SME Cluster Development A Dynamic View of Survival Clusters in Developing Countries
Mario Davide Parrilli
SME Cluster...
14 downloads
512 Views
525KB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
SME Cluster Development A Dynamic View of Survival Clusters in Developing Countries
Mario Davide Parrilli
SME Cluster Development
This page intentionally left blank
SME Cluster Development A Dynamic View of Survival Clusters in Developing Countries Mario Davide Parrilli
© Mario Davide Parrilli
2007
All rights reserved. No reproduction, copy or transmission of this publication may be made without written permission. No paragraph of this publication may be reproduced, copied or transmitted save with written permission or in accordance with the provisions of the Copyright, Designs and Patents Act 1988, or under the terms of any licence permitting limited copying issued by the Copyright Licensing Agency, 90 Tottenham Court Road, London W1T 4LP. Any person who does any unauthorised act in relation to this publication may be liable to criminal prosecution and civil claims for damages. The author has asserted his right to be identified as the author of this work in accordance with the Copyright, Designs and Patents Act 1988. First published 2007 by PALGRAVE MACMILLAN Houndmills, Basingstoke, Hampshire RG21 6XS and 175 Fifth Avenue, New York, N.Y. 10010 Companies and representatives throughout the world PALGRAVE MACMILLAN is the global academic imprint of the Palgrave Macmillan division of St. Martin’s Press, LLC and of Palgrave Macmillan Ltd. Macmillan® is a registered trademark in the United States, United Kingdom and other countries. Palgrave is a registered trademark in the European Union and other countries. ISBN-13: 978–0–230–00797–0 ISBN-10: 0–230–00797–X This book is printed on paper suitable for recycling and made from fully managed and sustained forest sources. A catalogue record for this book is available from the British Library. Library of Congress Cataloging-in-Publication Data Parrilli, Mario Davide. SME cluster development : a dynamic view of survival clusters in developing countries / by Mario Davide Parrilli. p. cm. Includes bibliographical references and index. ISBN-13: 978–0–230–00797–0 ISBN-10: 0–230–00797–X 1. Small business—Developing countries. 2. Industrial clusters— Developing countries. 3. Small business—Latin America. 4. Industrial clusters—Latin America. 5. Small business—Italy. 6. Industrial clusters— Italy. I. Title. HD2346.5.P37 2007 338.6⬘042091724—dc22 10 16
9 15
8 14
7 13
6 12
2006050506 5 11
4 10
3 09
2 08
1 07
Printed and bound in Great Britain by Antony Rowe Ltd, Chippenham and Eastbourne
To Nadya
‘Inflation became the nemesis of the Keynesian Consensus, and the control of inflation the icon of the Washington Consensus. In its turn the nemesis of the Washington Consensus, which has become visible in recent years, has been the increased income inequalities between and within countries involved in globalization and free markets. As a result the pendulum is beginning to swing away from the Washington Consensus, although not necessarily directly back to a new Keynesian Consensus …’ Hans W. Singer (Editorial: The Golden Age of the Keynesian Consensus – The Pendulum Swings Back, World Development, Elsevier – North Holland, 25 (3), 1997). ‘These speculations on the future of industrial districts may appear rather heretical against the background of their current economic success and their still-euphoric perception in scientific circles. However, they elucidate a potential dilemma concerning the embeddedness of networks. Too little embeddedness may expose networks to an erosion of their supportive tissue of social practices and institutions. Too much embeddedness, however, may promote a petrification of this supportive tissue and, hence, may pervert networks into cohesive coalitions against more radical innovations’. Gernot Grabher (‘Rediscovering the social in the economics of interfirm relations’, in G. Grabher, The Embedded Firm, Routledge 1993).
Contents List of Tables
x
List of Figures
xi
List of Abbreviations
xii
Preface
xv
Acknowledgements
Chapter 1:
xvii
Introduction 1. The SMEs, the fracture within the system and a systemic approach to development 2. The development of SME clusters 3. Empirical explorations in Latin American clusters
PART I:
Small and Medium Enterprise Development
Chapter 2:
Integrating the National Production System: The New Challenge for Chile 1. Introduction 2. A conceptual framework 3. Policy approach in the 1980s and the 1990s 3.1 The explicit policy approach 3.2 Implicit industrial policies 4. Enterprises’ reaction to crisis 5. Interpretations of the slowdown: ECLAC’s critique and the de-industrialisation process 6. The need to set up integrative industrial strategies
Chapter 3:
Inclusion Versus Fragmentation: Different Responses to Liberalisation in European and Latin American Small and Medium Enterprises 1. Introduction 2. Industrial restructuring and small firms in Europe vii
1 2 5 8
11 13 13 14 16 16 18 21 25 26
30 30 32
viii
Contents
3.
4.
PART II: Chapter 4:
Chapter 5:
The main processes in Latin America: liberalisation and industrial destructuring 3.1 Case study: Argentina 3.2 Case study: Chile 3.3 Case study: Nicaragua 3.4 Case study: Costa Rica 3.5 A synthetic view of the Latin American response A Strategic approach to industrial development
Small and Medium Enterprise Cluster Development Different Theoretical Approaches to SME Cluster Development: Relevance in the Case of Nicaragua 1. Introduction 2. The ‘spontaneous’ approach to SME cluster development 3. The ‘policy-inducement’ approach to SME development 4. The social approach 4.1 The social embeddedness of economic action 4.2 Trust in industrial districts and SME clusters 4.3 The social strengths of SME clusters 5. Empirical evidence from Nicaragua 5.1 An introduction to the country 5.2 The main programmes to support SMEs in Nicaragua 5.3 The limits to the division and specialisation of labour and their social roots 5.4 The conflictive political economy in Nicaragua 6. Preliminary conclusions on policy-making in SME clusters A Stage and Eclectic Approach to Industrial District Development: Two Policy Keys for Survival Clusters in Developing Countries 1. Introduction 2. The analysis of development stages
35 37 40 43 46 48 50
55 57 57 59 61 63 63 64 66 69 69 71 74 76 78
82 82 84
Contents
3. 4.
5.
Chapter 6:
Chapter 7:
ix
Summing up the eclectic analytical framework 85 The stages of development of the Italian industrial districts 89 4.1 The passage from rural life to artisanal clusters (1920s–early 1950s) 89 4.2 The passage from craft production to industrial concentration (1950s–1960s) 92 4.3 The passage from industrial concentration to industrial districts (1960s–1960s) 94 4.4 The passage from traditional IDs to new competitive IDs (1980s–2000s) 96 Two policy keys for ‘survival clusters’ in developing countries 101 5.1 Policy key I: the importance of upgrading through stages 101 5.2 Policy key II: the relevance of an eclectic approach 103
Empirical Explorations in Survival Clusters in Central America and in Competitive Italian Industrial Districts 1. Introduction 2. ‘Survival clusters’ in developing countries 3. Empirical evidence from three clusters 3.1 The context 3.2 The district of Forlì, Italy 3.3 The ‘survival clusters’ of Sarchí, Costa Rica, and Masaya, Nicaragua 4. Concluding remarks for policy-making Conclusions 1. SME development 2. SME cluster development
105 105 106 107 107 109 116 124 126 127 132
Notes
136
Bibliography
142
Index
155
List of Tables Table 2.1: Table 2.2: Table 3.1: Table 3.2: Table 3.3: Table 3.4: Table 3.5: Table 3.6: Table 4.1: Table 4.2: Table 5.1: Table 5.2: Table 5.3: Table 5.4: Table 6.1: Table 6.2: Table 6.3: Table 6.4: Table 6.5: Table 6.6: Table 6.7: Table 6.8:
CORFO’s development programmes in 2002 The growth of imports in Chile, 1992–98 (million pesos) EU actions implemented from May 1997 to end of 1999 Sales and employment concentration in major economic groups in 1997 The enterprises in Chile in 1997 and 2000 The structure of the national industry by size, in 1996 Weight of exports versus foreign direct investment ($million) SMEs participation in the manufacturing sector Identified clusters in Nicaragua The strongest sectors of industrial SMEs in Nicaragua, 1996 Stage I: the 1920s to the early 1950s Stage II: the 1950s to the 1960s Stage III: the 1960s to the 1980s Stage IV: the 1980s onwards Mode of some basic features in Forlì, Masaya and Sarchí Number of firms and workers in wood and furniture in the province of Forlì Producers’ perceptions of external economies in Forlì, 2002 Enterprises created by workers in Forlì ‘Innovative’ producers during 2000–02 (%) Duration of work relationships with suppliers, clients and workers Enterprises created by workers in Sarchí and Masaya Duration of work relationships with suppliers, clients and workers
x
19 23 34 40 41 44 45 47 71 75 90 92 94 97 108 112 113 113 114 115 120 121
List of Figures Figure 2.1: Figure 2.2: Figure 2.3: Figure 6.1: Figure 6.2: Figure 6.3:
Recent evolution in the number of enterprises Recent evolution of sales (UF) The recent evolution of firm productivity (UF) Growth rates in the furniture industry in Forlì Likert scale (5–1) of producers’ perceptions on the sources of external economies Innovative producers over two years (%)
xi
22 22 22 110 118 121
List of Abbreviations AESMEC BCC BCCR BCIE BCN BID CAD CCIAA CNC CONAPI CORFO CNA CSIL EC ECLAC ERVET
EU FNI FOGAPE GDP GTZ HDI
Asian-European Conference on Small and Medium Enterprises Banco Central de Chile (Chìle’s Central Bank) Banco Central de Costa Rica (Costa Rica’s Central Bank) Banco CentroAmericano de Integración Económica (Central American Bank for Integration) Banco Central de Nìcaragua (Nicaragua’s Central Bank) Banco Interamericano de Desarrollo (Inter-American Development Bank) Computer-aided design Camera di Commercio, Industria, Artìgìanato ed Agrìcoltura (Chamber of Commerce) Computerised numerical control Cámera Nacìonal de la Pequeña Industria (Nicaragua’s Chamber of Small Industries) Corporación de Fomento de la Produccìón (Chile’s Development Corporation) Confederazione Nazìonale dell’Artìgìanato (Italy’s National Confederation of Artisans) Centro Studi Industria Leggera (Lombardy’s Centre for Study of Light Industry) European Community United Nations Economic Commission for Latin America and the Caribbean Emìlìa-Romagna Valorìzzazìone Economica del Terrìtorìo (Emilia-Romagna’s Agency for Economic Valorisation of the Territory) European Union Financìera Nìcaragüense de Inversìones (Nicaragua’s Investment Fund Agency) Fondo de Garantía para Pequeños Empresas (Chile’s Guarantee Fund for Small Enterprises) Gross domestic product Gesellschaft und Technische Zusammen Arbeit (Germany’s development agency) Human Development Index xii
List of Abbreviations
ICT IDs IMF INE INCAE
INDE INN INPYME
ISI ISTAT MAGFOR MEDE MIFIC
MINSA NGOs PAMIC
PROCHILE PROFOR PROSEDE R&D SMEs SME clusters SERCOTEC
xiii
Information and communication technologies Industrial districts International Monetary Fund Instituto Nacìonal de Estadístìcas (Chile’ National Institute for Statistics) Instìtuto Centroamericano de Admìnìstración de Empresas (Central American Institute for Business Administration) Instituto Nicaragüense de Desarrollo (Nicaragua’s Institute for Business Development) Instituto Nacional de Normalización (Chile’s National Institute for Normalisation) Instituto Nicaragüense de Apoyo a la Pequeña y Mediana Empresa (Nicaragua’s Institute for the Promotion of Small and Medium Enterprises) Import substitution industrialisation Istituto Nazìonale di Statistica (Italy’s National Institute for Statistics) Ministerio Agropecuario y Forestal (Nicaragua’s Ministry of Agriculture and Forestry) Ministerio de Economía y Desarrollo (Nicaragua’s Ministry of Economic Development) Ministerio de Fomento Industria y Comercio (Nicaragua’s Ministry of Development, Industry and Commerce) Ministero de Salud (Costa Rica’s Health Ministry) Non-governmental organisations Programa Nacìonal de Apoyo a la Micro Empresa (Nicaragua’s Programme for the Promotion of Micro Enterprises) Programa de Fomento a las Exportaciones Chilenas (Program for Export Support of Chile) Program on Forests (Multi-donor partnership involving the World Bank) Programa de Servìcìos Empresariales (Programme for Business Development Services of Nicaragua) Research and development Small and medium enterprises Small and medium enterprise clusters Servìcìo de Cooperación Técnìca (Chile’s Technical Cooperation Service)
xiv
List of Abbreviations
UF UNDP UNIDO UNIPYME WTO
Unìdad de Fomento (Units of Development: type of currency used by Chilean government institutions) United Nations Development Programme United Nations Industrial Development Organization Unìón de la Pequeña y Medìana Empresa (Union of Small and Medium Enterprises) World Trade Organization
Preface This volume is the result of a long-lasting reflection on the causes of development and underdevelopment that we have undertaken over the past 15 years and in which we have been working in Latin America and on Latin America. This work has benefited from intense interaction with experts, entrepreneurs, workers, associations and organisations met over recent years and with which we have been developing and implementing several studies, actions and projects that may have an impact on the economic development of this and other developing regions. This book represents the synthesis of these multiple ongoing efforts. Small and medium enterprises are much more than simply economic agents; they are social and cultural conveyors of values, trust, selfrealisation and welfare. They represent a large part of the population worldwide and their desires to realise their own autonomous ideals, objectives and aspirations. Supporting their development process means helping the people to increase their own welfare which may be independent from quantitative economic success. We all know that aggregate and per capita income are not sufficient indicators of development and welfare, because the cost of living varies from country to country; we also know that many factors that are currently excluded from the economic assessment of development, affect the overall welfare of people (e.g., the weather conditions and the friendliness of the people). These and other aspects show that development and welfare are complex concepts that cannot be simplified to a limited range of numeric indicators. Development and welfare are much more than that; they are cultural dimensions. On these bases, the topic of ‘small and medium enterprise clusters’ is treated as one of the most interesting solutions to make small and medium enterprises and entrepreneurs reap benefits from their number, association, critical mass … as well as from their cultural and institutional background. In fact, the topic of clustering has acquired momentum when, a few decades ago, several local production systems, namely the industrial districts in central northern Italy proved to be extremely competitive and welfare-generating. This is why we chose to analyse the development dynamics of these kinds of firm agglomerations; they provide important insights to improve the interactions among economic, social and institutional agents that participate in the generation of higher levels of welfare in many local production systems. xv
xvi
Preface
In developing economies, these local production systems are often lacking some key aspects of development, such as factor endowment, legal status and social embeddedness; this situation produces clusters in which the interaction among individuals, firms and institutions is not particularly rich. In this context, our work reveals several key features in development that are often neglected for a number of reasons. We propose a ‘stage and eclectic approach’ as a way to take into account these determinants and generate a comprehensive strategic decision-making at local and national levels. Through this book, it is hoped to provide useful elements to entrepreneurs, workers, associations and local, national and international institutions which daily work for the promotion of small and medium firms in both industrialised and developing economies, especially those that identify small and medium enterprises as a thorough conveyor of development, conceived not only in economic terms, but more broadly in social, institutional and cultural terms. MDP, SME Birmingham, 2006
Acknowledgements The completion of this volume has benefited meaningfully from academic interaction with a number of special persons and institutions; in particular I am grateful to Roger Sugden for his precious support and comments over the past four years of intense research leading to a PhD at the University of Birmingham. The collaboration with him in the ‘Latin American Graduate School in Industrial and SME Development Policies’ that we set up in Central America in 2000 and in a number of others joint initiatives created many occasions that helped me deepen the analysis that I have carried out over these years. I acknowledge the importance of the Birmingham Business School, University of Birmingham, which sponsored my research. I would like to thank Patrizio Bianchi, at the University of Ferrara, where I lectured and did research during 2001–05. Patrizio’s insights discussed in joint papers, seminars and consultancies were extremely helpful to spur my thinking on related topics. I gratefully acknowledge the excellent comments received by Peter Knorringa, at the Institute of Social Studies in The Hague, and Lisa De Propris, at the University of Birmingham; they helped me improve the overall consistence of this research. I appreciate very much the important comments received from the anonymous referees and the editors that helped me refine the arguments of papers that have been published in the journals Review of International Political Economy (‘Integrating the National Industrial System: the new challenge for Chile’, 11(5)), and European Planning Studies (‘A Stage and Eclectic Approach to Industrial District Development: Two Policy Keys for Survival Clusters in Developing Countries’, 12(8)); these are relevant parts of this volume. A special thanks goes to colleagues and students at Birmingham and Ferrara universities, as well as colleagues and students at the Universidad Centroamericana in Managua, Nicaragua, where I lectured and did research during 1996–2001. I would also like to thank colleagues and students at the Latin American Graduate School in Industrial Development and SME Policies that I have been coordinating in Managua, San José, Tlaxcala and Guatemala City, where issues relevant to this work were and are intensely debated and contributed to the development and completion of this research. I am also grateful to the colleagues that participated in the joint publication ‘High technology, productivity and linkages: a systemic approach xvii
xviii
Acknowledgements
to the development of small firms’, edited by Patrizio Bianchi, Roger Sugden and I and which is part of the EU-ALFA project ‘European-Latin American Network for Research and Learning in Industrial Development Policy’ that I am coordinating. They have been contributing important analyses that helped me deepen specific parts of my own research. Among them I would like to thank Hugo Kantis, of the Universidad Nacional de General Sarmiento in Buenos Aires; Stuart Schweitzer from the University of California in Los Angeles; Pablo Angelelli from the Inter-American Development Bank; Renato García, Joao Furtado and Wilson Suzigan at the Universidade de Sao Paulo; Adolfo Arata at the Universidad Técnica Federico Santa Maria in Valparaiso, Chile; Felix Cadena at the Colegio de Tlaxcala in Mexico; Mari Jose Aranguren at the Deusto University; María Callejón at the Barcelona University; David Audretsch at the Indiana University; Marco Bellandi at the University of Florence; Carlo Pietrobelli at the University of Rome; Roberta Rabellotti at the University of Piemonte Orientale in Novara; and other colleagues with whom I collaborated in consultancies and publications on related subjects and that discussed with me relevant issues in small and medium-sized enterprise development: René Hernández at the Economic Commission of Latin America and the Caribbean (ECLAC) based in Mexico City; Giovanni Stumpo and Wilson Peres at the ECLAC in Santiago, Chile; Maria Cappiello at the agency Nomisma, Italy, and Juan Carlos Scapini at the Ministry of Economy in Chile. I am also grateful to Hubert Schmitz and Raphie Kaplinsky, at the Institute of Development Studies of Brighton, Sussex, that in the mid 1990s led me to appreciate and study the attractive topic of SME cluster development. Last, but first of all, I am happy to acknowledge the key role that over this long period of work and research across different countries and culture have played my beloved parents, wife and children. Without their support this interesting and important research would not have produced the same positive outcome.
1 Introduction
In the present world of globalisation, small and medium enterprises (SMEs) are key actors in almost every production system. In all continents they represent the largest number of firms, and also contribute significantly to both employment generation and GDP formation (AESMEC, 1998; Observatory of European SMEs, 2002/2; Peres and Stumpo, 2002). This volume focuses on the development of small and medium enterprises and small and medium enterprise clusters (SME clusters). Its objective is to propose some relevant suggestions for the development of SMEs and of a special kind of SME cluster that is found in Latin America and, more generally, in developing countries: ‘survival clusters’ (Altenburg and Meyer-Stamer, 1999; Knorringa, 2002). These clusters are formed by micro and small craft firms working with obsolete technology and manual techniques to produce, with no division and specialisation of labour, low-quality non-standardised goods for low-income consumers in local markets. Overall, these agglomerations of firms have little margin for capital accumulation and economic growth (Altenburg and Meyer-Stamer, 1999). The identification of this category of SME clusters is relevant for theoretical and practical purposes because they frequently populate the economy of developing countries. In our view, they can be treated as a general category that includes more and less dynamic clusters, the first involving embryonic kinds of clusters, endowed with development potential, and the second involving very static kinds of agglomerations, often based in subsistence traditional services. Considering relevant elements from the history of SMEs and SME clusters in industrialised countries (i.e., Italian industrial districts), we address key issues that may have a significant meaning for development economics and industrial development policy in the case of SMEs and of the most dynamic among the whole category of survival clusters in developing countries. 1
2
SME Cluster Development
Through this research, some important dynamics that occur in the growth of SMEs and SME clusters are investigated. The focus in Part I of this work, is on the development of SMEs in Latin America Part II analyses the theoretical and policy approach and the trajectory that SME clusters could follow in order to upgrade from lower to higher development levels. As a first objective of Part II, this research attempts to identify the possible stages that SME clusters may pass through so as to realise a trajectory of development. The important issue here is not whether the evolutionary stages are those followed in other contexts or they are brand new; what is at stake here is that ‘feasible stages’ are targeted and that comparisons with too distant models are avoided. In this way, we build up a theoretical framework based upon the identification of different stages that clusters may go through. This framework could help policy-makers identify in which stage of development SME clusters are and what policy schemes could be applied more successfully to less dynamic contexts. This aspect is particularly important for some types of SME clusters in developing countries, in particular the above-mentioned survival clusters, and, within this category, those clusters that show more potential for effective growth.1 Simultaneously, a second target of this study relates to the identification of relevant complementary factors that, in an ‘eclectic’ theoretical and policy framework, could contribute to the growth process of these local production systems. This additional target identifies a number of specific economic, social and policy drivers as essential elements of an integrated approach to the development of SME clusters.
