by the inadequate provision of infrastructure and
environmental services. It, therefore, remains doubtful whether the boom in
mining will be able to provide the basis for sustained settlement growth.
 Dynamics of the gold mining industry in Ghana 187
Note
1 Computed from the 1970 Census Report, Special Report ‘A’ on Large Towns.

References
Agbesinyale, P. (2007) ‘Two sides of Ghana’s gold boom: A case of Wassa West
District’, The Oguaa Journal of Social Science (JOSS), 4(2): 1–22.
Akabzaa, T. and Darimani, A. (2001) ‘Impact of mining sector investment in Ghana:
A study of the Tarkwa Mining Region’, Draft report prepared for SAPRI.
Aryee, B. (2002) ‘Statistical overview of the mining industry in Ghana’, in K. Barning
and B. Dorgbetor (eds), Proceedings of the national mining conference on ‘Mining,
the Environment and Sustainable Development’, Accra: Minerals Commission.
Aubynn, A.K. (2009) ‘Sustainable solution or a marriage of inconvenience? The
co-existence of large-scale mining and artisanal and small-scale mining in
the Abosso Goldfields concession in Western Ghana’, Resource Policy, 34 (1–2):
64–70.
Barning, K. (n.d.) ‘Case study on foreign investment in mining: The case of Ghana’,
mimeo, Accra: Division of Policy and Planning Analysis, Minerals Commission.
Davis, G. and Tilton, J.E. (2002) ‘Should developing countries renounce mining? A
perspective on the debate’, online at http://colloid.org/Academic/econbus/pdf/
Davis/Davis and Tilton/2002.pdf, accessed 23 April 2008.
Dickson K.B. (1969) A historical geography of Ghana, London: Cambridge University
Press.
D’Souza, K.P.C.J. (2005) ‘Artisanal and small-scale mining in Africa: the poor
relation’, in Marker, B.R., Petterson, M.G., McEvoy, F. and Stephenson, M.H.
(eds), Sustainable minerals operations in the developing world, London: Geological
Society, pp. 95–120.
Hilson, G. (2001) ‘A contextual review of the Ghanaian small-scale industry’, Mining,
Mineral and Sustainable Development, 76: 1–29.
Hilson, G. (2002) ‘Harvesting mineral riches: 1000 years of gold mining in Ghana’,
Resource Policy, 28: 13–26.
Hilson, G. (2004) ‘Structural adjustment in Ghana: Assessing the impact of mining
sector reform’, Africa Today, 51 (2).
Hilson, G. and Potter, C. (2005) ‘Structural adjustment and subsistence industry:
Artisanal gold mining in Ghana’, Development and Change, 36(1): 103–31.
Huq, M.M. (1989) The economy of Ghana: The first twenty-two-years since independ-
ence, London: Macmillan.
Larsen, M.N., P. Yankson and N. Fold (2009) ‘Does FDI create linkages in mining?:
The case of gold mining in Ghana’, in A. Sumner, D. Sanchez-Ancochea,
E. Rugraff (eds), Transnational Corporations and Development Policy: Critical Per-
spectives Basingstoke: Palgrave Macmillan, pp. 247–73.
Opportunities Industrialisation Centres International (OICI) (2004) ‘ “GAG-
CARES”: Response to terms of reference for community development implemen-
tation for GAGL mine, Tarkwa Wassa West District-Ghana’, presented to Ghana
Australian Goldfields Limited (GAGL).
Sarin, R., Reisch, N., Kalafut, J. and Slack, K. (2006) ‘Tarnished gold: Mining and
unmet promise of development’, online at http://www.bicusa.org/proxy/Document.
9518.aspx, accessed 23 April 2008.
188 Paul W.K. Yankson
Songsore, J., Yankson, P.W.K. and Tsikata G.K. (1994) ‘Mining and the environment:
Towards a win-win strategy. A study of the Tarkwa-Aboso-Nsuta mining complex’,
final report prepared for the Ministry of Environment, Science and Technology
(MEST-EPC) and the World Bank in support of the formulation of a growth
strategy for the Western Region.
Twerefou, D.K., Aryeetey, E. and Bafour, O. (2007) Impact of mining sector reforms on
output, employment and incomes in Ghana, 1980–2002, Technical publication
no. 75, Legon: Institute of Statistical, Social & Economic Research, University
of Ghana.
Wassa West District Assembly (2002) ‘Medium term development plan 2002–2004’,
Tarkwa (unpublished).
14 Frontier mining settlements
Livelihood promises and
predicaments
Deborah Fahy Bryceson and
Paul W.K. Yankson 1

Introduction
Newly discovered mineral sites, perhaps more than anywhere else, encapsu-
late a frontier mystique. At the outset, the geological and social unknowns
combine with a sense of enormous economic potential. People flock to the
location in large numbers with high expectations of gain. There they must
contend with the fact that others from far and wide have descended on the
location with the same hopes and expectations. The local economy, culture
and society that emerge are an aberration from the surrounding countryside
and culture rooted in a strong sense of the locational past. Mining settle-
ments pivot around the production and consumption patterns of miners
living in the immediate present, who seek to gain their livelihood and make
their fortune then and there.
However, newly discovered mineral sites are just one form of a wide array
of mining sites whose one commonality is that they are all locations of
mineral exploitation. The type of mineral being exploited, the richness and
spatial extent of the deposit, the duration and depth of the excavation and the
scale of production mould the ‘carrying capacity’ of the site with respect to
employment and population settlement. The Ghana case study in this section
focused on what could be termed ‘re-discovered’, large-scale gold mining sites
in Wassa West District where highly capitalised and technologically advanced
surface methods of extraction are deployed. Decades ago this area was part
of an important gold-producing area that became sub-economic with the
progressive depletion of the mineral supply. However, the high world price of
gold and new technology spurred the re-opening of these fields with foreign
investment. The Tanzania case study, based in the northwestern regions of
Mwanza (gold) and Shinyanga (diamonds), on the other hand, dealt with
much newer sites, where small-scale miners are digging at shallow levels with
relatively low levels of technology.
This thematic overview explains more of the background to the frontier
mining case studies: reviewing the position of mining in the national econ-
omies of Ghana and Tanzania, outlining some of the salient differences
between small-scale and large-scale gold mining, and contrasting small-scale
190 Deborah Fahy Bryceson and Paul W.K. Yankson
gold as opposed to diamond mining before discussing the rural–urban
dynamics of mining settlements.

Mining livelihoods in Africa

Post-colonial states of Africa have been divided between a few notable coun-
tries where mining historically took a leading role in the economy and the
majority of others where agriculture was primarily the driver of the economy.
In the latter, as small-scale agriculture has experienced serious decline, African
states have more actively sought foreign investment. Most recently, Chinese
investors have upped the stakes with their eagerness for African raw materials.
In this context, mining has come to the fore in an array of former primarily
agrarian African countries, where valuable mineral deposits were known to
exist or have recently been discovered.
Although Ghana has been famed for its gold for centuries and was formerly
called the ‘Gold Coast’ during the colonial period, in more recent years of the
post-colonial period, mining flagged, accounting for only between one and
three per cent of wage employment. In the more recent liberalised business
climate, however, several large- and medium-scale mining companies have
been attracted to the country. National gold export earnings have been on
an upward trajectory.
Tanzania has a far less prominent history in mineral exploitation. During
the first two decades after the country’s independence in 1961, mining was
negligible, with the exception of Williamson diamond mine at Mwadui in
Shinyanga. Declining returns to small-scale peasant farming combined with
the rising price of gold on the world market spurred some farmers to start
searching for mineral deposits in their home areas. Small-scale miners’ num-
bers have gained momentum with an estimated 500,000 people, mostly men,
practising small-scale mining now (Chachage, 2005; Jønsson and Bryceson,
2009). Mining became the fastest growing sector with a 34 per cent increase
between 2004 and 2005 alone. Mining export growth of over 16 per cent per
annum was achieved in 2006 amidst booming gold prices (Tanzania, 2006).
Given their different histories of mining, it is not surprising that the legisla-
tive contexts of mining in the two countries are distinct. In Ghana, small-scale
mining has been legalised since 1989 and miners are deemed legal in the
context of being registered and granted a mining concession. However, a
large majority of miners have not gone through these procedures and operate
illegally defying proscription. The total number of small-scale gold mining
operators is estimated at 300,000 people (Hilson and Potter, 2005).2 There
is considerable friction between the large legal and small-scale illegal sectors.
Employment in large-scale mining is very restricted where new surface mining
techniques have been introduced. Its land-extensive nature has engendered
land disputes between large-scale mining firms and local residents. Small-
scale illegal miners feel a sense of moral outrage at the loss of land and
livelihood.
 Frontier mining settlements 191
Tanzanian small-scale miners have been subject to less control. Small-scale
mining was illegal during most of the Nyerere period after independence
until 1979 when they were technically allowed to peg claims. Legislation in
19973 ordained a clear legal space for a small-scale mining sector, prompted
by Tanzania’s concern to delineate small- as opposed to large-scale spheres.
Recent mining legislation related to Primary Miners Licensing and efforts to
formalise Tanzania’s small-scale mining sector have generally touched only
the most prominent small-scale operators, the primary mining licence owners
(Fisher, 2007). This leaves the majority of small-scale mining operations
unregulated and open to highly exploitative labour practices, which imparts
a true frontier setting to Tanzanian mining sites. The legality of the land
and labour usage is often questionable. Capital investment in equipment,
labour management and the risks and costs related to miners working with-
out striking gold falls primarily on the pit holder.
Certainly on a day-to-day basis, the secure profitable position of claim
owners may generate tension amongst the rank and file; nonetheless, the
borders between the three groups of miners, namely: claim owners, pit holders
and diggers, can blur. Claim owners and pit holders may work as ordinary
miners in bad periods to get cash.4 In any case, over the longer run, claim
owners have insufficient capital to continue mining once a depth is reached
that requires high-powered water pumps. They will be forced to cease oper-
ations. At that point, or in anticipation of that point, the claim owners may
strike deals with large-scale mining companies which wait for the inevitable
time when technologically sophisticated large-scale mining techniques are
needed at greater depths.
Gold, which begins with alluvial panning, and proceeds to digging ever
deeper shafts and tunnels down to levels that are no longer technologically
possible for small-scale miners, is a more capitalised production process
with more teamwork and hierarchical structure than diamond mining. The
economic structure of diamond mining in Tanzania tends to be less hier-
archical because the technology and levels of investments for small-scale
diamond mining are less demanding. A two-tiered production system prevails
in which locally resident male household heads with farmland use part of
their land for gold mining and procure labour for the digging. Some may
mine the land themselves with male family members, but older men with the
financial means to become ‘diamond mine owners’ tend to provide equipment
to a team of ‘diggers’ – numbering two to four men generally – whom they
permit to dig on their land in specified spots.
The mine owner generally supplies digging equipment to the mine labour-
ers, having himself borrowed the equipment from a diamond buyer on the
condition that any diamonds found must be sold to that buyer specifically.
The diggers excavate and wash the sand looking for diamonds. When dia-
monds are found, the mine owner takes them to the local licensed buyer who
deducts the costs of the mining equipment. Some mine owners ask for one of
the diggers to accompanying him in order to demonstrate that the diamond
192 Deborah Fahy Bryceson and Paul W.K. Yankson
transaction and the apportioning of shares from the diamond sale thereafter
are fair. While varying from place to place, mine owners are likely to take
half the proceeds, distributing the rest to their diggers with deduction for the
living costs and loans that they advanced to the diggers up to that point.
The outstanding difference between the two country case studies is that
the Tanzanian sites based on small-scale mining are labour-absorbing for
unskilled diggers, whereas the techniques used on the much older Ghanaian
gold sites are labour-repelling. The Ghanaian government’s proscription of
small-scale illegal mining operations clearly favours investment by large-scale
mining concerns which tend to have been allocated land formerly worked by
numerous small-scale illegal mining concerns. Small-scale operators are per-
sistently moving from one area to another in search of gold. Their search
takes place in active mining areas and includes the concessions of the large-
scale mining companies. Operating as ‘gangs’ comprising both men and
women, their mine work within the company concessions has been a major
source of conflict in Ghana’s gold mining industry.
Ghana’s large-scale foreign mining companies deploy techniques that
restrict the use of manual labour. Employment opportunities in the large-
scale gold mining sector are now primarily skilled jobs leading to increasing
levels of general unemployment and poverty amongst the uneducated and
untrained. The gold mining sector dominates the local economy and does
help to generate employment in commerce and services in Wassa West District.
Livelihood sustainability in the gold mining sector has necessarily become
the central issue for Ghanaian households in mining settlements
This is experienced in Tanzanian small-scale mining to a lesser degree.
Mining settlements ‘age’ as their mineral deposits get deeper and/or depleted,
but in Tanzania there is the feeling of continually new horizons. Small-scale
miners usually have a green light to seek new sites. Older sites, such as the
case study settlements reviewed in Chapter 12 above, inevitably become less
mine-oriented. Youthful men, in particular, are likely to emigrate but those
left behind and the service infrastructure continue to provide a basis for
livelihood sustainability.

Urbanisation and migration

High-value minerals and gemstones – gold or diamonds – are superb catalysts
for capitalist enterprise, long-distance trade and urbanisation processes. They
attract entrepreneurs and mining labour in large numbers from within and
beyond national boundaries, despite the fact that the local price miners
receive for their mineral output may represent only half to three-quarters of
the international consumer market price (Phillips et al., 2001). The poverty of
the countryside where the labourers originate and their low opportunity
costs result in elongated gaps in international price formation. Nonetheless,
the high value of mineral production relative to local agricultural output and
the in-migration of mine labourers with the multiplier effects of the service
 Frontier mining settlements 193
sector that springs up around them catalyse the population growth upon
which urban populations, urban divisions of labour and urban exchange
relationships congeal.
Mining sites experience rapid changes in migration flows to and from the
settlement. There have been virtually no studies of mine-focused population
movement (Bryceson and Jønsson, forthcoming). The Todaro and Harris-
Todaro models of urban migration based on migrants’ anticipated urban
job prospects or wage levels have long been abandoned (Montgomery et al.,
2004). More recent work by Bigsten (1996) suggests that urban-based family
relations arrange or at least encourage a rural migrant’s urban work placement
through communication channels established within circular rural–urban
migration patterns. Familial mine labour recruitment on the basis of ‘bush
telegraph’ lines of communication between rural households, strengthened
by the ever-growing use of mobile phone networks, could be likened to the
labour recruitment through familial ties described by Bigsten (1996) for
African urban areas.
Labour absorption is the motor force of livelihood diversification and
urbanisation. Age and gender stereotyping strongly structure labour force
participation and in turn influence household composition in one of two
ways through: first, household members selectively involving themselves in
chain migration following the original migrant for purposes of household
reunion; and second, new households which are formed through marriage of
couples at the new destination. In the case of the latter, the chain migration
is most likely to happen when the household member is of the age, gender
or occupational inclination that will ‘fit in’ with the new settlement.
While the gender ratio in mine-led urbanisation is biased towards men in
the first instance, female migration soon follows male miners’ influx. Women
are attracted to the area by the sharp rise in the demand for services induced
by mining or to join their husbands. When returns start to decline and male
miners begin migrating to new mining sites, they leave behind a larger resi-
dent population than existed at the outset of the mining boom, including
wives and children. In so doing, the gender ratio of the settlement may tip
from a male to a female bias.
We can trace small-scale mining settlements over a number of stages of
what might be termed the ‘settlement cycle’ beginning with the initial rapid
‘primary growth’ to peak settlement levels of miners themselves; followed
by a ‘secondary growth’ of service providers which comes quite quickly
thereafter and proceeds simultaneously with the miners’ migration impetus.
‘Tertiary growth’ is characterised by ‘family reunion’ where wives and chil-
dren follow mining husbands, or more likely ‘family formation’ whereby
miners meet and have relationships with or marry women resident in the
settlement who give birth to their children. It is difficult to delineate these
stages of population growth because the changes happen quickly and often
overlap one another. Census data every decade or more obscure the dynamics
of the inter-censal period. In addition, in many places small-scale miners are
194 Deborah Fahy Bryceson and Paul W.K. Yankson
illegal or only quasi-legal, making undercounting in censuses highly likely.
Thus, most estimates of population growth in mining settlements have to be
based on local leaders’ often biased guesstimates and survey data.
In Ghana, employment opportunities for unskilled labour in the large-
scale underground mining during the 1960s and 1970s spurred urbanisation
of the key mining settlements. The revival of the gold mining industry is
not as labour-absorbing as before and the local multiplier effect in terms
of service sector work has not been as effective. The state has not provided
adequate levels of services and investment in infrastructure. Under pressure,
the mining companies have demonstrated a growing sense of corporate social
responsibility involving more sensitivity to the housing and service needs of
their workers and dependants in the mining residential compounds. This has
extended to mining firm investment in infrastructure and service-provisioning
for local settlements directly affected by mining operations.
Small-scale mining settlements in their initial phases are usually very
makeshift in character, but over time, government and the private sector are
likely to invest in infrastructure, be it roads, schools, churches, mosques,
medical clinics, etc. As and when primary growth of the mining population
starts declining, the secondary and especially tertiary populations are likely
to be more stable. Mining settlements that have attracted considerable infra-
structural investment and are on a road are likely to continue to experience
in-migration from the surrounding countryside from rural youth seeking
improved life circumstances and outsiders seeing business possibilities and
employment in the settlement’s service infrastructure.