1. The SMEs, the fracture within the system and a systemic approach to development Before targeting the specific topic of SME cluster development, it is necessary to make a more general reference to the wider topic of small and medium enterprise development that represents the economic agents that form those clusters. This is particularly important nowadays, because of the big problems that the majority of these enterprises face in accessing production inputs, technology and finance, human and social capital, policy support within their own markets. These problems, which have been exaggerated by the trend to globalisation and market liberalisation of the past two decades, end up jeopardising the present and future competitive capacity of SMEs as well as the distribution of welfare among the population in less developed economies. This does not mean that policy matters only at the macroeconomic level, but because also at the local level relevant policies can be adopted and combined with
Introduction
3
national policies in order to create a more purposive environment for SME development. Small and medium enterprises represent a very significant collective actor of development in all geographical contexts. They usually include more than 90 percent of the total number of enterprises and between 30 percent and 50 percent of total employment and sales (Peres and Stumpo, 2002; Observatoire des PME Européennes, 2000; AESMEC, 1998). Other reasons more linked to SME potential to promote social development, also explain the importance of these economic agents. In fact, under specific conditions (e.g., steady GDP and employment growth, cooperation and distribution mechanism, and so on), SMEs and their local production systems show a high potential to promote economic democracy and collective welfare. This depends on the more horizontal relations that grow in contexts that are populated by these rather homogeneous economic and social agents. In some cases, SMEs have been able to implement their ‘potential’ of promoting economic democracy.2 This occurred in well-known cases such as the Italian industrial districts and other developed country clusters, such as those in Baden-Württenberg in Germany; southwest Flanders in Belgium, southeast Jutland in Denmark, etc. (Schmitz and Musick, 1994). In addition, this social and economic homogeneity tends to stimulate a higher participation of these agents (particularly SME entrepreneurs) in the process of expressing their preferences (e.g., in elections, informal meetings, private business organisations, hobby associations, etc.) and of pushing the local political system to be accountable to the needs of the local society (Schumacher, 1973; Bagnasco, 1988; Putnam, 1993; Locke, 1995; Cowling and Sugden, 1999; Becattini, 2000a). These important social and economic features motivate our interest in these actors (SMEs) and their potential for economic development in developing countries. Within these countries, this sector (SMEs) faces a new relevant problem, which is affecting the prospects of growth of their national production systems. This problem refers to the fracture that is growing within production systems. This leads to the ex-post exclusion of a large part of the population from the overall development process. This type of exclusion is different from the ex-ante exclusion produced by the traditional division of society in classes that was taking place in developing countries.3 This ex-post type of exclusion is the result of market dynamics and market failures (e.g., problems in SME access to the financial system and the transfer of more advanced technology; accessibility problems to wider international markets; adequate infrastructures for SME development in rural areas, etc.). Our hypothesis is that these difficulties
4
SME Cluster Development
can be modified through appropriate industrial development policies (see Chapter 3) and complemented with proactive and wider local policies. Anticipating the answer that is given in the final sections of Chapter 3, ‘linkage policies’ are among the most meaningful ways to overcome such a problem (Bianchi et al., 2000; Bianchi and Parrilli, 2002b). Within these policies, the development of clusters (i.e., through cluster policies) is particularly valuable. This topic is analysed in Part II of this book. The study of relevant Latin American economies highlights that, during the past 20 years of market liberalisation, their production system tended to split in two separate parts: on the one hand, a small sector of large, medium and some small modern enterprises capable of adopting the best international production practices, employing the most skilled engineers and workers of the country and orienting its commercialisation towards the international market; and on the other hand, a large sector of traditional micro, small and some medium enterprises, usually adopting obsolete machinery, employing unskilled and cheap labour and orienting their traditional goods and services mainly towards the local and national markets, with special reference to their low-income segments of consumers (Bianchi and Parrilli, 2002b). In Part I, Chapter 2 and Chapter 3 are included and four country case studies are analysed in order to verify the effective process of fragmentation that these national production systems face. In particular, Chapter 2 details and discusses the case of Chile as an analytical case that clarifies the width of the problem at stake and the way this has been evolving and mounting over time. This analysis points to the need to structure a policy framework that helps to solve the problem of the growing fracture within the national production system. In general, the policy solutions that have been pursued up to now (mainly macroeconomic stabilisation policies: World Bank, 2002; Peres and Stumpo, 2000) do not seem to respond to the needs of SMEs, because those policies have not addressed the specific problem of the fracture that has arisen within the national production system during the recent market liberalisation era. In Chapter 3, the cases of Argentina, Nicaragua, Costa Rica and again Chile are analysed, in order to show the way the above-mentioned process of fragmentation has taken place in these countries. This may be explained on the basis of the higher conditions required by globalised and liberalised markets to allow traditional SMEs4 a wider access to finance (e.g., higher guarantees), advanced technology (e.g., financial capital and skilled human capital), and international markets (e.g., again skilled human capital and finance). These missing conditions for SME development implicitly point to the concept of ‘market failures’, which constantly affect the
Introduction
5
growth of these economic agents (SMEs) in developing countries.5 In this chapter the lessons that can be drawn from relevant international experiences (mainly the European experience) are also presented. With a view to solving the problem of a fracture within the national production system, we propose a strategic approach to industrial development, which emphasises the need to create strong production linkages between modern and traditional enterprises (Bianchi et al., 2000; Bianchi and Parrilli, 2002a; 2002b; Parrilli, 2004a). In particular, Chapter 3 tries to identify those industrial development policies that can push SMEs to overcome this negative fracture and develop as a whole (particularly in section 3.5). The potential of geographical and sectoral concentrations of SMEs, otherwise called SME clusters, is identified as one of the most interesting ways to achieve this objective (among others, such as the promotion of subcontracting between modern large firms and small traditional firms); of course once appropriate conditions for development are verified. Thus, the necessary systemic approach to SME development in the present framework of globalisation should include the clustering of firms and relevant cluster policies among its most significant industrial development strategies. This consideration links Part I of this work (on SME development) to the more specific Part II (on clustering and SME cluster development). In this latter, the analysis does neither neglect the findings (i.e., the fracture in the economic system) nor the policy suggestions made in the Part I (i.e., a systemic approach to industrial development policy). Rather, in Part II we complement what has been discussed in Part I by focusing on a more specific collective and territorial actor (i.e., the cluster) that well represents one of the most interesting options for SMEs to become competitive also in developing countries. In this case, we point the attention to what has been done up to now in specific developing countries and to what may be realised to promote more effective results in accordance with the above-mentioned wider macroeconomic system.
2. The development of SME clusters Part II of this volume focuses on the topic of SME cluster development, with special reference to the so-called ‘survival clusters’ (Altenburg and Meyer-Stamer, 1999) and, within these, to those clusters that show the highest potential for growth (i.e., a type of ‘embryonic’ cluster). At the beginning of this part (Chapter 4), we review the existing literature on SME clusters and highlight the key features of the main theoretical approaches to SME cluster development.
6
SME Cluster Development
Although in some cases these approaches tend to overlap, we identify three main approaches.6 The first is an economic approach focused on the concepts of ‘collective efficiency’ and ‘flexible specialisation’, which is based upon the forces of ‘joint actions’ and ‘external economies’ (Schmitz, 1992; 1995a; Pedersen et al., 1994; Nadvi and Schmitz, 1999; van Dijk and Sandee, 2002). The second stream focuses on the strategic role of governance and the policy inducement of development. Within this approach several legal instruments (e.g., laws, programmes) favoured the growth of these competitive local production systems (Best, 1990; Capecchi, 1990; Putnam, 1993; Bianchi, 1998; Cowling and Sugden, 1999; Guerrieri and Pietrobelli, 2004; Humphrey and Schmitz, 2004). Finally, we identify a social approach that expresses the importance of the ‘social embeddedness’ of economic development and the significance of the local social network in promoting a positive industrial atmosphere in the cluster (Bagnasco, 1988; Becattini, 1990; 2000a; Lorenz, 1992; Platteau, 1994; Humphrey and Schmitz, 1998; Nadvi, 1999b). Within this, we emphasise two complementary and interdependent forces (i.e., ‘social cohesion’ and ‘self-realisation’) that have been too often neglected in the development of SME clusters in developing countries. Recognising the interpretative value of each of the above-described approaches, we underline the importance of adopting an integrated approach that pulls these streams together (this will become the ‘eclectic approach’ in Chapter 5 and Chapter 6). This approach represents the basis to build up a framework for the analysis and promotion of SME cluster development in general and, specifically, of ‘survival clusters’. Within this framework, the lack of some relevant levels of analysis (e.g., lack of either the ‘collective efficiency’ approach or the ‘policy-inducement’ approach or even the ‘social embeddedness’ approach) may not prevent the development of SME clusters, but it is very likely to slow down the dynamic growth of the local economy. After this review of theoretical approaches, Chapter 4 analyses the specific case of a developing country, Nicaragua, in order to illustrate the development policies that have been implemented and their weaknesses in the case of specific cases of SME survival clusters. In particular, these weaknesses are related to the adoption of approaches that did not take into account all of the key development factors identified by the main theoretical approaches. The evidence shows that no systemic policy has been implemented in that country, but rather a certain number of unconnected support actions (García, 1999; MIFIC, 2001b). This work relates the poor impact of these various actions to the weak integration between the above-mentioned relevant approaches (Parrilli, 1999a; 2003). Within
Introduction
7
this perspective, an eclectic approach, which integrates the three former theoretical approaches to the development of SME survival clusters, should set up better conditions to allow them to undertake a more solid process of development at the local and national level. In Chapter 5, we clarify even further the theoretical framework that may help to explain the development process of SME (survival) clusters. This approach stresses the relevance of taking a ‘stage approach’ to the development of SME clusters and the need to take also an ‘eclectic’ stance. These two aspects constitute the main hypotheses singled out and that we empirically explore in Chapter 6. Through the first hypothesis, we point out that cluster development passes through stages, which need to be taken into account before defining feasible projects at the local level. Neglecting this aspect can mean setting up too difficult targets for producers and their associations, and can cause the cluster to fail, which is what happens when policy-makers try to apply too quickly the known model of industrial districts to less dynamic clusters in developing countries (UNIDO, 1994; Bouchier, 1998; García, 1999). However, we suggest the importance of an integrated approach to cluster development, meaning that this process needs to be based upon several forces simultaneously and complementarily. More specifically, this means that the main theoretical and policy approaches to the development of SME clusters may be taken together for their capacity to enrich jointly the growth process of SME survival clusters. Neglecting one or some of them would reduce the possible speed and strength of the development process in these local production systems. In Chapter 5, this theoretical structure is applied to the context of ‘industrial districts’ (ID) as they grew in central-northern Italy during the second half of the past century (what is called ‘the Third Italy’). For some time, the successful history of these industrial districts and their SMEs led the development process of Italy (i.e., from the 1960s to the 1990s) (Brusco and Paba, 1997). This success shows the successive and simultaneous application of relevant factors (i.e., ‘collective efficiency’, ‘social embeddedness’ and ‘policy inducement’) that increased the competitiveness of these local production systems over time. The study of the complex way in which this process has taken place in the Third Italy is likely to help the theory of development economics and the practice of policy-making in less dynamic clusters in developing economies. It could help governments and international and national development agencies to take a view of development that integrates different relevant approaches to SME cluster development, instead of focusing
8
SME Cluster Development
on one specific approach at a time. In this way, these institutional agents could be more effective in the adoption and adaptation of external development models to selected clusters. The empirical evidence coming from the Third Italy and its historical development certainly helps to hypothesise that also apparently static SME clusters have an opportunity to grow, notwithstanding their present level of development.
3. Empirical explorations in Latin American clusters Part II concludes with Chapter 6, in which we empirically explore the conceptualisation that we have made in the former two chapters, which refers to the eclectic theoretical framework and the potential trajectory of SME clusters through different stages of development. The empirical analysis is realised in three countries that rank at different levels of development: Italy, Costa Rica and Nicaragua.7 The long-term personal experience we had in a specific manufacturing sector (i.e., furniture) oriented the selection process within these countries. Three SME furniture clusters (Masaya-Nicaragua, Sarchí-Costa Rica and Forlì-Italy) are selected because of their relevance within their own sectoral economies, and these are analysed through a broad focus that involves economic, social and political/institutional aspects (involving ‘collective efficiency’, ‘social embeddedness’ and ‘policy inducement’). In this way, the relevance and impact of these various levels of analysis and development policy are explored. The conclusion of this work stresses that in the three clusters the above-mentioned types of factors (i.e., ‘collective efficiency’, ‘policy inducement’ and ‘social embeddedness’) are likely to matter. In particular, the Italian cluster of Forlì is a world leader that generates more than 250 million dollars of exports annually (Camera di Commercio, 2005). Quantitative and qualitative data collected show that this cluster benefits in a dynamic way from a variety of factors. ‘Collective efficiency’ has been pushing the competitiveness of the cluster through ‘joint actions’ and ‘external economies’. The social fabric promoted the creation of several small dynamic enterprises, while ‘social cohesion’ permitted the local cluster to maintain high levels of trust over time. In terms of policy support, both national and local governments have incentivised the initiative of the private sector and eased their concrete production and market operations from the early 1950s to now (Fauri, 1996; Bertini, 1997; CSIL, 1997; Belussi and Bertini, 1998; Bardi, 2000). Despite being defined a ‘survival cluster’ (Altenburg and Meyer-Stamer, 1999), Sarchí is the most recognised furniture production site in Costa Rica.
Introduction
9
Producers sell anywhere in the country. This SME cluster has been expanding more and more in the past 20 years (Pérez-Sáinz, 1994; Murillo, 2002). This process has been supported by a good endowment of economic (e.g., ‘joint actions’ and flows of clients, workers and information) and social factors (in terms of a significant ‘social cohesion’ and entrepreneurship). In contrast, policy support has been reduced up to now both at the national level (a national law for SMEs was approved only in 2002) and at the local level (instead of receiving public support from local government and other public institutions, the local private sector is rather charged with growing costs). In the Latin American context, Masaya is likely to represent the most typical ‘survival cluster’. Nonetheless, it has become the leading furniture cluster in Nicaragua over the past 20 years and continues to expand, little by little, its position within the national market (Barahona et al., 1998; Parrilli, 2002a). In this case, the good endowment of natural resources (less rich in recent years) and agglomeration economies (especially in terms of the labour force and clients) is not accompanied yet by appropriate frameworks at the social level (‘social cohesion’ is still compromised by the internal conflicts in the 1970s and the 1980s) and political level (national laws for SMEs do not exist and the local government does not act strongly for the promotion of the cluster) (Parrilli, 2003). All these bottlenecks tend to limit the fast-track growth of the cluster. Among these three cases Forlì represents the most advanced step in SME cluster development, while Sarchí and Masaya show lower levels of economic development. Nevertheless, these gaps are not likely to be permanent and their survival conditions could effectively transform into more dynamic conditions of embryonic kinds of cluster. The analysis of the latter clusters shows that also these are transforming and upgrading, but certainly they do so at a slower rate. These results are compatible with the hypothesis that clustering represents a relevant development opportunity for SMEs also in the case of developing countries, but that a significant development can take place only once appropriate conditions materialise. This empirical exploration is relevant for policy-making, since it establishes two main conclusions. The first indicates that in all kinds of cluster, but especially in those that are less dynamic, policy-makers might think in strategic terms about the development of their cluster. They could assess what feasible step can be targeted next, avoiding too quick and advanced steps that local agents cannot really achieve soon and setting up the appropriate support instruments that permit the cluster to achieve the most feasible target first.
10
SME Cluster Development
The second conclusion implies that development may be catalysed much better when the different types of factors (‘policy inducement’, ‘social embeddedness’ and ‘collective efficiency’) operate simultaneously through the adoption of an ‘eclectic development approach’. This could lead to structuring a more complex type of policy-making that puts together these different levels in a cumulative and self-reinforcing model that is more likely to produce higher performances and developments. This significant change proposed in the policy approach represents also an important example of the way local systems in developing countries can be organised to respond to the global challenge discussed in Part I (i.e., the fracture arising between the modern and the traditional sectors of the economy). Indeed, the ‘eclectic’ cluster policies represent a change in the paradigm of development policy towards a systemic approach that requires the structural involvement of proactive policy and that includes actors from both the modern and the traditional sector in the local and the national economic systems.
Part I Small and Medium Enterprise Development
This page intentionally left blank
2 Integrating the National Production System: The New Challenge for Chile
1. Introduction8 This chapter expands the argument made in the Introduction about the fracture arising nowadays within local production systems in developing countries, and more specifically, in Latin America. In this case, special reference to Chile is made because of the extreme clarity of process indicators that this economy presents with regard to that problem: the fracture in the national production system between a small modern sector of large and medium enterprises versus a large group of micro and small traditional enterprises. In particular, the conclusion of this chapter leaves a big question about what kind of policies may be used to restore unity within the national production system. In fact, Chile fits well the development model promoted over the past 20 years in the developing world by multilateral organisations and institutions from industrialised countries. After decades of import substitution industrialisation (ISI), the debt crisis of the early 1980s bred severe financial and economic crises in all developing economies. Neoliberal structural adjustment programmes (SAPs) became the chosen policy to carry this and other developing economies out of crisis. These promoted market liberalisation and thereby promised optimal resource allocation and national growth (Cornia et al., 1989; Dornbusch, 1996: 4). Notwithstanding, after two decades of attempts to start a new growth process, serious economic imbalances remain in this and various other countries, and new crises occupy the horizon. The certitude of organisations in the pursuit of these policies needs now to be questioned, and this approach to development reconsidered to generate new ideas, mechanisms and models. 13
14
SME Cluster Development
The author has in loco verified that the neoliberal policy framework and concomitant unilateral market liberalisation had a significant impact on small and medium enterprises. It seems this played a key role in forcing micro firms and SMEs to compete, in the absence of any structural defence, with a large number of international firms. As a consequence, enterprises have tended to divide in a dualistic system, in which a reduced number of big modern firms are able to compete in the international market, while the large majority, usually micro enterprises and SMEs, tend to be marginalised, and barely survive in the national economy. This study examines the enterprise system in Chile in an effort to identify both the effects of development policy over the past decades, and the need to search for alternative and more inclusive development policies that integrate all types of firms and actors in a process of shared growth and welfare (the next chapter expresses in more detail the policy framework that could help respond to such challenge). The chapter is structured as follows. The conceptual framework is established in section 2. Section 3 details the implementation of the main policy approaches (both explicit and implicit) within the time frame of the study. Section 4 analyses the relevant data. An interpretation of the present slowdown is offered in section 5. By way of conclusion, section 6 summarises the competitive gaps observed in the national production system and calls for an organised policy answer which solves the problem of the growing fracture between firms by means of helping to structure a stronger and more uniform production system.