Rural–urban transitions
The rural–urban dynamics of mining settlements are complex. Rapid in-
migration causes bush, green field or village sites to transform rapidly into
areas with urban characteristics. Mine labourers face many material insecur-
ities connected with taking up residence in a new locality and trying to earn a
living by mining. Large-scale mine labourers generally receive regular work
payments, but face the possibility of being laid off. Small-scale miners, on the
other hand, never know when they will get a mineral strike and they lead a
hand-to-mouth existence. Many engage in own or family farming for their
subsistence. However, combining mining and farming livelihoods is jeopard-
ised by small-scale mining techniques that damage the topsoil, compromising
local farming efforts in the long run. Large-scale surface gold mining is the
worst offender, causing the loss of large amounts of valuable topsoil, which
restricts farming in the mine concessions to the cultivation of basic subsist-
ence foods rather than perennial cash crops.
Mining settlements are premised on a non-renewable natural resource.
Large-scale mines have long-term plans and labour policies calibrated to
exploit deposits at specific sites within a particular time frame. Small-scale
miners, on the other hand, deal with changing mining conditions on a
 Frontier mining settlements 195
day-to-day basis. In both forms of mining, the methods used to extract the
minerals depend on the alluvial, soil or hard rock positioning of the mineral
in question, albeit at different levels of technology. Small-scale miners have
few technological problems with alluvial panning, whereas digging for gold
and diamonds involves descending to ever deeper layers in the soil and hard
rock, with increasing danger and technology costs. Small-scale miners, with-
out the capital to invest in sophisticated digging, blasting and water-pumping
machinery and skill improvement, face diminishing returns from their efforts
the longer they stay on a site. Many miners are tempted to leave the mining
settlement and venture beyond to a new frontier when news of new mineral
strikes reaches them. This mobility option is relatively constrained in Ghana
as opposed to Tanzania. (Jønsson and Bryceson, 2009)
Small-scale mining has very strong forward and backward linkages to other
sectors of the local economy. It is a boon to commercial agriculture because
it raises the demand for purchased food. It generates an array of mine-related
processing jobs in addition to the manual labourers digging the gold and
diamond pits, namely: rock crushing, sieving, amalgamating the metal, etc. It
spawns small-scale industrial activity with the production of rock-crushing
machinery and pumping devices. But above all, it causes a proliferation of
service activities catering for the food and entertainment needs of the mine
workers. The miners are known to be big spenders when they make a strike.
There are plenty of people on hand ready to be at their service, providing
alcohol, entertainment and sex for the miners’ enjoyment. The multiplier
effect of mine earnings facilitates widespread income generation and more
even income distribution within the settlement. Small-scale mining may be
temporally short-lived but it has a surprisingly broad and favourable impact
on the local economy through service sector expansion and trade. Large-scale
mining, like small-scale mining, provides cash injections into the local
economy. Although wage employment is far more restricted, what employ-
ment is offered is relatively more stable than self-employment in small-scale
mining.
Mining is an unpredictable driver of settlement transformation. Primary
growth of the mining population will taper off at varying rates depending
on how mineral-rich the mining site is, how fast the extractable levels of the
mineral are depleted, and how many alternative new sites there are. With
the limited capital and level of technology that small-scale miners have at
their disposal, they are primarily restricted to the extraction of surface and
near-surface levels of mineral deposits. If miners hear of a more promising
site with more readily extractable surface deposits, many will be tempted to
emigrate, usually leaving wives, girlfriends and children in situ, not knowing
what lies ahead.
Mining settlements worldwide tend to rapidly display a sub-culture in which
hard-working, risk-seeking men, oblivious to workplace health hazards5 and
financial insecurity, are observed becoming exceptionally happy-go-lucky
and pleasure-driven, dissipating their earnings on ‘high life’ social drinking
196 Deborah Fahy Bryceson and Paul W.K. Yankson
and sexual liaisons in the aftermath of striking gold. Marital stability is
likely to suffer as alluded to by Yankson.
There are two main factors acting to reduce the incidence of dire poverty:
first, the population is on average very youthful and mobile. If they see a lack
of promise in the local economy as a result of a slump in gold and diamond
mining they can simply move elsewhere; men to areas where minerals are
being successfully mined, whereas women are likely to move to the service
economies surrounding these boom locations. Second, as long as there is
sufficient land in the area for household subsistence farming, and climatic
conditions provide adequate growing conditions, extreme poverty and destitu-
tion is unlikely.

Conclusion
Weighing the above-mentioned conflicting tendencies, we venture to propose
that the capricious nature of small-scale mining can nonetheless ultimately
lead towards sustainable urban settlement. Johannesburg, South Africa’s big-
gest city, which started as a Rand mining settlement just over a century ago,
is an outstanding example. In the first instance, a sustained urban growth
trajectory depends on access to basic needs for water and food in combin-
ation with good mining output. As the settlement develops, the gender and
age composition of the mining settlement becomes less skewed. The settle-
ment’s economic foundation, initially catalysed and nourished by mining,
is not extinguished as the mining component of the local economy becomes
less dominant. Instead the urban settlement retains viability or even a growth
momentum through the trade, service and possibly semi-industrial activity
of a population concentration that is denser and availed of more economic
opportunity, social choice and cultural relativity than in the agrarian coun-
tryside – hence its continued ‘bright lights’ attraction to outsiders. Small-scale
mining as a frontier livelihood in any given place must be viewed as a tempor-
ally delimited phenomenon. Its long-term welfare impact on the resident
population extends beyond mining to the influence of urbanisation and the
growth of trade and services.

Notes
1 This chapter has been inspired by the late Professor Ndalawa Faustin Madulu’s
insight and determination to research Tanzanian mining communities. Thanks
go to Jesper Bosse Jønsson for his helpful comments on the first draft of this
chapter.
2 As many as 90 per cent of Ghanaian small-scale miners may be illegal.
3 The Mineral Policy of 1997 (Tanzania 1997).
4 It was not until 1990, after the monopoly of the State Mining Company had
ended in the late 1980s, that the then Ministry of Water, Energy and Minerals
liberalised mining and selling of gold completely (Phillips et al., 2001). From then
on, small-scale miners were encouraged to acquire claims (Lange, 2006).
 Frontier mining settlements 197
5 See Jønsson et al. (2008) for details on small-scale miners’ hazardous use of
mercury for gold extraction.

References
Bigsten, A. (1996) ‘The circular migration of smallholders in Kenya’, Journal of
African Economies, 5(1): 1–20.
Bryceson, D.F. and Jønsson, J.B. (forthcoming) ‘Gold digging careers: Small-scale
miners’ livelihood and lifestyle choices in rural Tanzania’.
Chachage, C.S.L. (2005) ‘Can Africa’s poor inherit the Earth and all its mineral rights?’
Paper presented at the General Assembly of the Council for the Development
of Social Science Research in Africa, Maputo, 6–10 December.
Fisher, E. (2007) ‘Occupying the margins: Labour integration and social exclusion
in artisanal mining in Tanzania’, Development and Change, 38(4): 735–60.
Hilson, G. and Potter, C. (2005) ‘Structural adjustment and subsistence industry:
Artisanal gold mining in Ghana’, Development and Change, 36(1): 13–26.
Jønsson, J.B., Appel, P.W.U. and Chibunda, R. (2008) ‘A matter of approach:
The retort’s potential to reduce mercury consumption within small-scale gold
mining settlements in Tanzania’, Journal of Cleaner Production, DOI 10.1016/
j.jclepro.2008.04.002.
Jønsson, J.B. and Bryceson, D.F. (2009) ‘Rushing for gold: Mobility and small-scale
mining in East Africa’, Development and Change, 40(2): 249–79.
Lange, S. (2006) Benefit streams from mining in Tanzania: Case studies from Geita
and Mererani, Bergen: Chr. Michelsen Institute.
Montgomery, R., Stren, R., Cohen, B. and Reed, H.E. (2004) Cities transformed: Demo-
graphic change and its implications in the developing world, London: Earthscan.
Phillips, L.C., Semboja, H., Shukla, G.B., Sezinga, R., Mutagwaba, W., Mchwampaka,
B., Wanga, G., Kahyarara, G. and Keller, P.C. (2001) ‘Tanzania’s precious mineral
boom: Issues in mining and marketing’, Washington, DC: USAID.
Tanzania, United Republic of (1997) The Mineral Policy of 1997, Dar es Salaam:
Government Printers.
Tanzania, United Republic of (2006) The Economic Survey 2006, online at Dar es
Salaam, http://www.tanzania.go.tz/economicsurveyf.html.
15 Conclusion
Niels Fold, Katherine V. Gough and
Jytte Agergaard

In this concluding chapter, we bring together the main findings from the
frontier studies presented in this book. First we highlight the benefits of
adopting a frontier approach to study rural–urban dynamics by summarising
some of the main findings of the research. Then we outline some of the
challenges of scaling this approach up to the national and continental levels.
Finally we draw some policy implications of the research to round off the
chapter and the book as a whole.

Rural–urban dynamics in frontiers

In this book, we have explored rural–urban dynamics through a focus on
frontiers. The research programme was built up around the conceptualisation
of frontiers as being regions where linkages to global markets are dominated
by one particular product (coffee, pineapples, gold, etc.) or a number of
similar products (handicrafts, fresh fruit). The only exception is the manu-
facturing frontier in Thailand, which is characterised by different industries
and a diverse product range. The purpose of this focus on frontiers was to
construct a relatively ‘homogenous’ spatial entity with a particular type of
linkage to world markets in order to examine the interaction between
global markets and rural–urban dynamics. As the chapters on the individual
frontiers have highlighted, this can be a revealing way to tackle rural–urban
dynamics. Although the experiences on each frontier are unique, certain
trends have been highlighted in the agricultural, handicraft/manufacturing,
and mining frontiers respectively.
The agricultural frontiers have largely been driven by market dynamics,
in some cases dominated by leading agro-industrial transnational com-
panies. In some agricultural frontiers, livelihoods continue to be highly
dependent on a specific crop. In the most dynamic settlements, however,
although the economic base is closely interlinked with a specific crop, the
majority of residents are engaged in non-farm occupations, with trade
and services being especially important. These dynamic settlements attract
in-migrants who work as agricultural labourers or establish their own enter-
prises. Trade and mobility clearly benefit from good transport infrastructure,
 Conclusion 199
which provides access to the surrounding rural region and a network of
larger cities.
In the handicraft/manufacturing frontiers, agriculture is declining in
importance as the production of handicrafts, which once complemented
agriculture in the slack season, is becoming the prime occupation. Rural
incomes are being boosted due to this non-farm work. However, the central
assumption of the virtuous cycle of farm–non-farm relations, namely that
the two facets of the rural economy are co-situated, does not hold true.
The profits from handicraft production are not being ploughed back into
agriculture but are being used to expand craft production and to consolidate
dwellings. The cases from Thailand and Vietnam highlight the manner
in which craft and factory production relies on, and is increasingly sustained
by, wider links and associations.
The mining frontiers are characterised by quite a different type of settle-
ment formation, as they are premised on a non-renewable natural resource. As
the cases from Tanzania and Ghana illustrate, mining as a frontier livelihood
is a temporarily delimited phenomenon and thus an unpredictable driver of
settlement formation. Settlements in mining frontiers typically pass through a
number of stages, from temporary settlement to consolidating settlement
to established settlement. When the mining frontier moves elsewhere, the
established settlement does not necessarily die. Settlements can establish a
momentum of their own as their main livelihoods switch to trade and services.
While specific characteristics of settlement formation can be identified for
each of the frontier types, some overall trends which cross frontier type can
be identified. A theme that runs through all the frontier settlements is the
importance of the ability of households to be flexible and to diversify or
specialise production according to demand for the frontier product. The
demand for these products is increasingly dependent on the vagaries of the
global market, making households highly vulnerable to sudden fluctuations
in demand. Furthermore, the settlements in all of the frontier types are
characterised by increasing social differentiation and inequality.
In all of the frontiers, globalisation is clearly an important force impacting
on rural–urban dynamics. In order to explore the ways in which this impact is
worked out in frontier settlements, livelihood analysis was combined with
global value chain analysis to varying degrees. By tracing the ways in which
actors on the global stage impact on value chains, and how actors in the value
chain influence the demand for certain products, we have been able to show
how global changes impact on households. This clearly shows the benefits
that can be gained by linking these two types of analysis through a focus on
frontier regions.