2. A conceptual framework This study investigates the harmful effects that neoliberal policies have on the productive structure of a dynamic developing country. It focuses on the tendency to split the system into a small modern and large traditional enterprise sector. This occurs because liberalisation policies lead defenceless national enterprises into competition with a large number of stronger international firms. In this situation, traditional SMEs and micro enterprises (see note 5 in Chapter 1) do not compete successfully and are marginalised from the market, due to their limited structural capacity to access finance, technology and markets. In contrast, a minority of modern, capital-intensive large firms can offset such competition and maintain their market share, through significant investment in technology, and sacrifices in terms of employment. The hypothesis reconsiders the concept of dualistic development to offer an alternative perspective to that used since the work of early
Integrating the National Production System
15
development economists. In the 1950s and the 1960s, Lewis, and Harris and Todaro emphasised the economic mechanisms of dual economies, such as the (expected) income gap between rural and urban life, consistent with the desired process of accumulation and growth (Harris and Todaro, 1969; Lewis, 1973). This approach considered economic dualism as ex ante (i.e., as a condition), and as a positive element in the process of development. In contrast, in the 1970s a stream of critical economists made it clear that Rostow’s notion of modernisation was too simplistic to explain development processes taking place in the developing world. In the Third World, the notion of dualism, though still ex ante, had also to be seen in social and political terms, as tiny elites within these countries reaped the benefits of growing industrialisation and linkages to the international market, while denying these advantages to the rest of the population (Prebisch, 1982: 470–2; 1984: 177, 184–6; Hirschman, 1984: 108–9; Frank, 1992: 8–9). This work introduces a structural modification to these approaches to dualism. The former views conceptualised dualism as an ex ante social and political condition of development, which was dependent upon the particular societal structure. By contrast, this work argues that dualism is more an ex post effect, due to the working of free markets, which create the conditions for possible market failures, such as in finance, technology, commercialisation, human capital, and reproduce a skewed division of benefits and opportunities within the developing world (Stiglitz, 1998: 5, 27; Peres and Stumpo, 2002).9 Classical, neoclassical and neoliberal theories hypothesise that, through the ‘invisible hand’ of the market, a positive sum game is naturally generated by interactions among individual agents. This happens under the assumption that the market is in perfect competition. Nevertheless, information asymmetries, market failures and regulatory obstacles make it clear that perfect competition is a theoretical construction, not apparent in developing economies. In these dual economies, the lack of a ‘dialogue’ among domestic firms, in terms of a lack of work linkages, is a frequent collateral effect of market liberalisation. This results in the isolation of a few large national and foreign firms, which use advanced technologies to produce a broad array of goods for the national and international market. This constitutes the modern sector. Simultaneously, many micro enterprises, and some SMEs continue to use second-hand machinery and low-skilled workers to produce generalpurpose goods for the local market. This represents the traditional sector. The overall effect is the generation of a two-speed economy, in which
16
SME Cluster Development
many people remain outside of the development process without opportunities for personal and collective growth (Cowling and Sugden, 1997). Every country has the ‘opportunity’ to undertake positive economic and social development, but only under the provision that there is an appropriate policy approach. The outstanding development of several Asian countries and the relevant debate that took place at the beginning of the 1990s, demonstrated that markets do not necessarily create dualistic economies and societies. In those cases, the coexistence of proactive industrial policies and strategies, oriented to enabling local industries to join the world market, had the capacity to stimulate the joint and gradual effort of national firms to enter more competitively in the world market (Wade, 1990; World Bank, 1993; Amsden, 1994; Rodrik, 1994). In order to build a more systematic industrial strategy, policy-makers need to go beyond recognising the heterogeneity of the enterprise system and the separation among its agents. Otherwise, these agents would continue to act separately and, therefore, lack the economies of scale and scope that would help them offset competition from larger firms and multi nationals (Schmitz, 1992; Bianchi and Parrilli, 2002b; Sylos Labini, 2000). In contrast, policy-makers can succeed by giving more emphasis to the importance of relations that link up the different types of enterprises (e.g., micro, SMEs and large firms). In this way, they can build a common and systemic strategy to gain market share and increase national wealth and employment. This chapter emphasises the need to establish a particular industrial strategy that, through an eclectic and wise use of various policies and instruments, helps to close the gap between modern and traditional firms, and shape a stronger and more cohesive industrial system. This is analysed in detail in the final section of Chapter 3.
3. Policy approach in the 1980s and the 1990s 3.1 The explicit policy approach In the past two decades, the neoliberal policy approach has dominated academic debate and policy-making in Chile, and more generally in Latin America. In the aftermath of the debt crisis, it was presented by the main international financial organizations (i.e., IMF and World Bank) as the only possible way out of crisis (Krueger, 1983; Balassa et al., 1986). The neoliberal position claims that the competitiveness of the national productive system depends mainly on two policy baselines. The first is founded upon the comparative advantage model proposed by HeckscherOhlin 70 years ago. It stresses the importance of specialising in producing
Integrating the National Production System
17
goods that are compatible with the country’s factor endowments (Krueger, 1983; Balassa et al., 1986). The second is based on good management of the national economy, which is defined by the rather strict control of monetary and fiscal policies (i.e., inflation, public expenditure, currency appreciation and balance of payments) and the reduction of state intervention (e.g., decreasing the number of state-owned enterprises and cutting public employment) (Singh, 1992). Neoliberalism considers the present economic and competitive problems in Chile unconnected to the absence of industrial policies. On the contrary, these problems are argued to be rooted in state intervention, the adoption of industrial policies, and the improper management of macroeconomic variables. On the whole, the proponents of this approach understand the basis of the recent slowdown of the Chilean economy to be the policy shifts that occurred over the past decade (Fischer, 2001). It is argued that the policy shift led to real appreciation of the peso from 1991 to 1998, and the consequent economic slowdown. They also observe that from 1999, better macroeconomic management led to a real depreciation of the peso, generating an export growth of 19 percent in 2000, and setting Chile once more on the road to growth (Fischer, 2001: 17; Dornbusch, 1996: 4–6). Similarly, they highlight another policy error, namely, an ‘undeclared’ shift in national trade policy over the past 10 years. From a neoliberal perspective, the Chilean authorities abandoned tariff barriers on imports, and shifted to the use of non-tariff barriers, such as safeguards, health regulations, price bands and anti-dumping regulations (Saez and Valdez, 1999; Ffrench-Davis, 2002). Simultaneously, Chile began parallel entry in numerous free trade agreements (FTA), not only with neighbouring Latin American countries, but also with North America and the European Union (Scapini, 1998). These policies were prompted by a fear of isolation from international and geopolitical arenas, particularly after the dictatorship, and when it was critical not to be excluded from the most important trade relations (Hachette, in Fischer, 2001: 6). These considerations led the country towards an active trade policy.10 As mentioned above, neoliberalism stresses the relevance of comparative advantage theory, which emphasises the capacity of a resource-based economy to generate exports and growth. From this perspective, if Chile effectively bases its growth on natural resources, successful exports in one sector will build new knowledge to further develop other production and export lines. This has occurred in the case of the fishery and mollusc industries (e.g., turbot and abalone), which are based on the outstanding success of the salmon industry (Fischer, 2001: 22–4; 29–30). The strategy is supported
18
SME Cluster Development
by the consideration that, ‘the technology required in these industries is quite sophisticated and in terms of value added, sending fresh fish from Chile to the US adds perhaps more value than a traditional industrial export’ (Perez-Aleman, 2000: 44; Falabella, 2001: 261; Fischer, 2001: 22). Before examining the implementation of these strategies, a further analysis must be made. This focuses on practical policies applied in Chile since the early 1990s, which do not fully reflect the prior neoliberal approach, because they imply some form of soft-state interventions. For this reason we call them ‘implicit’ industrial policies. 3.2 Implicit industrial policies Over the past 10–15 years, Chile, along with other Latin American countries, has implemented several ‘implicit’ policies, in addition to the preexisting explicit policy options. Although the government maintained its support of neoliberal policies, from the late 1980s onwards it began promoting two new kinds of initiatives. In order not to miss the growing opportunities presented by proliferating free trade and political agreements, it applied active trade policies. Simultaneously, it supported specific economic groups through particular development programmes (see CORFO operations in Table 2.1). In an ‘implicit’ way, this promoted an export-led model. The implementation of several measures (e.g., subsidies, etc.) by successive governments that tried to boost national exports shows that an active policy approach was in place even in the 1980s (Agosin, 1999; Saez and Valdez, 1999; Ffrench-Davis, 2002). The macroeconomic growth of the country from the late 1980s to the late 1990s indicates the good performance outcomes that this ‘implicit’ industrial policy contributes to generate. At the sectoral level, ‘implicit’ policies were also applied, but in a rather fragmented manner. To substantiate this, it is essential to understand the development operations of public (and private) support organisations. Therefore, this analysis focuses on the work of the biggest public organisation, the development corporation CORFO, because it best indicates the industrial policy approach applied in Chile up to now. CORFO is an independent institution, financially stronger than the Ministry of Economy and Energy, from where it is controlled. In fact, it manages a bigger budget (Ministerio de Hacienda, 2001). Its strength is justified by neoliberal thinktanks and the main international organisations that want to reduce the policy role of the Ministry of Economy in every developing country so as to more fully orientate them to market demand. This belief characterises CORFO (CORFO, 1999). As a matter of course, it is likely to reduce the possibility of state planning and intervention.
Table 2.1:
CORFO’s development programmes in 2002
Programme name
Programme type
Modalities
Direct management
PDP
Subcontracting development
Subsidies for up to UF 9000 per group of firms in 3 years
Private agencies
PROFO
Association and cooperation
Subsidies for up to UF 8100 per group of firms in 4 years
Private agencies
FAT
Technical assistance
Subsidies for up to 50% of the cost and UF 150 per firm
Private agencies
FAT clean production
Technical assistance for clean production
Subsidies up to 70% of the cost and UF 125 per firm
Private agencies
PAG
Enterprise management and relative actions
Subsidies for up to 50% of the cost up to UF 1,500/firm
Private agencies SERCOTEC
Programme for commerce
Commercialisation
Programme pre-investment
Subsidies for Pre-investment studies
Up to 60% of the feasibility study and US$ 12,000/firm
CORFO
Programme of credit insurance
Subsidies for credit insurance
Subsidise the eventual bankruptcy up to US$ 30,000
CORFO CORFO
Funds for innovation development
Public procurement
Funds product and service innovation
Subsidise up to 50% of the cost
CORFO
FONTEC II
Funds for investing in infrastructures
Subsidise up to 30% of the cost
CORFO
FONTEC III
Funds for associative technological transference
Subsidise up to 50% of the cost and up to US$ 100,000
CORFO
FONTEC IV
Funds for centres of technology transfer
Subsidise up to 50% of the cost up to US$ 400,000
CORFO
FONTEC V
Funds for innovation pre-investment study
Subsidise up to 50% of the cost up to US$ 15,000
CORFO
Programme for supporting credit
Subsidies for credit support
Subsidise up to 80% of the first payment and the commission for the guarantee
CORFO
19
FDI FONTEC I
(Continued)
(Continued)
20
Table 2.1:
Programme name
Programme type
Modalities
Direct management
Programme of training for new hired workforce
Subsidies for hiring workforce
Subsidise up to UF 140 per worker
CORFO
Programme for industrial urbanisation
Subsidies for urbanisation of regional lands
Subsidy up to UF 0.5 for square metre, 6,000 per firm
CORFO
Programme for industrial regional localisation
Subsidies for industrial regional localisation
Subsidise up to US$ 10 million for project generating more than 200 jobs
CORFO
Programme for buying industrial estates
Incentives for buying industrial estates
Subsidise up to 50% of the cost of land and UF 2,000
CORFO
COBEX
Borrowing guarantee for exporters
Guarantee that cover up to 40% of the total guarantee
CORFO (guarantee), private banks (credit)
Reprogramming SME debt and Credit Line B.13
Reprogramming debt with CORFO guarantee
Up to US$ 150,000 to be paid in up to 6 years
Banks for credit, CORFO for guarantee
Credit Line A.3
Financing leasing operations for SMEs
Up to UF 15,000 with repayment terms of 3–5 years
Leasing companies and banks
Credit Line B.14
Credit for environmental protection operations
Up to US$ 1 million and repayments term up to 12 years
Private banks
Credit Line B.22
Credit for input purchase and selling abroad
Up to US$ 3 million and repayment up to 8 years
Private banks
Credit Line B.21
Credit to foreign buyers of goods and services
No financial restrictions and up to 10 years repayment term
Private banks abroad
Credit Line B.11 and B.12
Financing SME investments
Up to US$ 5 million with repayment term up to 10 years
Private banks
Credit Line B.41 and B.42
Financing under- and post-graduate studies
Credit up to UF 3,000 to study abroad
Private banks
Source: Author’s compilation based on CORFO web page <www.corfo.cl/instrumentos> (2002).
Integrating the National Production System
21
Several studies highlight the proficiency of development activity carried out by CORFO, in initiating an upgrading process within the production system and related business associations. CORFO pushed the enterprises to adopt higher quality standards, efficiency and scales of production. Its work resulted in increased exports from more sophisticated agro-industrial production (e.g., salmon, abalone, wine). All these activities have been promoted by CORFO and other public agencies (e.g., PROCHILE in exporting, FOGAPE in credit) over several years (Perez-Aleman, 2000; Fischer, 2001). A study of CORFO’s development programmes reveals several characteristics. In the past 10–15 years, this institution, established in 1939, assumed more importance and a new development profile. It started to adopt a more market-friendly approach, becoming extremely reactive to market demand and abandoning the former ISI policies. It has recently implemented several support programmes in order to respond to a variety of demands from different economic groups. Table 2.1 indicates CORFO’s interventions, often in the form of subsidies. It shows that the Chilean government – through CORFO – acted in a way that is somehow distant from pure neoliberal prescriptions. However, although several industrial interventions have been implemented, these support programmes are rather fragmented and focused on responding to the demand of specific groups of national stakeholders (i.e., business associations).11 In this sense, CORFO follows a pattern, which does fit into the neoliberal approach. The analysis now addresses the main results of ‘explicit’ and ‘implicit’ economic policies implemented in Chile over the past decades.
4. Enterprises’ reaction to crisis This section provides evidence that there exists a tendency for the economic system to marginalise a large number of economic agents and reinforce a dualistic framework within society as a whole. This fracture weakens the productive and industrial structure of the country, and reduces its capacity to respond to international challenges in the age of globalisation. Table 3.3 in Chapter 3 indicates the importance of micro enterprises and SMEs within the national production system. They are the most numerous firms and account for a significant proportion of sales and employment. Interestingly, the data in that table also reflect the crisis that recently impacted upon the country and the entire continent. Figures 2.1, 2.2, and 2.3 illustrate this. On the one hand, large firms and SMEs have been shrinking in number while their sales have increased (in terms of firm’s average);
22
SME Cluster Development
on the other hand, micro enterprises have been growing in number while their sales have decreased (Bianchi and Parrilli, 2002a; see also Chapter 3). Within the micro enterprise and SMEs sector, there are two phenomena worthy of note. First, an increasingly large number of micro firms (240,000)
600000 400000 200000 0 Micro enterprises
Large firms 1997
Figure 2.1:
2000
Recent evolution in the number of enterprises
4000 3000 2000 1000 0 Micro enterprises
SMEs 1997
Figure 2.2:
Large firms 2000
Recent evolution of sales (UF)
1000000 800000 600000 400000 200000 0
Micro enterprises
SMEs
1997 Figure 2.3:
Large firms
2000
The recent evolution of firm productivity (UF)
Integrating the National Production System
23
did not declare sales in 2000. Second, many SMEs closed down their operations or became micro enterprises. Thus, on the whole, a dual trend can be seen within the same micro enterprise and SMEs sector. The more active firms show the same trend as large firms (i.e., increasing sales per enterprise and employee with a parallel decrease in total number of firms and employment), while the ‘non-declaring’ enterprises show the opposite trend, with increasing numbers of firms, rising employment and decreasing sales (INE-Gobierno-BancoEstado, 2001). In this sense, a tendency of growing dualism in the performance of national enterprises seems to be confirmed. In our view, the main explanation of these tendencies lies in the competitive shock suffered by a large number of enterprises, whose market has been increasingly squeezed by imported products. Indeed, Table 2.2 reveals that from 1992 to 1998, imports have grown by 133 percent. This sudden increase has taken place mostly in manufactured products, which represent 80 percent of total imports. Within manufacturing, the most relevant sectors are chemicals, machinery and equipment, textiles and clothing, food and beverages. The last two sectors are particularly interesting, because they are densely populated by micro firms and SMEs suffering increasing competition from East Asia and China (in the case of textiles) and from Latin America, Australia and the United States (in the case of food and beverages). The interviews with entrepreneurs conducted for the Inter-American Development Bank and the Ministry of Economy and Energy of Chile in 2001 and 2002, clearly identified the problems these industrial sectors encountered in the face of extremely cheap imports (Bianchi and Parrilli, 2002a). Table 2.2:
The growth of imports in Chile, 1992–98 (million pesos) 1992
1995
1998
Textiles and clothing
257,328
433,005 (⫹68.27%)
632,434 (⫹46.05%)
Food and beverages
162,165
294,256 (⫹81.45%)
480,893 (⫹63.43%)
Other manufacturing Sectors
2,881,735
5,029,384 (⫹74.52%)
6,932,348 (⫹37.84%)
Total manufacturing
3,330,228
5,756,645 (⫹72.86%)
8,045,675 (⫹39.76%)
Total imports
4,446,912
7,433,498 (⫹67.16%)
10,382,871 (⫹39.67%)
Source: INE (1999): 113.
24
SME Cluster Development
At the same time, domestic production in the manufacturing sector suffered a setback both in 1998 (⫺1.5 percent) and 1999 (⫺0.7 percent). These data indicate the overall negative impact that imports had on local producers by shrinking their space in the national market. This trend is even more pronounced in the case of micro enterprises, and within a segment of SMEs (INE, 1999: 28). In 1997, about 26,000 micro enterprises worked in the manufacturing sector; three years later this number had shrunk to about 9,000 (with sales above 600 UF12 per year or about US$ 15,000). This means that more than 17,000 micro enterprises lost competitiveness, ended up disappearing, or made too few sales to declare (CORFO, 1999: 17; INEGobierno-BancoEstado, 2001: table 2a). The same trend occurred in all economic activities. Table 2.3 in Chapter 2 shows that, on the whole, micro enterprises boomed in terms of number (⫹90,000 from 1997 to 2000), while many (i.e., 240,000) simultaneously declared that they did not make relevant sales volumes during those years. These firms represent the ‘informal sector’, which behaves anti-cyclically, growing with economic slowdowns and shrinking during expansions (Hart, 1977; Tokman, 1992). In other words, the strength of these firms lies in their resilience in times of economic crisis. They are informal enterprises that have the capacity to save on meaningful costs of labour, health, environment, taxes (e.g., by declaring less of what they sell) and industrial buildings (for instance, by using their own house as a workshop). In this way, they maintain some kind of sustainability and/or subsistence. In contrast, the remaining spectrum of firms reacted very differently. In fact, the wide majority of large firms and a small proportion of SMEs and micro enterprises were able to react by investing in new and labour-saving technologies, reducing their employment and maintaining, or even increasing their sales (see Figure 2.3). As a result, the levels of sales per firm and their labour productivity slightly increased, giving them a new competitive edge within a globalised market (INE-Gobierno-BancoEstado, 2001: table 5.1.3.a). In sum, these data highlight the divergent trend between these two sectors; modern up-to-date technology-intensive large firms and SMEs, and traditional labour-intensive SMEs and informal micro enterprises. There exists an identifiable and growing risk of a split in Chile’s social and economic production system. In the following section, some complementary evidence and benchmarking practices are examined in order to explain in more detail the inadequacies of current industrial strategies.
Integrating the National Production System
25
5. Interpretations of the slowdown: ECLAC’s critique and the de-industrialisation process The most important critique of neoliberal policies as applied in Latin America and Chile is made by ECLAC (United Nations Economic Commission for Latin America and the Caribbean). Some of its scholars claim that, since the early 1970s, national policy-making has been consistent with the orthodox neoliberal approach. This has been so despite soft governmental interventions through organisations such as CORFO. Stumpo (2000) argues that this policy regime led Chile into a significant de-industrialisation process. The same study identifies an interesting outcome; the share of industry in GDP declined from 24.9 percent to 17.2 percent between 1972 and 1997. This trend was not reversed in the 1990s; the share fell from 19.2 percent to 17.2 percent from 1989 to 1997 (ibid., 2000). It is also important to observe the peculiar transformations taking place within the industrial sector. The industrial composition of production changed from 17.7 percent to 5.1 percent in science-based industries (e.g., electronics, pharmaceuticals), from 45.5 percent to 66.5 percent in natural resource-based industries (e.g., copper processing, cellulose production) and from 36.8 percent to 28.5 percent in labour-intensive-based industries, such as textiles and shoes (ibid., 2000: 17; Reinhardt and Peres, 2000). It becomes clear that the bulk of industrial activities, which in Chile are not high-technology oriented, are increasingly linked to natural resources, and mainly focused on the exploitation and transformation of those resources through relatively basic processes. This contrasts with the neoliberal assumption, itself founded upon the theory of comparative advantage, of the possibility to shift from low value-added resource-based production to more sophisticated types of production. In terms of industrial sectors, some industrial chains have been neglected and even dismembered over this period (1972–97), while others have increased their share in the national economy. For example, industrial production of durable and capital goods has been shrinking. At the same time, production entailing less value-added, such as forestry and pulp, agriculture, fish, flowers, vegetable oil, canned fish, tomatoes and fruit, wine, cattle, milk, and dairy products has been increasing substantially (Reinhardt and Peres, 2000; Stumpo, 2000). The eventual outcome of this policy approach can be clearly identified, ‘[I]n the last twenty years, in Chile it has not been possible to find a clear tendency toward the creation of productive chains that start from
26
SME Cluster Development
the transformation of natural resources and indicate the birth of a more complex and diversified productive structure … At the same time, the Chilean production in more sophisticated markets and sectors does not seem to increase sensibly. In this sense, the growing trade deficit highlights a clear loss of competitiveness’ (Sachs and Warner, 1995; Meller et al., 1996; Stumpo, 2000, p. 19). On the whole, the present weakness of the Chilean economy, and in particular in the traditional and vast SME and micro enterprise sector, seems also to be caused by the absence of an industrial strategy that emphasises high-tech (i.e., science-based) industry, and enhances links between modern and traditional enterprises and sectors.