Rural–urban dynamics at national and continental levels

Our frontier approach is less applicable on the national level. National terri-
tories are constituted by regions, many of which may not fit into our frontier
200 Niels Fold, Katherine V. Gough and Jytte Agergaard
concept, for example agricultural regions with highly mixed land use and
cropping patterns, regions dominated by a dense urban network of towns or
cities, or regions dominated by out-migration and stagnating economies – in
short, regions with other characteristics than those defined by our frontier
concept. Consequently, we are unable to outline ‘national rural–urban
dynamics’, interpreted here as the totality of complex rural–urban dynamics
that are observable at the national level. Based on the findings in this book,
however, it is possible to identify two general types of rural–urban dynamics
that are integral parts of national development trajectories. The typology
is structured according to a simple bipartite division of rural–urban dynam-
ics in the context of national development, namely into hinterland areas and
areas which are resource rich.
Rapid and comprehensive national industrialisation processes are closely
interlinked with the dynamics of urban metropolises that encompass a highly
diverse set of economic activities. Rural–urban dynamics within the hinter-
land of national metropolises are strongly influenced by the processes of
rapid urbanisation and industrialisation that take place in the city and its
surroundings. Livelihoods, mobility and settlement development in hinter-
land areas are heavily influenced by changes in urban demand for consumer
goods (such as a shift in dietary preferences due to the emergence of new
social classes), as well as in demand for raw materials and labour by manu-
facturing industries that are oriented towards global markets.
Some settlements in the hinterland consolidate and expand as livelihoods
are diversified when households start to exploit the increasing number of in
situ income-generating activities. Spatial dispersion of manufacturing indus-
tries from congested areas in the city to new industrial zones in previous
agricultural frontiers enhances the possibilities for livelihood diversification.
The population is able to remain in accessible settlements or towns and
migrants are attracted by new employment opportunities. New skills and
working habits are implanted among the original residents and migrants who
enter into these non-agricultural activities, which are primarily located in
rural settlements. However, these settlements do not necessarily underpin the
industrialisation process by heralding a transformation into localities for
incipient manufacturing industries, even though similar organisational forms
of labour division are being introduced. The settlements form part of broader
processes in which rural areas are transformed and settlement populations
mixed with (labour) migrants and recast into new social relationships. In this
sense, these frontier settlements reflect new rural–urban dynamics at the
national level, influenced by their enhanced connections to global markets.
These improved opportunities do not prevail in all areas of the hinterland,
however, as previous income-generating activities are eroded resulting in
settlements stagnating and their residents having to look for employment
elsewhere.
Some resource-rich (e.g. agriculture, mining) frontiers may be of significant
importance for the national economy if export revenues generated in the
 Conclusion 201
frontier constitute a substantial share of total exports. Rural–urban dynamics
in these frontiers interact with broader development processes on the national
scale. The one-way flow of commodities from the area of production to
the ports of export corresponds to more complex cross-country and multi-
directional flows of labour and money, the latter in the form of both remit-
tances and individual investment capital. Some regions which are not
resource rich serve these frontiers as long-term labour reserves, as is the case
for northern Ghana, where remittances primarily support the reproduction
of migrants’ extended households. Other regions, however, may benefit from
investments by migrants who have the ability to accumulate a sufficient surplus
to invest in property or small businesses in their hometowns.
Within resource-rich areas, livelihood diversification and mobility patterns
tend to strengthen the position of particular settlements with suitable locations
for trade in consumer goods, inputs and tools for production. The growth of
these settlements into self-reliant regional centres may be a sustainable way
of creating more regionally balanced socio-economic development at the
national level. However, their potential may never materialise if the frontier
expansion is interrupted by significant changes on global markets, as hap-
pened in the case of smallholders growing pineapples close to Accra, Ghana.
In these cases, the diversification of livelihoods comes to a halt and settlements
stagnate. In other cases, the spatial needs of large-scale agricultural or mining
operations have resulted in the displacement of indigenous settlements and
the establishment of unviable resettlements as inhabitants are dispossessed
from their land.
These reflections have some important methodological implications.
Rural–urban dynamics examined through a ‘frontier’ lens reveal a fragmented,
although instructive, set of insights of different processes that form part of
the complex national totality. Although the frontiers studied do not provide
an adequate basis for outlining a comprehensive picture of rural–urban
dynamics at the national level, they nonetheless feed into the simple bipartite
division of rural–urban dynamics into the broad categories of hinterland
and resource-rich areas. Hence, if further research aims to understand rural–
urban dynamics at the national level, attention should be paid to the selection
of different regions with distinctive positions in the national development
process (including, for example, labour reserve regions, resource-poor
regions, and geographically isolated regions) before conducting analyses on
livelihood diversification, mobility and settlement transformation like those
in this book.
Turning to the continental level, differing rural–urban dynamics have
been shown to occur within Southeast Asia and Africa, in particular in
relation to migration and mobility practices linked to livelihood diversifica-
tion. Southeast Asian countries have developed new sources of income
in manufacturing industries and related service activities fostering labour
migration opportunities beyond the otherwise seasonal migration which
characterises migration in the more commodity-dependent economies of
202 Niels Fold, Katherine V. Gough and Jytte Agergaard
Africa. Labour migration creates new opportunities for livelihood diversifica-
tion, not least in the hinterland settlements where livelihoods can be more
flexible due to their proximity to urban centres. The additional economic
activities and increasing cash income make it possible for household members
to own their own means of transportation, thereby increasing their activity
radius and day-to-day mobility. The effects of this ‘mobility revolution’ are
also visible – albeit to a more limited extent – in the resource-rich areas, such
as in the coffee frontier in Vietnam, where access to transportation (particu-
larly motorbikes) influences employment opportunities, access to markets
and services, and relations with traders. This increased mobility and associ-
ated development trajectories are not as pronounced in commodity-
dependent African economies. Nonetheless, despite the positive effects of
industrialisation in Asia, it is important not to overlook the increasing
inequalities between households and settlements caused by different mobility
possibilities and patterns. Asian frontiers may be characterised by greater
mobility and more intense rural–urban dynamics than is the case in African
frontiers, but a higher degree of inequality is emerging in the former.

Policy implications
Rural–urban dynamics are influenced by various types of policies which
operate at different spatial scales and are directed towards differing sectors of
the economy. In this book we have examined variations in rural–urban
dynamics and how these are played out in a range of frontier types. Policies
and various forms of public regulation have been considered to the extent
they have been of importance for rural–urban dynamics. For instance, in
Vietnam the state has played a central role in facilitating population redistri-
bution and – as part of the de-collectivisation process – establishing a relatively
egalitarian redistribution of land for agricultural production. In addition,
crop and income diversification have been part of state and provincial planning
policies, in particular from the mid 1990s onwards as part of the continued
five-year planning process. While centralised planning is of major importance
in the economy, producers are not cushioned in any way from global market
dynamics. In contrast, immigration to the cocoa frontier in Ghana is less
planned, although the state is heavily involved in the regulation of the
purchasing system (through fixed prices, quality control, etc.). In general,
though, in Ghana agriculture takes place within a system dominated by
customary land rights which govern the differing access to land by migrants
and indigenous groups.
There are a number of general policy implications from the frontier
case studies, concerning various policy instruments and their impacts on
rural–urban dynamics, which will be highlighted here. First, all the studies
have illustrated the importance of mobility in the diversification of liveli-
hoods as households straddle several locations in their efforts to find income-
earning activities. For marginalised and poor households, increased mobility
 Conclusion 203
or permanent migration may be the only way to escape poverty, even
though they often possess fewer of the resources needed to partake in flexible
livelihoods, such as means of transport. Hence, it is important to understand
the actual barriers to the movement of people and goods and be aware
that different income groups might meet different barriers to their mobility.
Regulatory measures, such as road blocks that can only be passed after the
payment of some kind of duty, as well as all kinds of permits, licences, etc.
that are necessary in order to move goods, persons or money, should be
avoided to maximise opportunities for livelihood diversification over space.
For similar reasons, measures to counter urban-directed migration by, for
instance, limiting services to non-registered migrants, should be avoided. New
urban dwellers may contribute significantly to urban growth by expanding
the labour supply and entrepreneurial talent, especially in small towns, while
in some areas also acting as a safety valve against a surplus rural population.
Second, there is a need for sensitive state intervention into frontier settle-
ment fostered by the inclusion of land and resources for new purposes. The
resource-rich frontiers in our study (coffee, cocoa, and mining) have illumin-
ated various difficulties in managing space for different types of livelihood.
However, state-induced resettlement schemes with elements of the forced
movement of poor people are inadvisable, even though they can increase
mobility for groups who otherwise may be stuck in poverty. As experience has
shown, these schemes may be detrimental to indigenous population groups
and the environment. This does not rule out the introduction of carefully
planned state-sponsored re-settlement schemes if movement is facilitated for
those in need and willing to move. However, technical and financial support
should be on a continuous basis, extending beyond the initial phases of
the establishment of settlements to provide a cushion against fluctuations
in world market prices and different periods for taking up residence and
production. In addition, wide variations in agro-environmental conditions
should be reflected in differing degrees of state-financed support. These
concerns for state regulation and support are also valid for mining areas,
although usually they are not part of state-initiated development schemes.
Both small-scale and large-scale mining leads to competition for land and
often to the dispossession and poverty of indigenous households.
Third, as illuminated by all the frontier chapters, it is problematic to dis-
tinguish rigidly between what is ‘rural’ and what is ‘urban’ as they are inter-
linked and have fuzzy boundaries. Settlements are difficult to categorise as
either villages or towns and within these categories there are wide variations.
What matters is not the category of the settlement but its ability to perform as
a local growth centre and its potential to develop into a regional growth pole.
As documented in a number of the frontier studies, these local centres have a
significant development impact in terms of employment creation, skills
development, and the local diversification of activities. They also take some
of the pressure off the mega-cities by re-directing urban migratory flows.
These benefits should be acknowledged by a corresponding expansion of the
204 Niels Fold, Katherine V. Gough and Jytte Agergaard
urban physical infrastructure including electricity, piped water, and sanitary
facilities, etc. Also, measures to enhance and promote the capacity of local
enterprises to network and sell their goods within a national context
should be promoted so that these settlements act as mediation points for the
integration of the domestic market. The location of public-service provisions
would further strengthen the impact of local growth centres that are based on
commercial vitality instead of spatial allocations determined by bureaucratic
paperwork or opaque political deals.
Fourth, the interlinked nature of rural and urban areas requires an
approach to the design and implementation of policy mechanisms that
breaks with the traditional separation into sector policies. Agricultural
policies must go beyond narrow foci, like productivity gains or rural labour
retention, and take into account their impact on settlement dynamics and
broader regional growth patterns, including emerging regional inequalities.
Correspondingly, industrialisation strategies and physical planning must be
concerned with the impacts on agricultural production and possibilities for
diversification into non-agricultural economic activities in predominantly
rural areas in addition to small-town development. This form of cross-sectoral
or ‘horizontal’ approach to policies concerned with regional development
should be combined with a ‘vertical’ approach where public institutions
operating on different administrative levels exchange and coordinate their
ideas, activities and plans. Public institutions at lower levels are most
appropriate for drawing up policies that are intimately related to a particular
regional context, but there is a strong need for support and assistance from
institutions at the national level.
Finally, these brief reflections on the implications of the findings of this
book for rural–urban dynamics stress the need for a contextual and flexible
policy framework that acknowledges the importance of non-metropolitan
settlement dynamics. Policy mechanisms should be ‘pro-growth’ by creating
and expanding mobility opportunities for households and individuals as well
as stimulating development in settlements and regions with documented
growth potential. However, this approach does not render policies superflu-
ous that seek to regulate market dynamics if poor households, marginalised
groups or whole regions are at risk due to, for instance, excessive dependence
on the world market for a single product. Furthermore, policies have to
take into account the ways in which links to world markets are determined
by the nature of the dominant products and the relations of production
resulting in divergent impacts within regions and among settlements. In
short, there is a strong case for spatially targeted policies at both regional and
local levels.
This view contrasts with the main argument in the 2009 World Development
Report, which as a general precept recommends ‘spatially blind’ and universal
institutions as a first policy choice, infrastructure as second choice while
specific incentives (spatially targeted interventions) to stimulate economic
activity in agglomerations are only third choice (World Bank, 2009). What we
 Conclusion 205
have shown in this book, is that economic development takes place not
exclusively in agglomerations but also at various locations along the rural–
urban continuum. Hence, we would argue that it is detrimental to overall
development in the Global South if a ‘new economic geography’ means that
these dynamic regions are to be bypassed by national (and international)
policies (Bryceson et al., 2009; Rigg et al., 2009).