6. The need to set up integrative industrial strategies Although in the 1990s implicit governmental policies, well represented by CORFO, contributed to sustain Chile’s position as one of the most successful Latin American economies, now the country faces critical structural problems in its production system. In fact, Chile’s growth rate was 8 percent per annum from 1983 to 1998, but declined to negative rates over 1998 and 1999, and then rose to 3 percent in 2001 and 1.8 percent in 2002 (BCC, 2002).13 While this Chapter recognises CORFO’s positive work in the 1990s, it extends the analysis through a specific critique to propose an alternative industrial strategy for both Chile and other governments enmeshed in the process of globalisation. In the contemporary world of fierce competition, a ‘fragmented’, or market approach to economic development induces operational inefficiencies. These occur when too many development programs and tools hinder the creation of scale and scope economies, key to efficiency and competitiveness in the global market (Schmitz, 1992; Bianchi, 1998; Sylos Labini, 2000). If the country is not able to integrate a critical mass of firms (e.g., through export consortia, credit consortia, spin-offs programmes, etc.), the failure of the productive system will persist. This lesson can be drawn for the prospective success of several territories and related networks of small and medium enterprises (Humphrey and Schmitz, 1996; Ceglie and Dini, 1999; Bianchi and Parrilli, 2002a). The data presented in CORFO’s Annual Report seem to confirm these considerations. In fact, these programmes involve no more than 2–3,000 firms each, and 15,000 in total (CORFO, 1999). As 500,000 firms exist in the country, this service does not reach a significant number of firms and,
Integrating the National Production System
27
therefore, cannot produce organised and systemic change. The hypothesis has also been tested through interviews with several entrepreneurs and managers in recent fieldwork in Chile (Bianchi and Parrilli, 2002a). Most often, they confirmed that they did not know about important governmental development programs (e.g., CORFO, BancoEstado-FOGAPE, INN). In this sense, it is interesting to re-examine the experience of European countries, both individually and as a group. Greater success here is a result of a systemic approach to development directed to a critical mass of enterprises, rather than supporting individual firms or small groups of firms (Bianchi, 1998; Capaldo et al., 1998). One of the most successful examples of this approach is the Sabatini Law created in Italy in 1965 to promote the technological upgrading of small and medium firms. This catalysed more than 200,000 credit lines worth more than $100 billion in subsidies over a 30-year period (Bertini, 1998: 317–18). The record of thousands of firms associated in a number of credit and export consortia exemplifies a development institution based on a specific legal recognition designed to spur systematic changes across all enterprises (Federconfidi, 1997; Bianchi and Parrilli, 2002a). The same logic applies to any program. Supporting individual enterprises does not produce the same positive results as supporting groups of firms, networks, and territories. These alternative policies, implemented through new programmes in most European countries over the past decades, constitute more useful lessons. Programmes and activities oriented to credit, export, subcontracting, training and other aspects of competitiveness are here structured in a way that promotes the advantages of participating in association with a large number of enterprises. The approach is also taken at the territorial level (Best, 1990; Nadvi and Schmitz, 1999; Humphrey and Schmitz, 1996; Miller and Bianchi, 1998: 151–6). From this viewpoint, the fragmented and implicit industrial policies of the Chilean government, together with neoliberal policies, were successful in promoting a boom among the most dynamic groups of firms rooted in natural resource sectors, and fully oriented towards the export market for semi-processed goods and materials. This approach focused on the ‘modern sector’,14 given the higher capacities of these mainly large firms both to demand government/CORFO commitment, and satisfy their requirements. However, caught in the middle of a steady process of globalisation and the entry of the People’s Republic of China into the World Trade Organization, this old approach seems unable to link the two sectors of the economy; the modern and traditional. Thus, there exists an ever-increasing risk of boosting a modern sector, increasingly small and linked to international
28
SME Cluster Development
markets and production systems, in open opposition to a growing and larger part of the economy, characterised by a decreasing capacity to create value added. The serious risk is that this modern sector is unable to generate internal dynamism and a more reliable structural context in which to sustain broad national development. In fact, the traditional sector reacts to international pressure by destructuring its activity. This entails the adoption of less developed technology and management practices, which render production uneven in terms of quality (Bianchi and Parrilli, 2002b). In this way, the tendency of this sector to be isolated from global markets persists. In synthesis, this Chapter shows that a particular effort is necessary to connect the modern and traditional sectors. Although neoliberal policies and market liberalisation respond to the need to stabilise the national macro economy, to control public expenditure and foreign debt, and to attract inward foreign investment, they have not been enough to produce a balanced and homogeneous development in the production system. For this reason a development strategy is required to focus on this issue, one that voluntarily and rationally addresses the problem by pulling together all national forces and actors. At present, most national and international development organisations work in a rather ‘fragmented’ way, which does not promote the structural reinforcement of the national economy. These institutions may be able to respond to the requests of specific business groups within a country and to help them being more competitive in global markets; nevertheless, this kind of policy response may be not enough to create a systemic answer. A zero-sum game may occur: a few firms benefit while a large number of firms do not. Simultaneously, the consumers may be paying (taxes) to raise public revenues that are spent for these reduced sectors of the enterprise system. This is why the structuring of a common solution is required in order to respond to the fracture that can be observed today in Chile as well as in many other developing economies. In this way, the firms, their local systems (e.g., clusters) and the related public institutions can avoid fragmentation and shift to a coordinated and cooperative production formula. The next chapter analyses several cases (Argentina, Nicaragua and again Chile) where the process of fragmentation of the national production system is quite evident and compares them with two cases where this seemed not to happen (Europe and, in Latin America, the case of Costa Rica). An explanation for these different responses is also presented and discussed, and this emphasises the role of an active industrial development strategy organised by the national governments over the 1980s
Integrating the National Production System
29
and the 1990s. The final section is developing this idea through the discussion of the European experience and the identification of a systemic approach to SME development policy which targets the issue of ‘fragmentation’ in the national production system through active policies oriented to SME development in the global context.
3 Inclusion Versus Fragmentation: Different Responses to Liberalisation in European and Latin American Small and Medium Enterprises 1. Introduction In this chapter, we present a comparative analysis of the role played by SMEs in the processes of liberalisation and productive restructuring in Latin America and Europe in the past few decades. This process has produced different levels of performance by generating, on the whole, a large number of very competitive SMEs in relevant parts of Europe and a large number of less competitive SMEs in Latin America (Observatoire des PME Européennes, 2000; Observatory of European SMEs, 2002/1; 2002/2; Peres and Stumpo, 2002). Although clear differences in the history of SME development in Europe can be recognised, we would like to emphasise that, in general, in these countries SMEs have become increasingly important over time. The dynamism of SMEs became evident in the 1960s and the 1970s as a response of the socio-economic system to the crisis of Fordism. These firms attracted workers that had been laid off by large factories and organised them into a flexible system of very competitive SMEs, which benefited from the European agreements (i.e., the European Common Market) of those years and exported a large part of their production (Piore and Sabel, 1984; Best, 1990). In contrast, the industrial boom of the Post-Second World War period did not generate a competitive system in Latin America, but contributed to creating – on the basis of the import substitution paradigm – a rather closed and protected economic system, which proved to be inefficient. The market liberalisation of the 1980s, begun by Mexico’s default in 1982 and the resulting debt crisis, led to a system in which most large firms were inefficient companies, while small firms were a low-profile solution 30
Inclusion Versus Fragmentation
31
to recession, closing down of firms and growing unemployment (Tokman, 1992). In particular, the weakness of the latter (SMEs) is linked to the market-driven lack of access to financial markets, advanced technology and international markets, which are based on the typical market failures that affect SMEs especially in developing countries. By looking at several case studies taken from Central and South America (i.e., Argentina, Chile, Costa Rica and Nicaragua), this chapter explains the mechanisms that have operated from that point in history (i.e., the 1980s) onwards and produced the present socio-economic outcome and its differences from Europe’s. The conclusions of this chapter seem consistent with the indication that the structuring of a coherent industrial development strategy in a framework of gradual market opening, seems to produce higher development outcomes than following the route to unilateral opening without paying attention to industrial development (Bianchi, 2002a; Bianchi and Parrilli, 2002b). In this way, the wider system of traditional (i.e., micro, small and medium) firms could have time to adjust to the changing scenario of the globalised market and become more competitive. This means that the neoliberal market framework has been good as a general framework to stabilise the economy, but it has not been able to make more (i.e., promoting development). Therefore, a more proactive role of the government is required, especially in the context of developing countries. It may create better conditions for the growth of the enterprise system, including the traditional kinds of enterprises. This aspect of state proactivity (and of local governments) are analysed also in Part II of this work, although there it is studied in the more specific context of SME clusters. On the whole, this chapter does not intend to open the important discussion on the macroeconomic environment that supports the growth of local and national production systems.15 In our analysis we rather choose to start stressing the key role of industrial development strategies and, in particular, to illustrate the relevance of using strategies, policies and instruments to close the gap between the two groups of modern, often large firms, and traditional enterprises, often SMEs, and shape a stronger and more competitive industrial system. The study of the European support to SMEs indicates the different kinds of effort of various country members (first of all Italy) to include these firms within a systemic policy framework for the development of their national production systems (Bertini, 1998; Capecchi, 1990; European Commission, 1998; 2002; Bianchi, 2000). This approach relies upon cluster policies as one of its major development strategies (European Commission, 2002). These latter can help to avoid the fragmentation of
32
SME Cluster Development
the national production system between a small fraction of large and medium modern firms and a large sector of micro and small traditional enterprises and the consequent marginalisation of the latter from the overall industrial development process (Bianchi et al., 2000; Bianchi and Parrilli, 2002b).
2. Industrial restructuring and small firms in Europe The literature shows that small firms are a very heterogeneous sector. They can include firms that range from small rural businesses to medium subcontracting firms, from small and micro high-tech firms to academic spin-offs, from cheap subcontractors to medium exporting firms and up to the small and medium networked firms of well-known industrial districts. These firms are located in a local, national and international context of production and market relations. For example, industrial districts are systems of firms that produce at the local level and sell at the international level, while small rural businesses are those that both produce and sell at the local level. Academic spin-offs are firms capable of both producing and trading at an international level (Bianchi et al., 2000). This heterogeneous universe of businesses can work in a more or less structured and networked way. Within our approach to growth and development, if the firms work in the market as separate atoms, they remain weaker than larger firms (Schmitz, 1992; Sylos Labini, 2000). In this way, we can hypothesise that the local and/or national system as a whole loses competitiveness. The lack of long-term links between firms, which is often a collateral effect of market liberalisation, increases the risk of splitting the national production system into a two-speed economy divided between a modern sector, composed of a few isolated large national and foreign firms, which produce a wide range of standardised goods for the national and international market through advanced technology, and a traditional sector with many thousands of small and micro firms, that work as craft shops, with second-hand machinery and low-skilled workers, producing low quality goods for the local market. This all process can be seen in several Latin American cases, as the next sections show (Bianchi and Parrilli, 2002a; 2002b). The history of SMEs in Europe illustrates a different kind of process, which has been able – despite obvious differences across countries – to avoid the risks of the above-mentioned socio-economic fracture. Since the 1960s, specific member states within the European Community (in particular Italy) have promoted the geographical and sectoral concentration of small firms, which helped them overcome the constraints of their small size of operations.
Inclusion Versus Fragmentation
33
Several legal tools were established and linked to this kind of logic (Bertini, 1998; Bianchi, 2000; Observatoire des PME Européennes, 2000; Observatory of European SMEs, 2003). The best example for this approach is Italy, where several laws for joint SME actions were approved, such as the law for promotion of cooperatives (e.g., Law 49 of 1985), and consortia (e.g., Law 240 in 1981), among others (Bianchi and Parrilli, 2002a). In Table 3.1, we can see the set of policies currently at work within the European Union to help small firms develop in the global market. All of these policies highlight the thorough support that most of the European Union member states are currently giving to SMEs. But, more importantly, these countries have recently stressed the importance of inserting these policies and instruments within a wider framework of ‘structural policies’, which define a type of intervention oriented towards increasing the productivity of the traditional sector of micro, small and medium enterprises as a whole. The idea was not to create incentives for the subsistence of inefficient firms (as occurred in the 1950s), but to promote greater competitiveness by spurring on new economies of agglomeration, driven by processes of territorial and sectoral networking among enterprises and institutions (European Commission, 2002). These structural policies are being now accompanied by simultaneous ‘innovation policies’, that are oriented towards increasing the competitive ability of national and regional productive systems to participate in hightechnology markets (European Commission, 1998). Complementarily, new policies have been developed to connect the modern sector of large firms to the wide sector of traditional micro, small and medium enterprises. These ‘linkage policies’ have created fluid links among the most advanced and the most traditional firms in order to promote the growth of the system as a whole, instead of breaking it apart and allowing only some parts to grow (European Commission, 1998; Bianchi and Parrilli, 2002b; Parrilli, 2004a). Within these linkage policies, ‘cluster policies’ – that will be dealt with as a substantial part of this volume – assume a strategic position, since these latter build upon a large number of existing geographical and sectoral urban agglomerations (Garofoli, 1992; Schmitz and Musick, 1994; Markusen, 1996; van Dijk and Rabellotti, 1997) and push traditional and modern enterprises to reap the benefits of passive joint localisation (i.e., external economies) and consciously pursued joint actions (Schmitz, 1992; 1995a). In spite of the still evident differences existing among member countries (Parrilli, 2005a), these latter have been able to converge towards an increasing homogeneity of approach to SME development (see Table 2.1). This has
34
Table 3.1:
EU actions implemented from May 1997 to end of 1999 Administ. incentive
Belgium Denmark Germany Greek Spain France Ireland Italy Luxemb. Holland Austria Portugal Finland UK Switzerland Norway
X X X X X X X X X X X X X X X X
Delay in payment
X
X X
X X
Finance
X X X X X X X X X X X X X X X X
Internationalization
X X X X X X X X X X X X X X X X
Information
X X X X X X X X X X X X X X X X
Employment training X X X X X X X X X X X X X X X
R&D
Entrepreneurship
X X X X X X X X X X X X X X X X
X X X X X X X X X X X X X X
Source: Observatoire des PME Européennes, 2000: 17. This extensive European approach is maintained today (see Bianchi et al., 2006: 751, on the basis of the information provided in Observatory of European SMEs, 2003).
Inclusion Versus Fragmentation
35
taken place during almost fifty years of harmonisation within the European Union and has benefited greatly from the parallel process of commonly regulated openings, which the European countries chose to follow and that permitted the majority of the national productive systems to gradually adjust to the new context of global markets (Bianchi, 1998; 2000). On these bases, SMEs in Europe (especially SMEs based in clusters and industrial districts) have become global competitors. During the period 1988–2001, the turnover growth rates achieved a total of 2.2 percent and 2.3 percent in micro firms and SMEs respectively, just a little below the rate achieved by large firms (2.6 percent). These rates were induced by the export market, in which the growth rates were 5.8 percent, 6.3 percent and 6.2 percent respectively (Observatory of European SMEs, 2002/2: 22). The period 1990–93 represented a phase of crisis and economic adjustment. The overall employment level suffered a general reduction of 1.2 percent among micro, small and medium firms, while the value was positive and ranked at 0.5 percent in the 1993–98 period. In this case, it is interesting to note that large firms performed worse than SMEs, throughout 1988–2001, the first group showed an employment reduction of 0.1 percent, while micro enterprises and SMEs achieved a growth of 0.3 percent (ibid.: 26–8). In terms of profitability, the SMEs reached similar levels as large firms, with an annual increase of 0.4 percent (Observatoire des PME Européennes, 2000: 6). With reference to firm creation, 20 percent of micro firms are less than five years old, while this percentage falls to 8 percent amongst firms with less than 50 workers (small firms), and 5 percent among firms with up to 250 workers (medium firms). The larger the firm the bigger the percentage of firms that are more than 10 years old (59 percent, 81 percent and 88 percent respectively: Observatory of European SMEs, 2002/1: 9). Overall, these data indicate a good level of turnover and firm creation among SMEs, which help to structure a dynamic and competitive economy. These data and considerations highlight the fact that SMEs are not a weak player within the European economy, but participate actively in the process of growth. This outcome has been possible on several bases, which are rooted in the endogenous strength of local production systems and the supportive SME policy framework, which have been complemented by a stable macroeconomic regime.
3. The main processes in Latin America: liberalisation and industrial destructuring From 1982, the foreign debt crisis exploded throughout the Latin American region. The structural adjustments demanded by multilateral financial
36
SME Cluster Development
organisations implied abandoning the previous interventionist and protectionist state practices and spurred on the (unilateral) liberalisation of the internal market, the privatisation of state properties and enterprises, the reduction of public institutions and employment, the control of inflation through restrictive monetary policies and the control of the deficit and the foreign debt through restrictive fiscal policies (Singh, 1992). After two decades, many countries in Latin America and outside have been unable to follow a steady path to development (UNDP, 2000). A debate – promoted by the United Nations through its United Nations Development Programme (UNDP) and its Human Development Index (HDI) – arose about new practices needed for development (Cornia et al., 1989) and pushed the World Bank to partially modify its approach by incorporating project lines on the strengthening of public institutions as a means to generate a more efficient working of the market (World Bank, 1993; Stiglitz, 2001). Despite this, market liberalisation has been maintained and the unilateral opening has been considered the main principle for the whole set of more or less indebted developing countries that wanted to maintain good relations with the main multilateral financial organisations and the most important donor countries (Rodrik, 1994; Sugden and Wilson, 2002). In this way, many national production systems were caught unprepared and unprotected in the face of growing globalisation and the new competition that started to flow freely to every country, jeopardising the tiny spaces available to local SMEs (CEPAL, 2001). The entry of The People’s Republic of China into the World Trade Organization is just the last factor of the new global competition that is threatening to marginalise all players that cannot manage new production and trade practices. This perspective does not take into account alternative routes for growth and development, such as that followed by the European Community. The gradual and regulated opening that has taken place in the ‘old continent’ from the 1960s onwards allowed the weakest players (e.g., the SMEs) to join the market well prepared (Bertini, 1998; Bianchi, 2000). In Latin America, the liberalisation process is framed within very short periods of time (Parrilli, 2005). This approach does not give enough time to work for a progressive and gradual strengthening of the national production system (Bianchi, 2002a). In contrast, the natural consequence of market liberalisation is to block SME development by strengthening the conditions for some market failures to emerge; these in particular prevent SMEs from accessing finance and technology which in the end hinder their access to foreign markets.
Inclusion Versus Fragmentation
37
In order to explore our hypothesis on the contrasting development strategies adopted in the two continents and their divergent development trajectories, we analyse the weight and growth of small and large firms within selected economies, the policies that have been implemented in these contexts and the adjustment that SMEs have gone through. 3.1 Case study: Argentina Argentina was the last of the Latin American countries that fell into a deep crisis in the late 1990s. There are different views on this situation: there are, on the one hand, those observers that identify the mismanagement of national finances as the root of the crisis (Perry and Serven, 2002; Rogoff, 2004); on the other hand, those observers that consider this aspect as the tip of a problem rooted in the particular policy-making that has been implemented over the past two decades and the resulting production structure (Chudnovsky et al., 1999; Bianchi, 2002a). In our opinion, this second stream of analysis is particularly interesting. It emphasises that policy-makers did not focus enough on strengthening the national industrial system as a whole. This weak attention led to a sharp divide between a modern sector of mainly large and medium enterprises versus a traditional sector of mainly micro and small firms (Bianchi et al., 2000; Bianchi and Parrilli, 2002b). This fracture has been weakening the capacity of the national system to respond to international challenges coming in the new age of globalisation and has made it more easily attackable by speculative waves (e.g., the financial crisis in 2000) (Bianchi, 2002a). The rest of this section now analyses how this process has taken place. In the 1970s, the alliance between the agricultural oligarchy and the military led to the rejection of the import substitution industrialisation model implemented in the former decades. These sectors proposed a different view, focused on ‘agricultural specialisation and the proto-liberalism directed by a natural order’ (Bianchi, 2002a). At the beginning of the authoritarian phase (1976–83), the government promoted the resumption of production, particularly of consumer goods and investments. Nonetheless, in December 1978, restrictive policies led to recession; therefore, the military government suddenly decided to open up the economy, by abolishing custom protections, announcing currency devaluation and liberalizing the capital market. The immediate effect was a loss of capital, of both multinationals established in Argentina and national capital (ibid.).
38
SME Cluster Development
The opening of the internal market exposed the national firms to international competition. Conversely, the internal growth of interest rates, which was induced to control inflation, generated high debt levels, that brought about the worst crisis of the industrial sector in Argentinean history (Kosakoff, 2000: 46). In this context, a reorganisation process started and lasted for the entire mega-inflation period (the 1980s up to 1991). The system was polarised between a small minority of mainly large companies that were able to renew their technology, and the majority of mainly small and medium enterprises that reduced their activities to a minimum, or simply became importers or intermediaries. The multinationals left the country leading to an overall suspension in the entry of foreign capital; a simultaneous loss of capital overseas and the beginning of intensive financial speculation encouraged what has been called ‘the economy of sacking’ (O’Donnell, 1982; Bianchi, 2002a). In order to manage the crisis, the democratic government had to devalue continuously. In this way, it could boost exports in order to generate enough profits to compensate for the increasing costs of servicing the foreign debt. Inflation became ‘mega-inflation’, with further internal effects that led to destructuring production organisation in sectors that had expanded in the phase of import substitution (Heymann and Leijonhufvud, 1992). At the same time, the informal economy increased and a new industrial specialisation was consolidated, based upon the production of intermediate goods and inputs of production, such as petrochemicals, iron and steel (Kosakoff, 1993; Chudnovsky et al., 1999). One key effect of mega-inflation and institutional instability was the process of de-industrialisation, which can be measured in terms of the value of manufactured products related to total GDP. In the 1980s, the share of manufacturing production reached its 1940s levels, leading the GDP to fell significantly. The effects of uncertainty on mechanics and machine tools, but also on textiles, furniture and non-metallic goods led to the decline of both production and its quota on total industrial production, with a parallel increase in informal activities and traditional trade and services (Bianchi, 2002a). Manufacturing production, which in 1980 had reached about 30 percent of GDP, declined to about 20 percent in 1990 and to an even lower 16 percent in 2001 (CEPAL, 2003: 236–7). De-industrialisation was accompanied by a dramatic deterioration of the macroeconomic context. Between 1980 and 1990, GDP decreased by 9.4 percent, industrial production by 24 percent, consumer goods by 15.8 percent, imports and investments decreased respectively by 58.9 percent and 70.1 percent,
Inclusion Versus Fragmentation
39
per capita GDP was reduced by 25 percent and only exports grew by 78 percent (ibid.). On the whole, these data indicate the effects of the shift proposed by the World Bank and the IMF towards the comparative advantage model, which led to an increase in public and private policies and investments in agricultural production and natural resource-based industries, often dedicated to a first processing of raw materials and their export, which add little value, employment and knowledge (Reinhardt and Peres, 2000). The process of de-industrialisation can be seen through the decrease in the total number of industrial enterprises and employees. In 1984, there were 74,000 micro industries, 33,000 SMEs and 2,000 large industries; in 1993 these had shrunk to 64,000, 25,000 and 1,500. In terms of employment, micro industries accounted for 176,000 workers, SMEs 593,000 and large industries 611,000. In 1993, they were 140,000, 449,000 and 418,000 (Moori-Konig et al., 2002: 38). The most traditional micro industries and SMEs suffered a dramatic decrease in production. The most affected sectors had been the traditional production of shoes and leather, furniture and toys; while sectors occupied by large firms (e.g., chemicals, food and beverages, oil, automotive, etc.) improved their position within the national production structure (ibid.). In the past ten years, the policies applied by the government have been directed at organising an efficient programme of privatisation, thus dismantling direct state intervention in the economy. By the end of 1992, privatisation accounted for more than $16 billion of assets, some of which were paid in real outlays, but most through debt transfers from public to private firms. Moreover, foreign firms were allowed to transform the foreign debt of the country in capitalisation of national firms for more than $600 million (Kosakoff, 1993: 260). This process continued throughout the 1990s and supported the participation of large national economic groups in the privatisation process, often through alliances with international groups. It is the case of Techint, Perez Companc, Soldati, Astra, Macri. One effect of this programme was the consolidation of large groups, producers of industrial commodities (e.g., oil, steel, chemicals), as well as public services (e.g., telecommunications, transports, water, gas and electricity) (Bisang et al., 1995; Bianchi, 2002a). Analysing concentration in sales and employment within the large firms in the national economy helps to understand the way the process of restructuring/destructuring has taken place in Argentina. In particular, a growing concentration of industrial production among large economic groups can be observed, together with a decreasing rate of employment
40
SME Cluster Development
in those modern firms (against the opposite complementary trend in traditional SMEs): Table 3.2: in 1997
Sales and employment concentration in major economic groups
Total main 18 groups Total main 100 firms Total main 500 firms Total main 1,000 firms
Sales (000 pesos)
Employees
26,275 27,188 31,629 32,110
111,924 93,535 125,476 129,353
Source: Chudnovsky et al., 1999: 97.
These data indicate the relative strength acquired by these groups within national industrial production. In 1997, 18 major groups accounted for 96 percent of total sales of the first 100 firms, 83 percent of the first 500 firms and 81 percent of the main 1,000 firms. In terms of employment, they accounted for 89 percent of total employment generated by the first 500 firms and 86 percent of the main 1,000 firms (Chudnovsky et al., 1999). Nevertheless, considering the absolute number gives a different outcome. Indeed, 111,000 workers represent a low value for the 18 largest economic groups in the country, especially if compared to the 26,000 million pesos (or dollars for the parity) that they sold in 1997. The afore-mentioned data about the reduction in the number of firms and the employment concentration give a measure of the country’s production structure, which increases the risk of a growing dualism between a smaller and smaller group of modern, competitive, technology-intensive and internationalised sector of large national and foreign-owned companies versus a larger and larger traditional, labour-intensive and national market-oriented sector of mainly micro, small and medium-sized enterprises. The lack of industrial policies targeting this growing fracture and the need to strengthen the productive structure of the country as a whole complemented this market trend and pushed more easily Argentina into the crisis of the early 2000s (Bianchi, 2002a; Bianchi and Parrilli, 2002b). 3.2 Case study: Chile Chile is another example of the effect of a poor industrial policy oriented to strengthen the production structure of the country. This weakness helps to explain the recent serious problems faced by the country’s production system and its SMEs with respect to the increasing globalisation
Inclusion Versus Fragmentation
41
of markets and the entry of Asian products in Chile. This section shows this process synthetically. In Chile, as elsewhere, micro, small and medium firms are the most numerous enterprises, and represent a very important part of the economy in terms of employment and gross domestic product. Table 3.3 shows data that confirm this consideration. These data indicate quite clearly the importance of these kinds of firms within the national production system. As elsewhere, micro enterprises are the most numerous firms and play an important role in creating employment (90 percent of total industrial employment in 1997); to a much lesser extent they also produce a proportion of national sales and incomes. Table 3.3 shows that they generated 20 percent of sales in 2000, falling from 28 percent in 1997. Looking at the data, which were also corroborated by the qualitative information collected through an extensive consultancy work in 2002 (Bianchi and Parrilli, 2002a), it is evident that the national population of firms were passing through a phase of crisis and structural adjustment. In particular, micro, small and medium firms were suffering, while large companies increased both in number and sales. This situation is part of a national stagnation that has taken place in the past few years and which has also been triggered by the specific policy perspective that the Chilean government has been implementing over the past thirty years. In reality, for more than 20 years this framework worked very well, making Chile the most open, efficient and transparent Latin American economy. Even in terms of human development, the United Nations recognise the success of Chile, which ranks among the first Latin American countries and around 35th in the world (UNDP, 2001). Table 3.3:
The enterprises in Chile in 1997 and 2000 Number
1997
2000
Micro firms 432,400 521,820 SMEs 89,700 61,337 Large firms 4,800 4,200
Employment (%)
Sales Sales/enterprise (million UF) (thousands UF)
1997 2000/01 1997 2000 (growth) 40 50 10
⫺5.8 ⫺1.2 n.a.
201 135 1,085 786 3,287 3,599
1997
2000
465 258 12,095 12,822 684,700 857,000
Note: 1 UF is about US$ 25. Source: for 1997 CORFO, Ministerio De Economia (1999), Comité Público-Privado, Santiago, Chile. For 2000, INE-Gobierno-BancoEstado (2001) and CORFO, first half-yearly Survey of micro enterprises and third Survey of SMEs, 2001.
42
SME Cluster Development
Recently, however, this policy approach of unilateral opening has led to disturbing developments for the country. World competitors such as China and Korea have used the Chilean market as the entry door to Latin America. This situation created significant problems for national producers; in fact, the traditional sectors – textiles, clothing and shoes – have been the most affected and the sector of traditional firms (mainly micro, small and medium enterprises) have lost their competitive margins and/or are going to close down their operations. The modern sector (e.g., software, pharmaceutical) seems to be the most dynamic sector, together with some traditional natural resource-based sectors, such as copper, wine and fruits (Reinhardt and Peres, 2000; Cimoli and Katz, 2001). These considerations do not mean that the Chilean government did not intervene in the economic process to push the competitiveness of its firms. Indeed, recent studies point out the key role of CORFO, the development corporation of the Ministry of Economy, Energy and Mining in the promotion of the national production system (Bianchi and Parrilli, 2002b; Dini and Stumpo, 2002; see also Chapter 2). But government intervention has been very limited and has tended to respond to specific demands of some relevant economic groups, rather than forming part of a strategic approach to industrial development. Nowadays, this policy approach presents some weaknesses that do not put the national production system in a condition to respond to the international market challenges (Bianchi and Parrilli, 2002a; 2002b; see also Chapter 2). On the whole, the statistics express the effects of the crisis that has been affecting the country and the Latin American region in the past few years. The numbers of large firms and SMEs have been shrinking, due to the loss of competitiveness of the Chilean system in the face of imports from Asia. Simultaneously, the remaining large firms have increased their sales. The number of micro enterprises has been increasing, likely with the need of the population to create some niche solutions, often for survival, to raise their income and employment, but, at the same time, their sales have been shrinking (INE-Gobierno-BancoEstado, 2001). These indications mark the tendency of an increasing dualism within the national production system. Market liberalisation and market failures offer different development opportunities to two sectors within the national economy. This leads a rather tiny group of large, medium and small modern national firms to increase their competitiveness and alliances within international markets and companies, while pushes the wider majority of national micro, small and medium firms to continue producing in a traditional way for the local and national market. The latter are the ones that suffer the most from the imports of
Inclusion Versus Fragmentation
43
cheaper foreign goods for instance from China (BCC, 2002; see also Chapter 2). At present, the government is trying to identify and solve the main problem of most national firms (i.e., competitiveness). The policymakers realise the risks of splitting the national economy and losing the contribution to development of a large number of producers. This is the reason why they are currently discussing the need to promote structural policies for this traditional sector of micro, small and medium firms. At the end of 2001, they began to promote a detailed analysis of the normative fields that affect this sector’s place in the economy, as well as the market failures to be filled, in order to permit these firms to work in a more competitive way (Bianchi and Parrilli, 2002a). At the same time, the government created a ‘Statute for the Small Firm’ to define legal basis that promotes stronger actions for the growth of these traditional enterprises (Ministerio de Economía, 2002). On the whole, this policy approach presents weaknesses, because despite being oriented towards upgrading the competitive ability of the production system, it somehow neglects the effective demand of the market, which is taken for granted through the unilateral opening of the economy. The conflict of interests that exists between the modern sector of the economy (i.e., large companies), more interested in maintaining a deregulated market, and small firms, demanding some kind of protection from abroad, is a signal of an open question and problem for Chilean policy-makers and their production system. 3.3 Case study: Nicaragua Nicaragua is the third country that shows the above-mentioned fracture. The policies implemented by recent governments have not promoted a strengthening in the industrial structure of the country. This led the production system to split between a tiny modern sector and a big traditional SME sector. The most recent statistics on industrial production presented 1,800 small and medium industries at the national level, and 17,500 micro industries (firms that have less than four workers).16 These numbers are just a part of the bigger enterprise universe in Nicaragua (enterprises in agriculture, trade and services), that account for more than 158,000 firms, 99 percent of which are micro, small and medium enterprises (Banco Central de Nicaragua-GTZ-MEDE, 1998). The observation of various indicators (see Table 3.4) emphasises that SMEs and micro enterprises constitute an important sector within national
44
SME Cluster Development
Table 3.4:
The structure of the national industry by size, 1996 Number of enterprises
Large industries SMEs industries Micro industries Total industries
Number of workers
260 1,805 17,509 19,574
65,133 12,159 26,577 103,869
Source: Data from the manufacturing survey 1996 of BCN (Parrilli, 2002b).
industrial activity, which achieved more than 37 percent of total industrial employment in 1996. The SMEs in Nicaragua produce mainly baking products, metallic products (e.g., railings, roofs, window frames, etc.), furniture, chemical products (e.g., soap, and so on), shoes and non-metallic mineral products (e.g., cinder blocks). Of these, wood and furniture, bakery, leather and shoes are particularly relevant, since they produce the largest part of added sectoral value and employment (BCN, 1998; in Parrilli, 1999a; 2002b). In terms of added value, these sectors did not generate high values. The most relevant was bakery with about $10 million, while furniture and metallic products produced more than $6 million each. But the number of firms in each of these sectors was quite high (e.g., 1500 in furniture production), which means that the added value per firm is extremely low. The SMEs also produced low values in terms of exports: 8.8 million dollars in 1996 (BCN, 1998, in Parrilli, 2002b). Only in the past few years this value has increased significantly, but this is mostly due to SMEs in dairy production, which export more than $20 million alone (Artola and Parrilli, 2006). The reasons for these general weaknesses are quite evident. The SMEs in Nicaragua, are usually craft enterprises, which are small workshops with scarce and often obsolete machinery that produce reduced volumes of goods with poor qualitative standards, which are typically sold in the small local and/or national market. They have never participated in relevant transactions, neither inside nor outside local markets (Parrilli, 1999a; 2002b). Table 3.5 shows a changing trend that took place over the second half of the 1990s. The weight of foreign direct investments increased steadily, while the opposite occurred with exports. This process has been strengthening the lobbying capacity of the first group of players in the face of the government and the second group. The unilateral lowering of trade barriers that Nicaragua undertook over the past decade represents well
Inclusion Versus Fragmentation Table 3.5:
45
Weight of exports versus foreign direct investment ($million)
Exports FDI Ratio export/FDI
1996
1997
1998
1999
669.0 97.0 6.9
625.9 173.1 3.7
573.2 183.7 3.1
543.8 300.0 1.8
Source: Author’s compilation based upon data available at Banco Central de Nicaragua, 1998–2000.
this increasing power and the anti-productive bias of economic policy, which benefits national importers and foreign producers and negatively affects national producers (Solórzano, 2000; Parrilli, 2003). This aspect points to a lack of industrial development strategies and policies. Past governments emphasised the role of agricultural production as a way to strengthen the national economy, but they did not pay attention to the manufacturing sector, which transforms primary production and generates more added value, knowledge and employment, also through the activation of further niches of work and skills in the modern service sector related to dynamic industrial production (Sylos Labini, 2000). In this way, the weight of industry within GDP has progressively diminished in the past twenty years (Laenen, 1988; Reyes, 1993; BCN, 2002), in contrast to past trends in advanced countries and recent trends in the newly industrialised countries (Chenery and Syrquin, 1980; Cipolla, 1996; Parrilli, 1999a; Maddison, 2001). The failure of Nicaraguan governments to address these objectives is reflected in the weaknesses of various production chains, especially those linked to SMEs, which are less able to integrate the different phases of production. For example, the production chain that goes from cattle farming to the production of shoes is broken at the level of the tanning industry. Very little leather is processed and much less remains within the country. That is why SMEs have given up producing leather shoes and shifted to the segment of shoes in synthetic materials, which requires the importation of raw materials (Narváez, 2001; Parrilli, 2001). In the case of the wood and furniture production chain, there is a bottleneck at the level of woodworking. In this case, factories prefer exporting lumber rather than selling it to furniture makers (Barahona et al., 1998; Parrilli, 2002a). Similar situations occur in other production chains (e.g., coffee and fruit). The public institutions (the National Chamber of Small Enterprises – CONAPI – in the 1980s and the Support Programme for Micro Enterprises/SMEs – PAMIC/INPYME – in the 1990s and the 2000s) and private organisations (mainly NGOs) that work for the development of
46
SME Cluster Development
SMEs have focused their attention mainly on micro enterprises, since these are numerous, weak and show more visibly their need for support (García, 1999; MIFIC, 2001a; Parrilli, 2003). This practical approach did not take into account the lessons learnt from successful international experiences (e.g., industrial districts and networks of innovative SMEs), which indicate that development sprung mostly from SMEs, rather than from micro enterprises (Piore and Sabel, 1984; Best, 1990; Schmitz, 1992; Parrilli, 1999a). In fact, the strength of SMEs points also to an important internal division and specialisation of labour, that allows allocation of specialised personnel to new key activities that micro enterprises find more difficult to develop (e.g., marketing, design and innovation). In this way, the division and growing dualism among a very tiny sector of large modern firms (often commercial and importing companies) and the vast majority of micro and small traditional firms has been maintained over time and has created a gap that does not help the country to move forward as a whole by creating a common industrial development strategy (Bianchi and Parrilli, 2002b). Later on in this volume, Chapter 4 shows the Nicaraguan case in more detail with respect to clustering and cluster policies. Through that chapter, this work discusses the need for (cluster) policies to reactivate the traditional sector of SMEs to promote growth within clusters and, indirectly, within the national industry and production structure. 3.4 Case study: Costa Rica Within this regional trend represented by the cases of Argentina, Chile and Nicaragua, Costa Rica represents the exception. This is a small economy that has gone through a phase of expansion over the past 20 years. The support of the US government in the 1980s helped to implement the coherent economic policy that the national government was shaping and which targeted the opening of the economy and the investing in strategic sectors that nowadays produce very interesting results (BCCR, 2002). Since 1985 the national policies have primarily targeted the increase of exports. These policies promoted the insertion of several Costa Rican firms in the international market, regardless of their size and sector (Aguilar et al., 1998). Surprisingly, this process was aided by the Central American situation. During the 1980s, Nicaragua, Guatemala and El Salvador were in the middle of tough civil conflicts, which prevented Costa Rica from concentrating its export effort towards these smaller markets. Thus, the country targeted bigger and more demanding markets, such as the United States and Canada, which promoted the standardisation and restructuring process of the country’s industrial firms (Aguilar et al., 1998: 34).
Inclusion Versus Fragmentation
47
At the same time, the government made significant efforts to promote foreign direct investment within the country by supporting not only investments in traditional sectors, but also in strategic industries, such as software. For example, in recent years Costa Rica has been able to promote the creation of a software cluster around the transnational company INTEL in the San José airport area through a set of public investments in the upgrading of infrastructures (e.g., the international airport, the setting up of industrial areas) and human resources (e.g., engineering courses at university). As a consequence of these kinds of policies, the employment trend is different from that of other Latin American countries, since no difference can be seen in the behaviour of large, medium, small and micro firms during 1986–97 (see Table 3.6). Industrial micro enterprises decreased their employment by about 10 percent, small firms by about 4 percent, medium firms by about 7 percent and large firms by about 60 percent (Aguilar et al., 1998: 6).17 All these data appear to be quite different from the previous case studies. Costa Rica did not present signals of crisis until three years ago, when worldwide economic stagnation had permeated almost everywhere (BCCR, 2002). The above-mentioned data do not present signals of an economic split between a modern competitive sector and a traditional sector of micro firms and SMEs. Indeed, apart from micro industries, which have had a negative trend (in terms of number of firms) from 1986 onwards, SMEs and large industries have all grown. The policy target of promoting the export capacity of the national production system regardless of the size of firms, prevented the creation of a barrier between these two worlds and favoured more integration among different types of firms and sectors. In fact, all kinds of firms had the possibility to access government incentives, which helped them to upgrade their productive and competitive capacities (Aguilar et al., 1998; Wong Chacon, 2002).
Table 3.6:
SMEs participation in the manufacturing sector
Micro enterprises Small firms Medium firms Large firms
1986
1990
1997
2556 1851 376 212
2361 1524 412 271
2177 1996 402 308
Source: Based on Aguilar et al., 1998: 5.
48
SME Cluster Development
Building upon a common target (i.e., the strengthening of the production and export capacity of national firms) eased this process. This approach is very different from that implemented in other countries, where some important economic sectors had different and conflicting objectives with the national production system (see, for instance, the national importers of brand names and the transnational companies under free exporting regimes in the case of Nicaragua). This process has led all types of firms to increase their place in the international market. For instance, from 1990 to 1997, the percentage of exports on total sales passed from 14.98 percent to 25.54 percent (Wong Chacon, 2002: 248). Another interesting factor is represented by the type of sectors that have been involved in the international market. These are heavy machinery (from 26 percent to 31 percent of total sales), general machinery (from 23 percent to 39 percent), plastics (from 27 percent to 34 percent), metallic products (from 14 percent to 22 percent), chemicals (from 13 percent to 41 percent), and wood (from 11 percent to 30 percent) (ibid.). Another very significant element of this country case study refers to the productivity gap among large firms and SMEs. In other countries (e.g., Chile and Nicaragua) there is a widening gap and considerable differences, which are not visible in the case of Costa Rica. In this case, from 1990 to 1997, SMEs increased their productivity and reduced the gap with large firms (Peres and Stumpo, 2000; 2002; Wong Chacon, 2002: 246). In particular, large firms have increased steadily from an average of $19,760 (in 1991 terms) to an average of $23,434 per worker; small and medium firms have increased from $12,409 in 1990 to $17,099 in 1997. In percentage terms, it means that the first have increased by 18.5 percent, while the second increased by 37.8 percent (Wong Chacon, 2002). On the whole, Costa Rica shows a different trend to the other three countries (and the Latin American region in general). The country structure presents a number of competitive small, medium and large firms inserted in both traditional manufacturing and high-technology sectors. Their success is linked to a national policy focused to a common target, which is competing in the international market. This clear target was set up and pursued by the government from 1985 onwards. This treated firms homogeneously and spurred linkages among firms, clusters and production chains as a means to achieve the process of industrial restructuring in the mid term (Aguilar et al., 1998). 3.5 A synthetic view of the Latin American response This chapter analyses an important effect of the liberalisation process that has taken place in Latin America over the past 20 years. Many Latin
Inclusion Versus Fragmentation
49
American economies reacted to the shock of unilateral opening by destructuring their industrial systems. The cases of Argentina, Chile and Nicaragua seem to show this (see sections 3.1, 3.2 and 3.3). A restricted number of large, medium and small modern firms have been able to compete in the global economy; they have diminished in number and increased their sales. In contrast, most existing small and micro enterprises in traditional production and those born from the restructuring of large firms (i.e., employment reduction) have managed to survive at the local level with poor productivity performances and often by joining the informal economy. They have increased in number (especially via micro enterprises) and decreased in sales (this trend is particularly strong in the case of Chile). This process hides the risk of an important fracture within the economy (and society). In essence, the few large firms that remain competitive do so by increasing their productivity without developing innovations and complementariness with other firms. They merely focus on the reduction of costs, obtained through on-the-spot market transactions (at the interfirm level) and the simple substitution of capital for labour (at the intra-firm level). They develop production and market transactions that neglect the local level and that take place at the international level (e.g., with global suppliers and buyers). This behaviour limits the ability of these modern firms to lead the economy to growth because they end up representing a sector that is too small and has very few relations with the rest of the local production system (i.e., traditional enterprises). Argentina, Chile and Nicaragua seem to be good examples of this trend (Bianchi, 2002a; Bianchi and Parrilli, 2002a; 2002b; Parrilli, 2004a). Simultaneously, most SMEs cannot access the credit market to undertake the same route to production upgrading (i.e., substitution of labour for capital) and thus tend to lose productivity. Also for this reason, these enterprises are only capable of working with local partners in both production (i.e., purchasing inputs, producing and subcontracting) and commercialisation (i.e., local buyers and consumers). This situation reduces their sales and incomes and ends up affecting the overall stability of the social environment (Pyke and Sengenberger; 1990; Kaplinsky and Readman, 2001). In conclusion, a large group of traditional SMEs and micro enterprises remain isolated from a small group of mainly large (but also small and medium) modern firms. This situation can lead to a stiff dualism that may also be protecting the interests of regressive coalitions and transnational powers, but that stops or, at least, slows down the whole process of industrialisation and economic development (Cowling and Sugden, 1997; Bianchi, 1998).