References
Bryceson, D.F., Gough, K.V., Rigg, J. and Agergaard, J. (2009) ‘Critical commentary:
The World Development Report 2009’, Urban Studies, 46(4): 723–38.
Rigg, J., Bebbington, T., Gough, K.V., Bryceson, D.F., Tacoli, C., Agergaard, J. and
Fold, N. (2009) ‘The World Development Report (2009) “reshapes economic
geography”: Geographical reflections’, Transactions of the Institute of British
Geographers, 34(2): 128–36.
World Bank (2009) World Development Report 2009: Reshaping economic geography,
Washington, DC: World Bank.
Index

abusa system 64 Asia 4, 134; rural–urban dynamics

Accra–Takoradi railway line 66–7 201–2; see also Thailand, Vietnam
actor-oriented approach 14–15 Ayuthaya 135, 138, 140
Adey, P. 16
Adieye 177–86 bamboo crafts 119
administrative approach 11–12 Ban Khan Haam 135–48
Africa 4, 190; decentralisation Ban Khokmayom 135–48
programmes 38; rural–urban Ban Mor 118–22; mango wood-craft
dynamics 201–2; see also Ghana, 118, 119–20, 123–5, 128, 152–3
Tanzania Ban Pasakluang 118–22; saa paper
age: migration and Tanzania’s mining industry 118, 121, 123, 125–7, 128,
frontier 168–70; and occupation in 152–3
Thai villages 129–30 bananas 60, 75
agricultural frontiers 4–5, 6, 90–8, Barbier, E.B. 3
198–9; cocoa 6, 23–40, 66, 90–8 Bigsten, A. 193
passim; coffee 6, 41–58, 90–8 passim, Binh Minh industrial zone 80
202; fruit 6, 74–89, 90–8 passim; Blue Skies 70
livelihood transformations 96; Blunt, A. 17
markets 91–4; pineapples 6, 59–73, Bodi 26–8, 28–30, 33–7
90–8 passim, 201; settlements’ Bonsu Nkwanta 26–8, 29, 30–2, 33–7,
trajectories 94–6 38
agriculture: alongside craft production in boom-and-bust 3
Vietnam 107–8; GAP production Borras, S.M. 141
systems in Vietnam 82; Ghana’s Bowrie, K.A. 117
mining frontier and 181–2; mining Bramall, C. 155
and 194; policy implications 204; Brazil 42
post-colonial Africa 190; rice Buon Ma Thout 46, 47, 48, 49
farming in Thailand 118–19, 120, Burma 125
121–2, 137–8; sectoral change on buyer-driven chains 13
Thailand’s factory frontier 140,
141–2, 143–4; Tanzania’s mining Carney, D. 15
frontier and 163–4, 166–7, 168, 171, cashews 54
172; virtuous cycle of farm–nonfarm category managers 61
relations 151–4 Chachage, C.S.L. 173
American development, frontier thesis Chambers, R. 14, 15
of 2 Chatthip Nartsupha 137
Amoaya 26–8, 29, 30, 31, 33–7 chemical dyes 125, 126
Anglogold-Ashanti 179 Chiang Mai 154
Anh, D.N. 113 China 75, 155, 190
 Index 207
churning machine 125, 126 DiGregorio, M. 100–1
citrus fruits 75 dissemination 20–1
claim owners 160, 171, 191 dissociation 149; villages in Thailand
cocoa 6, 23–40, 66, 90–8 passim; diverse 139, 140–4, 147–8
trajectories of frontier settlement diversification 94–5; income see income
development 33–7; history in Ghana diversification
23–5; small towns and rural division of labour: handicrafts frontier
settlements 25–32 124, 153; mining frontier 165–7
Cocoa Marketing Company (CMC) 24, Doi Moi policies 41, 42, 74–5, 99, 154
25 Dole 62
Coe, N. 13 domestic markets, for agricultural
coffee 6, 41–58, 90–8 passim, 202; products 94
coffee sector in Vietnam 42–4; dormitories 124, 140
livelihoods in the coffee settlements drug abuse 180
51–4; migration and mobility 44–5,
51, 54–5; settlement formation in the Ea Nam commune 46, 47, 48, 49, 51,
coffee belt 46–51 51–5
coffee crisis 43–4, 53 Ea Sien commune 46, 47, 48, 49, 50–1,
common methodology 18–20 51–5
comparative approach 17–18 Ea Tu commune 46, 47, 48–9, 51–5
continental rural–urban dynamics 201–2 earthenware pots 119
Conway, G. 14, 15 education 83, 144
co-operatives 103 Effuanta 186
corporate social responsibility (CSR) embroidery 101, 103, 103–5, 107–13,
185, 194 152–3
Costa Rica 62 entrepreneurs 105, 124, 126, 129
Côte d’Ivoire 24, 62, 63 European Union (EU) 60–1, 62–3, 70
craft villages: Vietnam 6, 99–115, 151–7 exporters’ outstanding payments 70
passim; Thailand 6–7, 116–33, 151–7 extension services 52–3
passim
crop diversification 54 factories: saa paper 126, 127; Thailand’s
cross-sectoral policy approach 204 rice-growing region 7, 134–50, 151–7
customary land tenure 64 passim
familial ties, migration via 193
daily mobility: Thailand’s factory ‘Far Eastern’ company 106
frontier 145–6, 148; Vietnam 55 Farmers’ Unions 52
Dak Lak Province, Vietnam 41–58; financial support 203
immigration and resettlement financial transfers/flows 10
44–5 focus groups 18, 19
Dar es Salaam 159 food: crops on Vietnam’s coffee frontier
data analysis 20–1 52; production on mining frontier in
de Bruijn, M. 16 Tanzania 166–7; Thai factory villages
de Haan, L. 15–16 144
de-agrarianisation 137–8, 140–4 foreigners, marriage to 83–4
decentralisation: programmes in Africa forward selling 25
38; Tanzania 159 Fresh Del Monte 62
demand for fruit, in Vietnam 74–6 frontiers 1; comparison of frontier types
demographic safety valve 3 17; defining 2; frontier regions 2–5;
Dercon, S. 146 rural–urban dynamics 198–9
development communes 50–1 fruit: pineapples 6, 59–73, 90–8 passim,
diamond mining 7, 158, 161–4, 164–73, 201; Vietnam 6, 74–89, 90–8 passim
189–97 passim
diamond smuggling 162 GAP (good agricultural practice)
diggers 191–2 production systems 82, 88
208 Index
gender: and handicraft production 108, illegal small-scale mining 184, 192
109; migration and mining frontier in-depth interviews 18, 20
168–70, 193 in situ urbanisation: Tanzania 167–70;
Gereffi, G. 13 Vietnam 85–7
Ghana 4, 5, 201, 202; cocoa 6, 23–40, income diversification 96; Vietnam’s
90–8 passim; Economic Recovery coffee belt 53–4; Vietnam’s Mekong
Programme 176; gold mining 7, Delta 81–2
175–88, 189–97 passim; pineapples 6, indigenous settlements 28–30
59–73, 90–8 passim, 201; structural Indonesia 134
adjustment programmes 24, 176 industrial zone plan 111–12
Ghana-Australian Goldfields (GAG) industrialisation strategies 204
179 inequality 202; Thailand’s handicrafts
Ghana Cocoa Board (COCOBOD) frontier 128–9; Vietnam’s fruit
24–5, 30, 33, 38 frontier 81–2; Vietnams’s handicrafts
global markets 60–3, 75, 91–4 frontier 112–13
global production networks (GPNs) 13, information flows 10
14 infrastructure 31, 155, 185, 194, 204
global value chain (GVC) analysis Ivory Coast 24, 62, 63
12–14, 16, 91
globalisation 1, 11 Johannesburg 196
Gold Coast 24; see also Ghana Juaboso District, Ghana 26–37
gold mining: Ghana 7, 175–88, 189–97
passim; Tanzania 7, 158, 160–1, Kefass 26–8, 29, 32, 33–7
164–73, 189–97 passim; trends in the key informants, interviews with 18, 19
Ghanaian gold mining industry
176–7 lacquer mini-zones 111–12
governance forms 13 lacquerware handicrafts 101, 103,
105–6, 107, 107–13, 152–3
Ha, D.T. 3, 42, 43 Lamphun 131
Ha Thai 101, 102, 103, 105–6, 107–13 land: access to 34–5, 37, 64, 95–6; de-
Ha Tay Province 101 linking from its central role in
handicraft and manufacturing frontier Thailand 141–2; holdings in Tanzania
4–5, 6–7, 151–7, 199; forces shaping 167, 168; landlessness in Vietnam 76;
rural and urban spaces 154–6; prices 138; shortage on Tanzania’s
Thailand’s factory frontier 7, 134–50, mining frontier 163–4, 165, 167, 168,
151–7 passim; Thailand’s handicrafts 171
frontier 6–7, 116–33, 151–7 passim, Laos 125
198; Vietnam’s handicrafts frontier 6, large-scale mining 189–97 passim;
99–115, 151–7 passim; virtuous cycle diamonds 161–2, 163, 171, 190; gold
of farm–nonfarm relations 151–4 176–7, 182–3, 185, 189, 190, 192, 194;
Hannam, K. 16 livelihoods 190–2; rural–urban
Hart, G. 152 transitions 194–5
health centres 30, 67 laterite feeder road 137
health and safety 111 legislation affecting mining 190–1
hinterland areas 200–1, 202 Lerise, F. 11
Hoa 78, 79–80, 81–8 licensed buying companies (LBCs) 24–5,
Hoa Hung commune 79–80 30, 32, 33
horizontal approach to policy 204 linkages, rural–urban 9–12
household goods 112, 113 livelihood and economy approach 11–12
household studies 14–15 livelihood trajectories 15–16
housing: coffee frontier in Vietnam 49, livelihoods 7, 14–16, 96; changes on
50; handicraft and manufacturing Thailand’s factory frontier 138–9,
frontier 110–11, 155; mining frontier 141–4; coffee frontier in Vietnam
179 51–4; diversification 6, 201–2;
 Index 209
Ghanaian pineapple frontier 64–71; mining frontiers 4–5, 7, 189–97, 199;
handicrafts in Vietnam 107–9; mining Ghana 7, 175–88, 189–97 passim;
frontiers 165–7, 180–4, 190–2 livelihoods 165–7, 180–4, 190–2;
livestock production 53, 54 rural–urban transitions 194–6;
local economy 11; handicraft/ Tanzania 7, 158–74, 189–97 passim;