50
SME Cluster Development
In the context of Latin America, Costa Rica represents a very interesting example since it has been able to adopt a different growth path. It was able to strategically and systemically organise its industrial development process around a set of specific priorities and policies in which all kinds of enterprises were included. As a consequence, this country could overcome the risk of growing economic and social dualisms and conflicts.
4. A strategic approach to industrial development In relevant European experiences (widely in the case of Italy and in a more scattered form in France, Germany, Belgium, Denmark, Spain) the transformation of the 1960s and the 1970s, led to processes of competitive restructuring of large firms. Their massive employment reductions were compensated by the significant rise of SMEs. This situation was overcome through the progressive emersion of forms of networking (e.g., clustering, associations, subcontracting, consortia) that spurred the process of individual specialisation and functional complementariness among firms (Piore and Sabel, 1984; Best, 1990; Becattini, 2000a). In this way, the system incentivised the growth of new leaders (even SMEs) in a condition to lead the growth process through innovation, maintaining flexible practices and conquering new markets. The systemic framework for industrial and SME development organised in Europe over the past decade is currently producing important results in several European countries and SMEs and their capacity to contribute to regional development (see section 2 in this chapter). This outcome makes it a very interesting benchmark practice for industrial development policies in many Latin American countries. This experience may be encapsulated within the concept of ‘active industrial strategies’ applied by many industrialised countries in a more or less explicit way over the past several decades. Notwithstanding, this discussion does not mean that the macroeconomic framework of market liberalisation implemented by neoliberal governments and international organisations has been completely negative. What we argue in this chapter is that the neoliberal policy has not been sufficient to promote development, because it has neglected a variety of problems that can arise in liberalised markets (e.g., market failures, market concentration, and so on). Having recognised such a shortcoming, we would advocate a more proactive role of the state and of local governments to help national and local production systems to develop as whole. If this issue is not addressed properly, the likely outcome will be a ‘zero-sum-game’ where a few firms succeed (typically the large modern companies), while the large majority lose
Inclusion Versus Fragmentation
51
competitiveness and are marginalised from the market (typically the micro, small and medium traditional enterprises). It is evident that a particular effort is necessary to connect the modern and traditional sectors. Although neoliberal policies and market liberalisation respond to the need to stabilise the national macro economy, to control public expenditure and foreign debt, and to attract inward foreign investment, they have not been enough to produce a balanced and homogeneous development in the production system. This is why what is required is also a national development strategy that voluntarily and rationally addresses the problem, pulling together all national forces and actors to resolve it. At present, most national and international development organisations work in a rather ‘fragmented’ way, which does not promote the structural reinforcement of the national economy. A significant example of an increasingly integrated industrial strategy can be identified in the policies of most European countries (see Table 3.1), which – despite traditional differences, which are still present – have structured a more uniform production system (i.e., fighting possible market failures, information asymmetries and, as a consequence, the exacerbations of a kind of ex-post dualism). These policies are of three types. To begin with, there are actions that can promote the competitiveness of the traditional sector. Worldwide, one of the most interesting examples of this type of actions is the programme of Mise-à-Niveau under implementation in North Africa with the support of the European Union. This programme encourages all companies and institutions that interact in a local system to work with identical quality standards, sufficient to penetrate the European market (Bianchi, 2000; Ministère de Economie de Tunisie, 2000; Di Tommaso et al., 2001). This programme utilises several development instruments (e.g., technical assistance, credit for technology purchase, human resource training, and so on) and integrates them in a national industrial upgrading program. Of course, several other programs might be considered appropriate, such as the Sabatini Law or credit consortia applied in Italy (Federconfidi, 1997; Bertini, 1998), but all of them must maintain the general objective of promoting the collective consolidation of the traditional sector. This is why these policies are called ‘structural policies’ (European Commission, 2002). Furthermore, effort is required to increase the number of small firms in international networks of innovative enterprises. For example, academic spin-offs with a high level of technological content allow a system of firms to penetrate the knowledge economy, upgrade human and physical resources, and identify new opportunities for change in the economic
52
SME Cluster Development
structure. In this way, national and local productive systems can refocus competitive and growth goals on higher value-added sectors and reduce participation in low-tech manufacturing sectors, in which strong price competition crowds out many national enterprises (Baroncelli, 2001; European Union, 2001). This calls for a critical criss-crossing of relations among local, national and international institutions and (groups, consortia, associations of) enterprises in order to develop a more dynamic and coherent environment to nurture these initiatives. This second type of policy can be called ‘high-technology policy’, otherwise called ‘innovation policies’ (European Commission, 1998). Finally, structural and innovation policies require a third strategy aimed at linking the traditional and the modern sectors, in order to create a unique integrated system of national firms. This is the case, for example, in the subcontracting relations that take place among large and small firms (very successful in Japan: Caddy, 1998) or among only small firms (very successful in Italy: Lazerson, 1990; Innocenti and Labory, 2002). Such relations allow economic actors to work together in a specific project of competitive development. They also favour the creation of higher levels of specialisation and complementariness among firms, essential to integrate them in competitive networks producing for the open market (Bianchi and Parrilli, 2002a). In this regard, it is important to recognise the role that public institutions can play in upgrading these relations, even through specific laws for the defense of small subcontracting firms, as occurred in Japan in the early 1950s and in Italy in recent years (Caddy, 1998; Bianchi and Parrilli, 2002a). These policies, which focus on stronger linkages between traditional and modern firms, can be called ‘linkage policies’. Through them, it becomes easier to avoid both two-speed development in the country, and increasing income and welfare gaps within the population. The strategic approach commended herein is more likely to result in the fulfilment of growth potential in the traditional sector of the production system, namely, among micro and small enterprises. This potential lies in more efficient and further agglomerated economies. Efficiency results from the individual specialisation of enterprises alongside their collective complementariness in clusters of firms. The advantages of agglomeration are generated by the networking of territorial and sectoral chains of production, making possible economies of scale and scope even without the leadership of large enterprises (Brusco, 1982; Piore and Sabel, 1984; Becattini, 1990; Best, 1990; Pyke and Sengenberger, 1990; Schmitz, 1992; Nadvi and Schmitz, 1999). The industrial policies of most European countries over the past two decades reflect these considerations.
Inclusion Versus Fragmentation
53
Within linkage policies, cluster development policies are particularly important for two main reasons. On the one hand, clusters represent very common local production systems in both developed and developing countries. This situation stresses the need to set up strategies and actions to promote their further growth. On the other hand, clusters represent the easiest form to strengthen SME position within global markets, since they create those agglomeration economies that the literature indicates as the main force to help SMEs face the competition from large firms in the open market (Schmitz, 1992; Bianchi, 1998; Sylos Labini, 2000). Chapters 4, 5 and 6 focus specifically on this issue, which constitutes the second core topic of this book. On the whole, this systemic framework for development policy can be better realised when these greater linkages are supported by a consensus around the need for cooperation among parties to the project (e.g., enterprises, business associations, public agencies, universities, and so on). This cooperation refers to collective complementariness, and the creation of an efficient framework for productive and commercial networks and chains. The generation of competitive economic systems and the consequent development of the country, in Latin America as well as in other parts of the developing world, is more likely when all firms and public support institutions avoid fractures and cooperate for the common goal of local and national economic development.
This page intentionally left blank
Part II Small and Medium Enterprise Cluster Development
This page intentionally left blank
4 Different Theoretical Approaches to SME Cluster Development: Relevance in the Case of Nicaragua
1. Introduction This chapter constitutes the connection between the first part of this work and the second. It links the first part on SMEs to the second part on SME clusters and SME cluster development. In doing so, in this chapter we analyse the concept of ‘cluster’ and describe its main features, making reference to the main theoretical approaches to cluster development. In this work, the concept of ‘cluster’ is taken in a broad sense. It refers to ‘geographical and sectoral agglomerations of firms’ (Schmitz, 1992; 1995b). In this sense it includes a variety of possible configurations that have been identified by Markusen (1996), Knorringa (2002), Guerrieri and Pietrobelli (2004), among others, and that includes: (1) ‘Marshallian/Italian industrial districts’, in which we find only SMEs that operate in a dense network of relations amongst firms and between firms and institutions; (2) ‘hub-and-spoke clusters’, in which one or a few large firms locally based lead a network of subcontractors; (3) ‘satellite clusters’, in which agglomerations of firms are led by large firms operating outside the clusters; (4) ‘state-anchored clusters’, in which agglomerations of firms are led by state enterprises or institutions. As a basis for Chapter 5, in which the key hypotheses of Part II of this work are identified, in Chapter 4 we investigate the way the main theoretical and policy approaches to cluster development have been implemented in a specific developing country, Nicaragua, and in its SME clusters. In this context, particular importance is given to the ‘social approach’, which is 57
58
SME Cluster Development
analysed in detail and emphasised as an approach that has not been given enough attention in the work that national and international support agencies have developed to promote clusters in this specific country (UNIDO, 1994; INCAE, 1997; García, 1999). More generally, worldwide empirical evidence shows that SMEs are not individually successful, but flourish by working together. They combine forces and exploit agglomeration economies that permit them to compete with larger firms in the global market (Schmitz, 1992; Sylos Labini, 2000). Historically, the most relevant SME clustering experience concerns the late nineteenth century industrial districts of Sheffield and Lancashire (e.g., cutlery) as studied and theorised by Marshall (1916). Similar industrial conglomerations have grown and developed in central northern Italy since the 1970s, and are known as the ‘Third Italy’ (Bagnasco, 1977). In the past two decades, these have been among the most studied cases of dynamic local production systems, and have motivated several theorists to study and develop different analytical and development policy approaches (Brusco, 1982; Piore and Sabel, 1984; Becattini, 1990; Best, 1990; Schmitz, 1992; Pedersen et al., 1994; Humphrey and Schmitz, 1996; Cowling and Sugden, 1999; Bianchi et al., 2000; van Dijk and Sandee, 2002; Pietrobelli and Rabellotti, 2004). In this chapter we investigate the most important theoretical approaches to SME clustering in order to identify the most important variables for better understanding this economic reality and its development process. We also highlight the way these different perspectives can orient the development policy framework. These approaches are not necessarily separate from one another. Several theorists are increasingly studying different aspects of SME cluster development because they recognise the interrelation among these aspects and the importance of understanding them all (Becattini, 1990; 2000a; 2000b; Humphrey and Schmitz, 1998; 2004; Bellandi, 2005). On the basis of the insights coming from our work experience realised from 1996 onwards in Latin America, we have considered to take into account these and other significant contributions as a way to frame a systemic approach to SME cluster development, in particular for its possible effects on development policy in developing countries’ clusters. This constitutes a specific contribution of this volume. This approach to SME cluster development integrates Part I of this work through the analysis of a more specific topic; SME clusters. The macroeconomic and macropolitical trends and obstacles to development (i.e., the fracture in the production system) discussed in Part I do exist also in SME clusters. Nonetheless, in this case these obstacles present less strength. More favourable conditions (to solve the risks and limitations that SMEs
Different Theoretical Approaches
59
face in developing countries, for example in accessing finance and technology) may be linked to the cooperative opportunities of clusters that can help firms become more competitive in the global market. For instance, in clusters there are more opportunities to spur cooperative linkages among firms and among firms and institutions that in contexts where SMEs operate in isolation. This is one of the main policy lines underlined in the previous chapter when discussing ‘linkage policies’. Despite the opportunities offered to SMEs in clusters, their local-specific advantage does not dilute the role of policy (in particular, national policy and programmes) as a fundamental element to promote economic and social development among SMEs. In this chapter, we start by analysing separately the different streams of interpretation and development policy in order to show the importance of each of them (it is in Chapter 5 that there are pulled together through what is called the ‘eclectic approach’ to SME cluster development). The policy scheme developed by each of these approaches is then considered in the specific case of a developing country: Nicaragua. This interest is motivated by the limited success of international and governmental agencies in promoting the development of competitive industrial districts and clusters in this and other developing countries. An explanation could lie in the basic hypothesis of this study, that limited results could be due to the low consideration attributed to relevant types of development factors, such as the governing environment (i.e., local, national and international power structures and institutions) and the social features of the ‘real society that exists behind the market’ (Platteau, 1994). As a consequence, current theoretical, methodological and policy frameworks may need to be modified to take into account the lessons learned from a variety of notparticularly successful clusters. The basic features of three main approaches are highlighted below. The Nicaraguan evidence is then considered, and the three approaches analysed with reference to their effective or potential impact in this country and in its clusters. In the final section, a brief discussion is offered on some prospective consequences for policy-making in Nicaragua and, more in general, in developing countries. However, this final section represents only the beginning of a discussion on SME cluster development policy that will be deepened and clarified in the next two chapters.
2. The ‘spontaneous’ approach to SME cluster development Over the past decades, the mounting difficulties caused to developing countries by the internationalisation of markets have motivated neoliberal
60
SME Cluster Development
think-tanks to criticise the old approach of import substitution industrialisation. This has led the main international organisations to apply free market policies such as the elimination of trade barriers, inflation and exchange rate management, the reduction of state deficits and privatisation programs (Krueger, 1983; Balassa et al., 1986).18 Over the past 20 years, this approach enjoyed widespread recognition within academia and the relevant international agencies. A group of development economists and sociologists builds its theory of cluster development upon the basis of the inevitability of market liberalisation. They recognise the prime role of the market in the development of these networks. For them ‘the answer is unequivocal … effective demand has been the transforming force’, the driver behind the dynamics of growth at the cluster level (Bessant and Kaplinsky, 1995: 129–30; 139; Humphrey, 1995: 1; Schmitz; 1995a: 553–4; Nadvi and Schmitz, 1999: 1503). These development theorists refer to the ‘Third Italy’19 as the main reference point for their analyses and growth models. In this context, development of the European Common Market in the 1970s (i.e., demand side) was the catalyst for the structural upgrading of SME clusters from agglomerations of craft enterprises to the well-known ‘industrial districts’ (Parrilli, 2004b).20 At the same time, these theorists recognise that the capacity of local production systems to respond to market demand was not automatic. The industrial districts were those local systems which were capable of reacting to the new changing scenario of the 1960s and the 1970s. This was a result of the ‘spontaneous’ capacity of small firms to create linkages among themselves to respond flexibly and efficiently to growing market demand (Piore and Sabel, 1984; Best, 1990; Brusco, 1990: 11–13; Lazerson, 1990). A concept is taken to synthesise the endogenous forces of these industrial agglomerations: ‘collective efficiency’ (Schmitz, 1992). This is specified in the ‘joint actions’ and the ‘external economies’ that are put to work in SME clusters so as to ensure competitiveness in the open market. ‘Joint actions’ reflect the strong inter-firm cooperation that takes place within the cluster through, for example, production cooperatives,21 joint marketing abroad through common showrooms, credit and export consortia, etc. These ‘joint actions’ promote an accurate and competitive division and specialisation of labour among firms and, on the whole, the collective strength of the cluster. The second concept reflects the external benefits that firms reap for free from being part of a specialised productive context. These are, for example, the local pool of specialised workforce, the rich flow of information and innovation; and the intense flow of clients to the locality (Marshall, 1916; Ghani and Stewart, 1991; Schmitz, 1992; 1995a).
Different Theoretical Approaches
61
From the mid-1980s onwards, however, spontaneity was no longer sufficient to respond to the new market challenges. Competition, stronger due to market globalisation, forced the districts to re-organise and develop new working methods based on innovation, marketing, information and communication technologies. These changes define a new environment also for the policy support. Effectively, in this new stage ‘industry requires support from local and regional institutions in order to speed up innovation, expand into new markets and thus consolidate growth’ (Brusco, 1990; Schmitz and Musick, 1994: 902). Within this theoretical stream the most fruitful and relevant policy level refers to the support action of local economic institutions and not to that of national or international agencies. In particular, the abovementioned experts agree that local policy should prioritise the delivery of real services to local producers as a way of enhancing the economic efficiency of the local system (Brusco, 1990: 16–17; Capecchi, 1990; Schmitz and Musick, 1994: 903–4; Schmitz, 1995a: 553).22 This represents a support action that respects the market paradigm and, even more, attempts to build up a more developed market at the local level, without attempting to create or develop clusters from scratch.
3. The ‘policy-inducement’ approach to SME development A second important interpretation of SME cluster development derives from political economy, although it takes two lines of argument that are quite different from one another. The first focuses on the ‘policy inducement’ of development and the second on the governance environment. For some theorists (the first line of argument), effective local development requires appropriate national and international institutions, policies and strategies. Otherwise, efforts to expand face a series of bottlenecks linked to the existence of ‘regressive coalitions’ and ‘transnational powers’ that tend to marginalise SMEs and their local business associations (Capecchi, 1990; Cooke, 1996; Bertini, 1998; Bianchi, 1998; Cowling and Sugden, 1999). These theorists show that specific agents in the system, such as the SMEs and their associations, can encounter enormous difficulties in trying to grow. This is due to the path dependency that exists in this context which limits the options available to firms and their systems. In such a framework, the strongest agents use the market to maintain control over resources and decision-making. The capital advantage that they enjoy permits them to invest in more innovative and costly activities (e.g., investments in R&D, promotion campaigns) that contribute to strengthening their edge over weaker competitors (Cowling and Sugden, 1997: 126–7).
62
SME Cluster Development
These considerations explain why policy intervention (i.e., policy inducement) matters at the widest level (i.e., international and national) as a means for changing ‘path’ and allowing new entrants into the market. In this sense, proponents of the ‘policy inducement’ approach affirm that competitive industrial districts succeeded not only on the basis of the spontaneous efforts of SMEs, but also thanks to the robust involvement of policy-makers at the international, national and local levels. The benchmarking experience of the ‘Third Italy’ reinforces this hypothesis. In fact, the first Italian nationwide laws for the support of craft enterprises were introduced in the 1940s, for SMEs as early as 1952 (Bianchi, 1992: 191; Arrighetti and Serravalli: 1997: 358; Brusco and Paba, 1997: 324) and were supported over time through a rich set of national and regional laws for SME development (Capecchi, 1990; Bellini, 1996; Bertini, 1998: 317). The second line of argument recognises the role of governance. This is represented by theorists (many share the spontaneous approach) who studied and emphasised the important relation between global value chains and the process of cluster upgrading. This means emphasising the role of international and transnational powers in the upgrading process of clusters. They have identified the need to distinguish among different types of cluster governance (e.g., hierarchical, quasi-hierarchical, network, market), due to the particular insertion that these governance systems guarantee within global value chains (Gereffi and Korzeniewicz, 1994; Humphrey and Schmitz, 2004; Pietrobelli and Rabellotti, 2004). Upgrading opportunities for local firms and clusters depend strongly on the type of governance. For example, product and process upgrading tend to be easier in the context of ‘quasi-hierarchical’ systems (i.e., clusters led by one or few large firms inside or outside the cluster), while functional upgrading becomes easier in ‘network’ systems formed by SMEs only (Humphrey and Schmitz, 2004). In terms of policy framework, this second group within this (political economy) approach uphold the ‘spontaneous’ approach, by emphasising the importance of setting up stable macroeconomic policies together with local and sectoral policies in the case of ‘network’ governance systems in order to promote the upgrading of the cluster and its capacity to offset better eventual large external partners, without hiding the bigger difficulty in influencing ‘hierarchical’ types of systems (ibid.; Furtado et al., 2005). In contrast, the first stream of experts (i.e., policy inducement) within this approach maintains the importance to discuss the wider framework of production relations and to set up policies at the national and international level (e.g., the European Union, and the World Trade Organization). These are essential to create economies of scale and scope in the promotion
Different Theoretical Approaches
63
of an actor which is a key within the social and economic structure of any country (Bianchi, 1998). These national and international legal frameworks protect the interests of SMEs and their clusters in front of the afore-mentioned regressive coalitions and transnational powers, justifying an easier access of SMEs and their clusters to important means of competitiveness (e.g., financial support, market information, communication technologies). In this sense, building up an efficient set of business associations and real services is important, but needs to be complemented with national and international institutions, rules and policies that favour the development of SMEs and their local production systems (Bianchi, 1998; Sugden and Wilson, 2002). We call this latter sub-stream the ‘policy inducement’ approach to development.
4. The social approach 4.1 The social embeddedness of economic action Certain sociologists, in discussing the social embeddedness of economic action, have proposed important general considerations for the development of SME clusters. Simultaneously, other sociologists and economists, in directly studying the environments of SME clusters, recognise the importance of specific sociological aspects in their success. The first group emphasises the need for a new perspective that attempts to understand ‘the social embedding of economic action’ and the context that lies ‘behind the market, where real societies exist’ (Granovetter, 1985; Platteau, 1994). These wider analyses of development favour an approach that takes into account the ‘social rationality of economic actors’ (Dore, 1983), who act not only to maximise earnings or control markets, but for reasons related to existing social rules and institutions that acquire a strategic importance in specific territories. Their analytical work discusses particular national societies and social features, such as the Japanese ‘goodwill’, that takes the form of ‘obligational contracting’ and ‘welfare corporatism’ (Dore, 1983: 464–6; Sako, 1992), or the Italian ‘city-states’ that take the form of ‘merchant guilds’ (Platteau, 1994: 560). These are examples of national and local environments that have been able to construct strong social linkages and behavioural rules that, in our case, are likely to support the development of SME clusters. In this sense, policy frameworks are proposed to restore ‘moralised trading relationships’, based upon honesty and transparence of objectives, such that ‘these would spur the role of competition policies and job-tenure rights as opportunities for developing a sense of the community in business
64
SME Cluster Development
enterprises’ (Dore, 1983: 480). In a complementary way, these authors recognise the need for mechanisms of ‘co-ordination and enforcement required for thriving trade’, which were organised, for instance, by the merchant guilds in Italy. For this purpose, the state should take the role of leading the system towards a renewed ‘honesty … by organising and strengthening information-pooling systems … enacting rules and designing measurement devices that ensure uniformity of interpretation and narrow down the scope for fraud opportunities’ (Platteau, 1994: 566). This approach captures the importance of emphasising aspects of co-ordination and enforcement in moralised trading relationships as a way of easing systemic economic efficiency (e.g., common codes of conduct within business associations to avoid opportunistic behaviours). Nevertheless, this approach leaves unanswered the question about what actually encourages local economic agents to stay together and work to reach a higher competitiveness as a coordinated unit (i.e., the cluster). This aspect is analysed in the next two sub-sections through a stream of literature that stresses the aspect of trust in industrial districts and the importance of the social push to entrepreneurship (i.e., self-realisation). 4.2 Trust in industrial districts and SME clusters There is a group of sociologists and economists who are specifically interested in the social features that characterise SME clusters. Some of them use this analysis to confirm the hypothesis of the ‘spontaneous’ approach to SME cluster development, which emphasises the importance of the characteristics of specific local production systems (e.g., industrial districts) as the bases that permit these systems to develop competitively (Dei Ottati, 1994; Humphrey and Schmitz, 1998; Nadvi, 1999b; Schmitz, 1999). In general, these theorists highlight the role of the cooperation among local actors that characterises successful SME clusters and districts. Such social factors partly explain what compels economic agents to stick together and work as a coordinated unit. These theorists study factors that facilitate economic transactions and constitute key aspects of the Marshallian industrial atmosphere. In this sense, they underline the concept of trust as a facilitator of economic transactions. Trust allows firms to reduce their operating costs and creates what is known as the cluster’s ‘collective efficiency’ (Becattini, 1990; Lorenz, 1992; Dei Ottati, 1994; Nadvi, 1999b; Schmitz, 1999; Bellandi, 2005; De Propris, 2005). These scholars highlight the role of the local social environment in the creation of an ‘ascribed’ type of trust, which turns out to be a key aspect of the spontaneous growth of the cluster in its first phase of development.
Different Theoretical Approaches
65
The shift to an ‘earned-type’ of trust, however, provides the foundation for a more advanced stage of industrial development, in which the district finds itself amidst more competitive global markets. In this phase, a new type of trust arises to gain competitive advantage and to overcome ‘the costs of non-cooperating’ (Schmitz, 1995b; 1999). This new type of trust seems to be based upon a sort of ‘cost/benefit analysis’. The first type of trust (i.e., ascribed trust) tends to blur with what is also called ‘systemic trust’ (De Propris, 2005). This view links the creation of trust to the social environment of small town clusters, in which people create their own norms of behaviour and interaction (Taylor, 1982, in Lorenz, 1992: 196). In particular, ‘it is based upon feelings of empathy, reciprocity and morality’ (De Propris, 2005), and its origins are rooted in linkages of friendship and kinship at the individual level, and common norms and culture at the collective level (Noteboom, 2002). In this first conception of trust, although punishment for free-riders is not necessarily established through contract penalties, it is recognised in this system by means of a socialised formation of professional and entrepreneurial CVs. The second conceptualisation of trust (i.e., ‘earned trust’) can also be called ‘calculative trust’ (De Propris, 2005). This form of trust is tied to the economic interactions only and, more specifically, to the repetitive positive exchanges that help forecast the result of future interactions and, in this way, reduce the risk of opportunistic behaviour. This kind of trust is justified through self-interest at the individual level and the existence of an authority having the power to impose certain behaviours at the collective level (Noteboom, 2002). In this second conception of trust, punishment for opportunistic behaviour is reduced in a significant way through the extended use of contractual mechanisms and penalties (Axelrod, 1984; Lorenz, 1992: 198). The debate is still open on whether ‘calculative trust’ supports the implementation of concrete cooperation initiatives within the local SME system better than ‘systemic trust’ or vice versa. Possibly both of them promote trust (see 4.3 below); although the impression is that systemic trust presents the higher potential for the construction of a purposeful cooperation. Some evidence shows that assistance from state and/or local level institutions (i.e., systemic type of trust) can help to produce such a co-operation (Tendler and Amorim, 1996).23 The systemic type of trust correlates with the sociological school of industrial districts, whose theorists recognise the importance of the ‘Marshallian industrial district as a socioeconomic notion’ (Becattini, 1990; Bagnasco, 1988; 1999; 2000a; Trigilia, 1992). They stress the importance
66
SME Cluster Development
of ‘the spread of family-based agricultural smallholdings, helping to create the original supply of a cheap and flexible workforce, whose skills and motivations were well-suited to the development of small businesses’. They also indicate the relevance of local political traditions and institutions, such as the Catholic Church and the Socialist and Communist parties in the early twentieth century (Capecchi, 1990; Trigilia, 1992: 36–7; Bagnasco, 1999: 93–4). Relevant studies emphasise the significance of these forms of social capital as a basis for economic democracy and competitiveness (Putnam, 1993; Locke, 1995; Cowling and Sugden, 1999). If policy-making does not take trust into account, social capital can be destroyed (Bagnasco, 1999: 75). These theorists reaffirm the need to construe the social and institutional environment in local production systems as part of a competitive industrial strategy. In this sense, this stream of the literature mixes with the point of view of ‘policy inducement’ theorists on institutional building as a way to upgrade the conditions of clusters and localities (Sugden and Wilson, 2002). ‘It is necessary to stimulate the creation of strong interlocutors within civil society … The political construction of them is an essential part of the development strategy because existing organisations might eventually internalise the advantages deriving from the production of collective goods and might, as a result, develop a longer-term view of individual interests, inducing them to cooperate more effectively … Unless a more autonomous and better-organised civil society is created and unless a more autonomous and technically trained bureaucracy emerges, any new policy is liable to prove inefficient and to stimulate the growth of political entrepreneurship rather than economic development’ (Trigilia, 1992: 46–7). In spite of these peculiar considerations, policy making tends to follow a well trodden path. Within this stream of the ‘social’ approach, regional development is enabled through policies aimed at encouraging the creation of collective goods that lead to the spread of technological knowhow, trade promotion, business skills training, worker training, and so on (Trigilia, 1992). Interpreting these words, however, economic development is less a result of social forces applied to market, and more the result of economic mechanisms and agents coordinated in an efficient way. 4.3 The social strengths of SME clusters The analysis now turns to the ‘black box’ of society. This is important since the analysis of economic determinants alone cannot explain all the effective elements that promote dynamism in SME clusters.
Different Theoretical Approaches
67
For example, let us discuss the issue of what promotes cluster dynamism. Under certain approaches, this aspect could be analysed through the economic concept of ‘entrepreneurship’ (Schumpeter, 1934; Kantis, 2004; Aranguren et al., 2005). In our view, however, this is insufficient because we need to know where strong entrepreneurship comes from. This does not only depend on economic variables (e.g., maximisation of incomes, financial incentives or market opportunities), but also and mostly on social institutions and rules (Brusco, 1982; Bagnasco, 1988; Becattini, 2000b). In dynamic SME clusters (e.g., Italian industrial districts) this is shown by people’s desire to set up their own businesses, to be their own boss, to decide their own timetable, products, clients, suppliers, workers and to be in a position to succeed on the basis of their own skills and capacities (Hjalager, 1993). These aspects cannot be understood through an analysis of the economic incentives to entrepreneurship. They represent an individual’s move to ‘self-realisation’ that takes place in SME clusters. In other societal contexts (e.g., big cities), people possibly dream of careers in large companies and institutions. In SME clusters, usually within rather small towns, the status of being an entrepreneur is very much respected and considered a desirable position (Bagnasco, 1988; Hjalager, 1993; Becattini, 2000b). These individual choices, although often contradictory to economic objectives, are nevertheless important factors in decisions to create small firms. That is why we need to investigate the social characteristics of successful societies. What social features can help SME clusters to develop? In our view, two main interdependent forces are at work: social cohesion24 and self-realisation. With reference to the first for example, there is a close link to the concept of trust. The above-mentioned literature (4.2) emphasises the concept of trust as ‘oil’ that facilitates transactions. Research by Schmitz concerning the passage from ‘ascribed’ to ‘earned trust’ indicates that trust can change, depending on particular social and economic conditions (Schmitz, 1995b; 1999). In his study, ‘ascribed trust’ comes from ethnic ties, while ‘earned trust’ stems from economic advantages. Although their origins are very different, both represent trust. Trust is a sort of ‘homogeneity’ in the behaviours of local people that helps the working of the economy, because it makes the people understand each other easily and thus leads to lower transaction costs. It does not seem to matter upon what trust is based, as long as all local agents know each other and share common social and economic values (Dasgupta, 2002: 35). In this sense, a wider concept than trust needs to be discussed, that of social cohesion. This represents the force that helps a cluster to develop as a unique, collective agent, avoiding/overcoming internal fractures and dualism. Trust is rather an effect. It is like a beautiful dress, but just a dress,
68
SME Cluster Development
which needs ‘content’ (a body) in order to be appreciated. This content is ensured by the individual and collective values and objectives shared within the local community. It refers to the ‘commons’, the ‘values’ or even the ‘priorities’ of local societies at a certain point in time that define the way the people interact with one another (Lorenz, 1992; Dasgupta, 2002). Cooperation, conviviality, family work and personal and collective reputation represent these values within a trustworthy society. If these are not agreed upon within the local culture, it becomes hard to produce common development plans and outcomes. In effect, the cooperative initiatives that permit small firms to be competitive in the open market seem to require more than (calculative) trust. Active participation in credit and export consortia, production cooperatives, trade fairs, showrooms, etc., need trust, but of a different kind in respect of ‘calculative trust’. Indeed, this helps individuals to reduce their transaction costs in bilateral exchanges, but does not seem enough to push them to participate actively in joint initiatives with many other producers and economic agents over long periods of time, unless they can reap short term economic returns.25 To have a kind of cooperation that surpasses the ‘calculative trust’, people need to agree at least implicitly on the abovementioned commons, values and priorities of development at the cluster level. These values point to the systemic concept of ‘social cohesion’ (Dasgupta, 2002), which resembles the concept of systemic trust pointed out by De Propris (2005). ‘Social cohesion’ can be an element of a static environment in which people get on well with each other, as more likely occurs in traditional surroundings, where people had known each other well for long (e.g., small country villages). In SME clusters, the widespread adoption of entrepreneurial habits (that represent the local tendency to ‘self-realisation’) represents a key for collective success (Brusco, 1982; Becattini, 2000b). This pushes the locality and its agents towards dynamic and competitive practices (e.g., innovation, exporting, and credit consortia). For these reasons ‘social cohesion’ is not enough to explain the dynamism of SME clusters; thus, ‘self-realisation’ needs to be involved (i.e., interdependence between the two factors). Thus, the discovery of the social features that motivate dynamic entrepreneurship becomes essential for development policy purposes. The concept of ‘self-realisation’ could be used to perform this needed function. In fact, many people in SME clusters prefer running their own business rather than participating as an employee in someone else’s. They aim at maintaining an independent business (Brusco, 1982; Hjalager, 1993; Becattini, 2000b). It is a matter of choice, however, since in other contexts many people prefer alternative ways to succeed.26
Different Theoretical Approaches
69
Several studies on the history of very successful experiences in the ‘Third Italy’ highlight mechanisms that helped to create a strong social fabric and the corresponding economic concept (i.e., entrepreneurship). For example, the long standing practice of sharecropping and the municipal history of arts and corporations helped peasants and urban dwellers to familiarise themselves in managing economic activities directly and responsibly (Brusco, 1982; Putnam, 1993; Platteau, 1994). The reading of the Italian, but also of Latin American and East Asian contexts (Kantis et al., 2002; Kantis, 2004) nurtures a complementary hypothesis which points to society being responsible for the approval and implementation of policies, institutions and rules. Understanding the essential social features (including more generally the historical and institutional background) at a given moment in time therefore becomes crucial to determining the potential of SME clusters, and the mechanisms and institutions likely to promote their development process. In conclusion, ‘social cohesion’ is essential because it permits the system to grow as a unique, integrated agent. ‘Self-realisation’ is vital to a dynamic local environment moving towards higher levels of individual competitiveness and collective satisfaction. Having one without the other would generate a sub optimal outcome, because society could end up being static (in the absence of ‘self-realisation’) or result in an aggregate ‘zero-sum game’ in terms of local development (without ‘social cohesion’). Targeting a set of social policies to support these interdependent characteristics becomes a strategic objective for motivating a proactive social environment to actively respond to and benefit from economic policies. Through the case of Nicaragua, the next section examines how the different approaches to SME cluster development work in reality and considers the need to integrate them in a framework for interpretation and effective policy-making (which, from the next chapter, is called ‘eclectic approach’).
5. Empirical evidence from Nicaragua 5.1 An introduction to the country Nicaragua is a developing country that has similar problems to many other Latin American countries. GDP, per capita income, exports, trade balances and foreign debt offer little stability, as does the governance environment. Notwithstanding this, a set of positive elements is today visible. For example, a recent wave of public investments in basic infrastructure has eased the transportation of production from the countryside to towns and export markets. A growth of foreign direct investment in traditional
70
SME Cluster Development
manufacturing and agro-industrial products has also occurred (BCN, 2000). These sector specific factors are complemented by the rather successful management of the main macroeconomic indices such as inflation, the exchange rate and the budget deficit. The economy grew by an average of 5 percent per annum in the 1990s, during which time the government successfully applied a controlled exchange rate devaluation to hold down inflation. Devaluations of 12 percent per annum (10 percent since the beginning of 1999), allowed a significant reduction of inflation to less than 10 percent per year (ibid.). Other fundamental macroeconomic measures have also been successfully applied, among which, trade liberalisation has turned Nicaragua into the most open Central American economy. Other measures have included the privatisation of state enterprises (including public utilities), improvements to basic infrastructure, control of public expenditure and public deficit reduction. Complementarily to these efforts the country has been able to generate continuous growth over the past 10–15 years (ibid.). Despite these positive indicators, income distribution remains a serious problem, since a large part of the population lives below the subsistence level, often earning a living through informal activities such as petty trading and manufacturing. The growing informal sector is another indicator of the failure of macroeconomic policy to address the problems of competitiveness of SME clusters. Non-existent under the former regime, the informal sector is currently estimated to employ around 50 percent of the economically active population (Speer, 1997; Agurto and Guido, 2002). From the point of view of economic performance, SME (and SME clusters) do not seem to have significantly improved. Low competitiveness is suggested by the increasing gap between imports and exports in relevant production sectors.27 On the whole, SME clusters (i.e., the supply side) cannot yet respond adequately to national and international consumers (i.e., the demand side), who therefore prefer imported goods. This situation leads to ever increasing trade deficits (BCN, 2000; Narváez and Parrilli, 2000). As described in Chapters 2 and 3, the economy is divided between a small part able to adhere to international production and consumption standards, and a large majority of the population living below or slightly above the subsistence level and which is anchored to the traditional local economy or to the informal sector (Bianchi and Parrilli, 2002b). In Nicaragua, the 158,000 existing SMEs are representative of this large sector of the national population that suffers from a lack of higher income opportunities (Banco Central de Nicaragua-GTZ-MEDE, 1998). In this context, several small towns form classic configurations of SME clusters (Parrilli, 1999a; 1999b). This is presented in Table 4.1.
Different Theoretical Approaches Table 4.1:
71
Identified clusters in Nicaragua
Sector
Existing clusters
Number of firms
Value added (VA) or exports (E)
Coffee
Matagalpa, Carazo, Nueva Segovia
US$ 120 million (E)
Dairy products
Boaco, Chontales, Rivas, Rio Blanco, Managua
Furniture
Shoes
Masaya, Masatepe, Estelì, Chinandega, León, Ocotal Masaya
60 SMEs ⫹ thousands of farmers 20 craft firms (only 3–4 industrial) ⫹ thousands of cattle-farmers 600 craft firms
US$ 5 million (VA)
Leather
León, Granada
Pottery
Salmon
San Juan de Oriente, San Juan de Limay, Mozonte Puerto Morazan
500 craft types of firms 3 medium ⫹ 50 craft firms 60 craft firms
Bricks Hammocks
Nagarote Masaya
30 large firms and SMEs 30 craft firms 30 craft firms
US$ 25 million (E)
US$ 3 million (VA)
million 2.5 dollars (VA) US$ 0.5 million (VA)
US$ 10 million (E) US$ 0.5 million (VA) US$ 0.5 million (E)
Source: BCN-MEDE-INEC, 1996 and BCN-GTZ-MEDE, 1998; for exports and value added: BCN, 2003.