manufacturing activity and 153–4; urbanisation and migration 192–4
mining’s linkages to 195 mobility 16–17, 96, 202; Ghana’s cocoa
local governments 11 frontier 36–7, 37; handicrafts in
local growth centres 203–4 Vietnam 109–10; mining frontier 196;
local traders see traders/trading policy implications 202–3; Thailand’s
localism 117, 131 factory frontier 138–9, 144–7, 148;
location 33–4, 37, 94 Vietnam’s coffee belt 54–5; Vietnam’s
Long, N. 14 Mekong Delta 83–5; see also
Lynch, K. 11 migration
multiplier effects 172
Ma, L.J.C. 155 multi-spatial households 10
macro-economic policies 11 Mwadui Williamson diamond mine
Maganzo 158, 161–4, 164–73 161–2, 163, 171, 190
Malaysia 134 My Hoa commune 80
mango wood-craft 118, 119–20, 123–5, My Thoi 1 settlement 78, 79, 80–1, 81–8
128, 152–3
mangoes 75, 77, 78, 79–80, 86 Na Ayuthaya 116
manufacturing 83; see also handicraft national development, rural–urban
and manufacturing frontier dynamics and 199–201
market chains 85–6 New Atuabo 186
markets 6, 12–14; domestic for New Economic Zone (NEZ)
agricultural products 94; fruit markets programmes 44
in Vietnam 86, 87; global for ‘new mobilities’ paradigm 16–17
agricultural products 60–3, 75, 91–4; Nijman, J. 17
handicrafts 105, 106 Nyarugusu 158, 160–1, 164–73
mature pioneer settlement 50 Nyerere, J. 160, 162
MD2 pineapple 59, 62–3, 69, 70
mechanisation 119 Obom 65–71
Mekong Delta 6, 74–89, 90–8 passim; occupational status 129–30
income diversification 81–2; OTOP (One Tambon, One Product)
migration, mobility and remittances policy 120
83–5; settlements 78–81; traders and Owusu, G. 25–6
rural development 85–7 Oxfam 46, 57
Melhuus, M. 17
methodology 18–20 Parnwell, M.J.G. 117
middlemen 105 pepper 54
migration 3, 10, 155; Ghana’s pineapple peri-urban settlement 48–9
belt 66, 67–8, 70; migrant settlements Philippines 134
on Ghana’s cocoa frontier 30–2, pieceworkers 127, 129
33–7; mining frontier 164–5, 185, pineapples 6, 59–73, 90–8 passim, 201;
192–4; Thailand’s factory frontier boom years 68–9; dynamics in the
137, 140–1, 144–5; Thailand’s global market 60–3; Ghanaian
handicraft frontier 124–5; Vietnam’s pineapple frontier 64–71; post-boom
coffee belt 44–5, 51, 54–5; Vietnam’s phase 69–71
Mekong Delta 83–5; see also pioneers 2; mature pioneer settlement
mobility 50
Mikesell, M.W. 2 pit holders 160, 191
milk fruit 77, 78, 78–9, 86 planned resettlement programmes 44–5,
mine owners 191–2 203
210 Index
plantations, pineapple 68–9, 70 development paradigms 127–30;
Pokrom Nsaba 65–71 Vietnam’s fruit frontier 85–7
policy implications 202–5 rural poverty-alleviation programmes
policy seminars 20–1 117
pollution 111, 112 rural–urban dynamics 6, 9–22, 198–205;
pomelo 77, 78, 80, 81–2, 86 common methodology 18–20;
population growth 26 comparative approach 17–18;
por saa (paper mulberry tree) 125 conceptualising 9–17; data analysis
pots, earthenware 119 and dissemination 20–1; in frontiers
poverty: mining frontier and 184–6, 196; 198–9; livelihoods 14–16; markets
Thailand 128–9 12–14; mobility 16–17; at national and
power 15–16 continental levels 199–202;
Precious Minerals Marketing Company performing comparison 17–21; policy
(PMMC) 184 implications 202–5
preferred suppliers 61 rural–urban linkages 9–12
prices: coffee 43; fixed price for cocoa in rural urbanisation 100–1, 114
Ghana 24, 25; fluctuations 3
Primary Mining Licences (PMLs) 160, saa (mulberry) paper production 118,
173, 191 121, 123, 125–7, 128, 152–3
processing technologies 92–3 sand tailings 162, 173
Produce Buying Company (PBC) 24, sedentarisation programme 44–5
38 self-employment 182
producer-driven chains 13 semi-structured interviews 18, 19
pro-growth policy mechanisms 204 services: agricultural frontier 32, 35–6,
prostitution 180 86, 95; connected to mining 172, 195;
public services 32, 172, 204 services sector in Vietnam 83
settlement cycle 193
Quang Hiep commune 46, 47, 48, 49, settlements 3–4, 199; formation and
50, 51–5 growth on mining frontiers 193–4,
Quat Dong 101, 102, 103–5, 107–13 196, 199; Ghana’s cocoa frontier
questionnaire survey 18, 19–20 25–32; Ghana’s pineapple frontier
64–71; gold mining in Ghana 177–80;
railways 66–7, 177 selection for study 18–19; Tanzania’s
recreational services 172 mining frontier 159–64, 170–2;
Red River Delta 6, 76, 99–115, 151–7 Thailand’s factory frontier 135–8;
passim; changing nature of Thailand’s handicrafts frontier
handicraft production 101–6; 118–22; trajectories on agricultural
craft villages 100–1; livelihoods frontier 33–7, 94–6; urban momentum
107–9; mobility 109–10; settlement of frontier settlement growth 170–2;
110–13 Vietnam’s coffee belt 45–51;
regional development policies 154 Vietnam’s handicraft frontier 110–13;
remittances 10; Ghana 185; Vietnam Vietnam’s Mekong Delta 78–81
83, 84–5, 109–10 Shen, Xiaoping 155
resettlement schemes 44–5, 203 Shinyanga 172
resource-rich areas 200–1, 202, 203 Shively, G. 3, 42, 43
retailing 60 skilled/semi-skilled work, shift to 142–3
rice farming 52; Thailand 118–19, 120, small-scale gold-buying firms 183–4
121–2, 137–8 small-scale mining 172, 189–97 passim;
Rigg, J. 12, 16–17 diamonds 161–4, 191–2; gold in
‘Robusta’ coffee 42 Ghana 176–7, 183, 185, 190, 192; gold
Rojana Industrial Park 136–7, 140, in Tanzania 160–1, 191, 192;
145 livelihoods 190–2; rural–urban
rural development: on Thailand’s dynamics 194–5; settlement growth
handicrafts frontier and rural 193–4
 Index 211
small towns: development on Ghana’s tourism 105, 106
cocoa frontier 6, 25–38; Vietnam’s traders/trading: agricultural frontiers
fruit frontier 85–7 31–2, 85–7, 94; mining frontiers 166,
smallholders: and agricultural markets 180–1
93–4; coffee production 42, 43, 48, 51, traditional frontiers 91
53, 56; pineapple production 59–73; transnational companies (TNCs) 91
producer groups 92 transport 55, 145–6
Smooth Cayenne pineapple 59, 62–3, 68, turbulence 138–47
69, 70 Turner, F.J. 2
social and cultural approach 11–12
social networks 128 unemployment 184–6
space: forces shaping rural and urban United Kingdom Department for
spaces on handicraft and International Development (DFID)
manufacturing frontier 154–6; spatial 15
disaggregation 128; spatial processes United States pineapple plantation
127–8 companies 62–3
standards for agricultural products 61, unskilled labour 184
92 upgrading 13, 14
state: and agricultural markets 93; policy urban ‘becoming’ 155
implications 202–5; state strategy and urbanisation: in situ 85–7, 167–70;
frontier development 3 mining frontier 167–70, 192–4; rural
state-owned enterprises (SOEs) 43 100–1, 114
‘stretched’ settlement 50–1
structural adjustment programmes 24, vertical approach to policy 204
176 Vietnam 4–5, 134, 202; centralised
sub-contractors 124, 126–7 planning 202; coffee frontier 6,
sub-culture, mining settlements’ 41–58, 90–8 passim, 202; coffee
195–6 sector 42–4; Doi Moi policies 41,
sufficiency economy 127 42, 74–5, 99, 154; fruit production in
supermarkets 60–1, 91–2, 144 the Mekong Delta 6, 74–89, 90–8
supply chains 61 passim; fruit sector 74–6; handicraft
survival strategies 15 frontier 6, 99–115, 151–7 passim;
Ministry of Agriculture and Rural
Tacoli, C. 9–10, 17, 151 Development 99; Ministry of
Tambon Khan Haam 135–7, 140–1, Industry 101
144 Village Chiang Mai 116
Tanzania 4, 5, 7, 158–74, 189–97 passim; Vinh Kim commune
Mining Act 1979 173; Mining Act 78–9
1998 173; Small Scale Mining Vinh Long Province 76–8, 80–1, 81–8
Policy Paper 173; urbanisation trends Vinh Thoi 78–9, 81–8
159 virtuous cycle of farm–nonfarm
Tarkwa 177–86 relations 151–4
teak-based crafts 119–20, 123
Teberebie 177–86 wage labouring 53–4, 70, 119, 128–9
Teberebie Goldfields (TG) Ltd 179 Ward, K. 17
technical support 203 Wassa West District, Ghana 175–88,
textile factories 138, 140 189
Thailand 4, 5; factory frontier 7, 134–50, water shortage, on Tanzanian mining
151–7 passim; handicraft frontier 6–7, frontier 163, 170, 171
116–33, 151–7 passim, 198; regional ‘wet coffee’ 43
development policies 154 Williamson diamond mine 161–2, 163,
Thompson, E.C. 12, 147, 149, 155 171, 190
Tien Giang Province 76–8, 78–80, wood-based handicrafts 119–20; see also
81–8 mango wood-craft
212 Index
workshops 18 yields, coffee 52, 53
World Bank 116 young people: Ghana’s mining frontier
World Development Report 2009 154, 184–5; handicraft production 108–9
204–5
World Trade Organization (WTO) 75 Zoomers, A. 15–16