These data show the empirical bases that could be utilised in view of spurring the development of clusters through appropriate cluster policies and actions that can be set up in the context of Nicaragua. Conscious of this situation, international organisations and cooperation agencies have been setting up support programs for SMEs and clusters alongside the government, national business associations, NGOs and universities.28 5.2 The main programmes to support SMEs in Nicaragua After decades of import substitution policies, aiming to create and support national sectoral ‘champions’ (Laenen, 1988; Reyes, 1993), it was only during the late 1980s and early 1990s that successive Nicaraguan governments started to implement policies to support SMEs, at both the national and local level (Bouchier, 1998). Nineteen-ninety, for example, saw the creation of the Programme for Supporting Micro Enterprises (PAMIC), which evolved in 1998 into the
72
SME Cluster Development
National Institute for Supporting the Small and Medium Enterprises (INPYME). This institution, supported by various international cooperation agencies (e.g., from Spain (AECI), Germany (DED) and others from Austria, Norway, Holland, etc.) has mainly promoted the work of micro enterprises through micro-credit (indirectly, through non-conventional micro-finance institutions) and other initiatives. Among these, the main activity is ‘Microfer’, a trade fair organised twice a year in Managua in order to promote micro enterprise sales. Moreover, INPYME is supporting the participation of micro enterprises in trade fairs abroad, especially in Central America, to offer micro enterprises more opportunities to know the international market and to commercialise their production (INPYME, 2000). Another relevant project, for SMEs involved in furniture production, was set up by the governments of Finland, Chile and Nicaragua (in the project Nicamueble). Through this project they attempted to train producers and promote their production standards. For example, the project supported a group of artisans (35 small firms selected at the country level without a cluster approach) through a set of activities linked mainly to training and technical assistance. Moreover, the project helped the group to set up a shop in the capital to sell their production to high-income consumers (INPYME, 2000; Narváez and Parrilli, 2000). The financial environment is often indicated by SME entrepreneurs as the weak link in the policy support for SMEs in Latin America and elsewhere (García, 1999; MIFIC, 2001b). In 1999 the Nicaraguan Ministry of Development, Industry and Commerce (MIFIC) was allocated a fund of US$ 34 million from the Taiwanese government as an instrument to upgrade the technological standard of SMEs (FNI, 1999). Another relevant fund, US$ 10 million, was allocated in the same years from the Central American Integration Bank (BCIE) to Nicaragua (BCIE, 1999). Another important financial support (in the form of a subsidy) also arose at the beginning of the 2000s. This is based upon a project of the Inter-American Development Bank jointly with the German development agency GTZ and the business association Nicaraguan Institute for Entrepreneurial Development (INDE). It refers to the Programme of Development Services (PROSEDE) that is targeted to support supply and demand of business development services through vouchers paid by these agencies and to which the enterprises are required to contribute in certain proportions (BID, 1999; MIFIC, 2001). Further recent policy guidelines need to be taken into account. In recent years, with the newly elected government in 2002, an overall policy to develop competitive clusters has been approved and is now in its phase of implementation (INCAE, 1997; MIFIC, 2001). In reality, this
Different Theoretical Approaches
73
conceptualisation of clusters is quite different from the standard one, which is used at the international level to mean ‘geographical and sectoral concentrations of mainly small and medium firms’ (Schmitz, 1992). In contrast, this idea is based upon a wider geographical framework, which can include vast areas of territory (also territories of 10,000 square kilometres). In some way, it depends on the specific interpretation that Central American academic and policy institutions made of the concept of cluster, which tends to be identified with the ‘diamond of competitiveness’ elaborated by Porter (1990) and which tends to blur with the concept of production chain). This policy identified four main types of products: textiles, tourism, fishery and agro industry, namely coffee, dairy products and non-traditional products (INCAE, 1997). But the choice of these sectors seems to be problematic. For example, the textile sector includes mainly transnational companies from Taiwan, Korea and the United States working in free exporting zones. Agro industry is mainly focused on the promotion of coffee and dairy products (De Franco et al., 1996; Parrilli, 1998a; Mendoza, 2002; Artola and Parrilli, 2006). These sectors are crowded by thousands of small farmers and rather tiny processing sectors, mainly controlled by large and medium enterprises, which can also be controlled by transnational capital (e.g., Parmalat and Nestlé in dairy production; Nestlé in coffee production). Tourism is another non-manufacturing sector that is mainly controlled by large national and transnational firms. This overall perspective makes clear the difficulty of strengthening the position of SME clusters within the national production system. Finally, special attention needs to be devoted to the important work realised by the United Nation Industrial Development Organization (UNIDO). This organisation has been working for SME and cluster promotion from 1993 in Nicaragua and supporting the governmental organisation that promotes small and medium enterprises (PAMIC/INPYME). In particular, UNIDO set up programmes led by network brokers who try to encourage small producers to work together to find joint solutions to common competitive problems (Humphrey and Schmitz, 1995; Ceglie and Dini, 1999). In the first few years (1993–98), UNIDO project was oriented to spur a general type of cooperation among small producers in some selected clusters (e.g., Masaya in textiles and hammocks, León in leather production, Masatepe in furniture, Granada in leather goods, Boaco in dairy products), in order to promote their cooperation to participate in trade fairs and training programmes. This type of programme led to the creation of relatively small groups of firms (between 15 and 30 craft firms) mainly
74
SME Cluster Development
oriented to improve their access to wider markets (Ceglie and Dini, 1999). Interesting cases are, for example, the marketing efforts of a group of producers in Granada to sell their leather craftworks in the United States; and also the effort to improve the materials (e.g., natural varnishes) and access international markets (e.g., Canada and Europe) in the hammock cluster of Masaya (Parrilli, 2003). From the end of the 1990s, UNIDO complemented the programmes to support horizontal networks with programmes to support vertical networks. This latter case was oriented to promote the creation of production linkages (e.g., subcontracting chain) between SMEs and large firms. This has been made especially in the shoe and dairy industry with alternate success (Artola and Parrilli, 2006). Simultaneously, UNIDO focused its attention on the institutional system (e.g., business associations, governmental bodies, universities, municipalities, etc.), trying to boost its participation in the promotion of economic development. In the case of dairy production, this effort led, among other activities, to the organisation of the first meeting of the producers of dairy products in 2001 (ibid.). After years of support, the economic performance has improved significantly for only a reduced number of firms, while the rest have not undergone a substantial upgrading process (see Table 4.2). At the cluster level, success is even less evident. Most small firms remain subsistence workshops or slightly better. They continue to work by themselves, doing everything internally. They often completely omit certain phases of production or shoddily attempt them internally, rather than subcontracting to or partnering with more specialised SMEs in the cluster. For example, in furniture production, many producers tend to omit an appropriate drying of the wood or the necessary wood lathing; in dairy production they omit pasteurising, etc. (De Franco et al., 1996; Barahona et al., 1998; 1999; Parrilli, 2002a; 2003; Artola and Parrilli, 2006). 5.3 The limits to the division and specialisation of labour and their social roots The above-mentioned kinds of firms (craft shops) tend to disregard the benefits of inter-firm cooperation that the institutions have tried to stimulate. This posture is due to a few main reasons. One of these is linked to the negative attitudes that entrepreneurs hold towards SME clustering. They fear the close presence of other firms due to the risk of losing clients and to the risk of imitation from competitors owing to the absence of legal protection (see also section 4.3 in Chapter 6). This fear is also linked to the market where producers sell their goods. Since this is often the poor local market, they are forced to manufacture cheap products. The only
Different Theoretical Approaches Table 4.2:
75
The strongest sectors of industrial SMEs in Nicaragua, 1996
Sectors
Bakery products Clothing Leather and shoes Chemicals Non-metallic mineral productions Metallic products excluded Machinery Wood and furniture Total
Number of SMEs
Number of workers
Value added (US$ 000s)
463 66 133 34 217
2,383 891 1497 357 1435
10,529 3,562 4,387 4,594 4,373
101
561
6,643
502 1,805
2,866 12,159
6,115 51,170
Note: All data have been converted with the average exchange rate of 1995: C$ 6.72 ⫽ US$ 1. Source: Database of the Central Bank of Nicaragua, 1998, in Parrilli (2002b).
way to cope with this condition is by holding down production costs, buying cheap materials, hiring cheap labour, saving on production techniques and technology and dedicating as little time as possible to producing each final good (Pyke and Sengenberger, 1990; Kaplinsky and Readman, 2001). Most of the national producers, especially those localised in SME clusters, have not experienced the benefits of division and specialisation of labour and, hence, tend to be rather suspicious about this possibility. They want to rely as little as possible on other producers in order to control all possible risks (e.g., bad quality subcontracting, low quality of inputs, late delivery of intermediate goods, etc.) and therefore control directly their productive activity (Narváez and Parrilli, 2000; Parrilli, 2002a). These difficulties also relate to the social environment of the country. That is why this further aspect of SME cluster development needs to be analysed. In this sense, we utilise the theoretical framework developed in section 4.3, and attribute special relevance to the aspects of ‘social cohesion’ and ‘self-realisation’. The recent history of military and civil conflict within the country, that started in the early 1970s (revolution against a dictatorship) and ended up in the early 1990s (civil conflict), left an overall atmosphere of distrust among people in the localities/clusters (Mendoza, 1990). This atmosphere also affected the cooperative system that was created and developed by the Socialist regime in the 1980s. The sudden and dramatic market liberalisation that took place in the early 1990s led many cooperatives to close and caused much conflict regarding the assignation of capital between members, a situation which led most of the people/ artisans to deception and distrust with this type of productive and market
76
SME Cluster Development
organisation. Nowadays, the setting up of cooperatives is not well-received in the urban context,29 which is a further indication of the country’s low level of ‘social cohesion’ (Mendoza, 1990; Parrilli, 1998b; 1999a; 1999b; Narváez, 2001). At the same time, however, market liberalisation altered the social and economic prospects of the most dynamic and richly endowed individuals. It inspired them to develop their own initiative, skills and capacities, which had remained dormant under the centralised system. There are several interesting individual cases of producers on a dynamic entrepreneurial growth path in Nicaragua. They are relatively young people, determined to learn and apply the best international practices to their own activity, by investing in technology, buying appropriate materials, hiring and training good workers, and searching for customers outside the local market (MAGFOR, 1999; Parrilli, 1999b).30 However, the concepts of ‘entrepreneurship’ and ‘small and medium enterprise’ are not yet sufficiently respected. Being a small entrepreneur is still associated with ‘living in subsistence’ (Parrilli, 1999b). Many young people therefore prefer to seek career opportunities elsewhere within large corporations and public institutions (e.g., finance, banking, and international cooperation projects) or even within international NGOs (Rocha, 1999). In Nicaragua, ‘self-realisation’ through entrepreneurship seems more prevalent than ‘social cohesion’ through cooperation. Such individual dynamism will not go far, however, unless incorporated within a medium to long-term project agreed among all the local economic agents. The economic system would otherwise tend to produce a ‘zero sum game’, with no positive local effects. In fact, within a system that manages scarce resources (i.e., developing countries), in which producers compete for the same local consumers, the success of a few enterprises tends to be complemented with the failures of many others. A few producers would be better off, while the large majority remain in subsistence conditions (Parrilli, 1998b; 2001). On the whole, few ‘social’ policies and initiatives have been implemented in Nicaragua. Aspects of ‘social cohesion’ and ‘self-realisation’ have been left to the spontaneous mood of producers and institutions, a situation which offers local economies little potential to develop. 5.4 The conflictive political economy in Nicaragua With respect to the issue of governance systems and power relations, in Nicaragua operate different interest groups whose conflict has important impacts on SMEs and cluster development. A few examples can suggest
Different Theoretical Approaches
77
these considerations. At the end of the 1990s, the German development agency GTZ supported a group of business associations (INDE, UNIPYME, etc.) to design a law project for SME promotion that was passed on to the Congress for approval. This law project got stopped at the Congress and was never approved (Bouchier, 1998; Parrilli, 2003). This failure suggests that some groups within the Congress do not consider this law a priority; otherwise they would have simply requested adequate changes and improvements in the project before approving it. Still at the end of the 1990s, the importers and traders of international brands had been fighting for lower tariff and non-tariff barriers to trade. Partly for this reason, Nicaragua has incurred the fastest trade liberalisation in Central America, which has led to ‘tariff perversion’, as the former Ministry of Industry Norman Caldera called it, allowing the entry of imported intermediate inputs with higher tariffs than for the related final goods. This has discouraged national production and led domestic players to import final goods to the national market and therefore become traders themselves (Solórzano, 2000).31 Also this case suggests that some interest group (i.e., importers of international brands) have different and conflicting purposes in respect of those of small and medium manufacturing enterprises. A third example of regressive coalitions at work involves financial policy, which affect SME access to credit at acceptable interest rates. The oligopoly maintained by the national banks prevents the growing sector of nonconventional financial institutions (NGOs) from collecting people’s savings (Rocha, 1998). Opening this market would make the banking sector more competitive and offer lower interest rates to SMEs, which nowadays are still between 20 percent and 30 percent for manufacturing production, reducing sharply the return on their technological investments (BCN, 2002). Effectively, only 5 percent of the available funds had been delivered to SMEs two years after starting the afore-mentioned BCIE financial program (see section 5.2). Also, this third example suggests that other interest groups operate with different and conflicting objectives (versus SMEs) and usually tend to succeed. On the whole, these three examples clarify the importance of analysing a country’s governance system, in this case revealing the conflicts of interest between economic and political lobbies and indicate the relative weakness of the SMEs, their associations and clusters in front of these other lobbies in the country. In this context, the policy-inducement approach offers insights into understanding the real dynamics behind the development of SMEs and their clusters. In particular, this approach leads to assess the difficult
78
SME Cluster Development
governance environment that SMEs are facing in Nicaragua nowadays. Other more powerful lobbies plea for different kinds of policies and succeed. Few legal incentives and mechanisms have been put in action in order to legitimise SMEs in the country and promote their growth process; even less for SME clusters in specific localities.
6. Preliminary conclusions on policy-making in SME clusters This chapter has discussed the relevance of some important approaches to SME cluster development in both the interpretation and encouragement of the development process. The approaches individually highlighted herein are proposed as necessary aspects of an effective integrated development policy scheme (this proposal is made more explicit through the formulation of an ‘eclectic approach’ in Chapter 5). The ‘spontaneous’ approach proves to be of strategic importance at the local level where action and development in fact take place. It focuses on the responsibility of entrepreneurs and their associations to set up strategic joint initiatives to reap the benefits from collective economies of scale and scope, and from external economies. The policy-inducement approach is strategic too, since SME clusters need adequate legal and institutional frameworks and tools to overcome the more or less open opposition of competing economic interests. The social embeddedness approach is equally fundamental because it allows us to understand the entrepreneurs on the ground, and their inner motivations and incentives for economic action. This makes it easier to involve them in the development processes and projects in their own local production system. Not applying these three approaches together is an action equivalent to an engineer omitting one of the supporting columns of a large building. Although the building would stand, it would be somewhat shaky and unable to support subsequent extensions. Likewise, SME clusters might always grow with limited support, but will do so at a much slower pace than if they benefit from integrated policy support. It is important therefore to bring together all the relevant aspects in a framework of industrial development policy. This consideration is consistent with what is argued in Part I of this volume, where a proactive and systemic approach to the development of national production systems is advocated as a way to overcome the wider problems that SMEs face in the open market (i.e., the fracture within the production system due to unequal access to finance, technology and markets). In linking this discussion to the more specific context of SME clusters we may argue that,
Different Theoretical Approaches
79
if policy does not involve strategies and actions to strengthen the social environment, local people (particularly in less developed countries) will find more difficult to benefit from the economic policies in action. The case of Nicaragua seems consistent with this hypothesis. Different sets of policies and programs have had limited success over the past 10–15 years, but remain unaccompanied by actions targeted to prepare and motivate society for their benefits. As a consequence, small and medium industries and SME clusters advance more slowly than necessary (Parrilli, 1999a; 2001). Table 4.2, making special reference to the value added generated by SMEs, suggests that these sectors, often organised around clusters (e.g., wood and furniture, leather and shoes, clothing), are not yet able to achieve significant standards of development. A policy-making approach involving a social aspect is more likely to produce effective results in the development of SME clusters. This is because such an approach focuses on strengthening the relevant social features (i.e., ‘social cohesion’ and individual ‘self-realisation’), which increase the motivation of the local society to actively endorse the economic policies at work (Parrilli, 1999a; 1999b; 2001). For example, one of the most common programs in Nicaragua is the ‘networking programme’, which has been encouraged by important organisations (e.g., UNIDO, INPYME). SMEs are pushed to work together through the activity of ‘network brokers’ who provide specific support to help them produce, export, participate in trade fairs and so on (Ceglie and Dini, 1999). No social policy is present, however, to make these small producers more sensitive to the many advantages of working together. Thus, they remain suspicious about imitation risks and the predatory attitudes of other producers in the cluster. Within this context, a comprehensive development framework could involve parallel social and networking programmes (Cooke and Wills, 1999). This could mean creating forums, especially at the local level, in which several agents are invited from local governments to participate in the setting up of a medium- to long-term development plan. In this way, the producers and their associations, banks, local authorities and support agencies, retailers and service providers are more likely to get to know each other better, and start creating a more cohesive and dynamic ‘industrial atmosphere’ within their SME clusters (Aranguren et al., 2005). Another application of a more comprehensive framework could imitate the recent trend in development programmes of setting up business incubator programmes in developing countries and clusters (Schmitz and Musick, 1994; Kantis et al., 2004). These programmes have also been discussed and applied in Nicaragua, but with contentious outcomes. In the
80
SME Cluster Development
tanning industry in León, the country’s second largest town, groups of producers have been pushed to join an initiative supported by the Austrian cooperation agency and UNIDO. It was de-emphasized two years later, however, and local producers returned to work in their obsolete workshops (Parrilli, 2001). A typical interpretation by producers was that a lack of coordination caused such an outcome (ibid.). This response seems consistent with comments about the absence of local ‘social cohesion’ and a lack of dialogue between the designers of the project and the local people. The hypothesis offered is that people accept the development program as a gift, from the outside, rather than as the final outcome of a participatory decisionmaking process. This situation occurs because policy-making tends to be a top-down process in which SMEs are not involved (Sugden and Wilson, 2002). On the whole, the large majority of the population does not seem strong enough (i.e., socially cohesive and individually dynamic) for a place at the decision-making table. On this basis, a comprehensive scheme for policy could involve a social programme that accompanies the business incubator programme at its outset. This strategy would help to prepare and motivate people for the economic programme to work effectively. For example, social policy could involve not just some of the ideas mentioned above, but also a campaign inside and outside the formal education system in León. This might guarantee more respect for entrepreneurial action and attitudes enabled by the organisation of important local and regional events in which a few successful entrepreneurs and enterprises are rewarded with a local and/or national prize, and are asked to present their success stories in different schools, universities and institutions.32 This strategy is likely to increase the number of young people; both students and workers, involved in private initiatives and to consequently drive the dynamism of SME clusters in developing countries. On the whole, these are examples that indicate some of the missing links that can be identified in the growth process of SME clusters and that can help to set up effective development policies in the near future. An approach to SME cluster development that integrates the relevant and interdependent economic (i.e., ‘joint actions’ and ‘external economies’ or ‘collective efficiency), governance (i.e., national and local policies or ‘policy inducement’) and social (i.e., ‘social cohesion’ and ‘self-realisation’ or ‘social embeddedness’) targets is more likely to meet the objectives of both the development organisations and the local people within a reasonable timeframe.
Different Theoretical Approaches
81
The next chapter identifies the main hypotheses of Part II of this volume in a synthetic and systemic way and applies them to a specific context, that of Italian industrial districts. In this way, we present the possibility to apply the identified theoretical framework (‘stage and eclectic approach’) also to the relevant context for this work, which is that of (a specific set of) ‘survival clusters’ in developing countries. The latter stage is explored in Chapter 6.
5 A Stage and Eclectic Approach to Industrial District Development: Two Policy Keys for Survival Clusters in Developing Countries
1. Introduction33 Through this chapter we intend to continue the analysis on clustering begun in Chapter 4 and to identify more clearly the main hypotheses of this Part II of this volume, which refer to a specific type of cluster that is localised in developing countries: the ‘survival cluster’. This represents the less dynamic kind of cluster, composed mainly of craft producers oriented to local markets and, therefore, achieving a very low rate of capitalisation and growth. Nevertheless, this kind of cluster is very common in developing countries (Altenburg and Meyer-Stamer, 1999; Knorringa, 2002) and, hence, deserves special attention. The objective of this chapter is to identify the development policy that, based upon specific conditions, can promote a sustained growth, at least in one part of these less dynamic clusters in developing countries (the most embryonic clusters). Its purpose is to avoid the mistakes policy-makers often make in developing countries, which is to encourage too fast a development of local SME clusters (often ‘survival clusters’) without passing through viable stages. The creation of too high expectations and unfeasible models are likely to lead these projects to failure. For this purpose, we analyse the development of the Italian industrial districts, through whose historic evolution (two) relevant lessons are identified for the development of the potentially most dynamic sector within survival clusters. On the whole, the study of the Third Italy experience ends up proposing a ‘stage and eclectic’ theoretical approach that integrates the key characteristics of the main bodies of literature on this subject. The resulting analytical structure delivers two main hypotheses: the first recognises the importance of upgrading through gradual and feasible stages in the growth process of small and medium enterprise 82
A Stage and Eclectic Approach
83
clusters; the second identifies the complementarities among economic, social and policy factors (more specifically, those that refer to ‘collective efficiency’, ‘social embeddedness’ and ‘policy inducement’, as they are described in Chapter 4) that systemically contribute to the growth of these production systems. This analysis benefits from Brusco’s discussion of different models of SME clustering (Brusco, 1990), through which he identifies four historic models that synthesise the debate on industrial districts in Italy from the 1950s onwards: the simple agglomeration of craft enterprises between the 1920s and the 1950s; large firm-led industrial agglomerations in the 1950s and the early 1960s; traditional industrial districts from the late 1960s to the 1980s; and new competitive IDs from the late 1980s onwards. Although other significant industrial districts typologies have been identified more recently (Asheim, 1994; Markusen, 1996; Guerrieri and Pietrobelli, 2004), this paper refers principally to Brusco for two main reasons. First, it helps explain the specific growth trajectory of Italian industrial districts, other typologies being conceived for different and wider contexts. Second, some newer typologies (e.g., ‘hub-and-spoke’ and ‘satellite’ clusters) resemble the second model envisaged by Brusco,34 which is broadly characterised by a dependence of the local SME cluster on a (few) large firm(s). In our view, the four models identified by Brusco, represent a ‘sequence’ of stages in the growth trajectory of Italian industrial districts. Thus, we argue that passing through stages has been an essential aspect of their development. This consideration can have real meaning for cluster development in developing countries, where few or no steps have yet been experienced and where policy-makers often attempt to cross too many stages at once. The sequence of stages followed by many Italian industrial districts is not meant to be the only possible path for cluster development, and for that matter, we do not advocate it to be the best. There may be other development trajectories that can allow clusters to grow. Small hightechnology firms could be the leaders of a development at the local level, making the presence of large firms redundant. Business incubators, technology parks and public procurement programmes may be other solutions that could to stimulate local investment to shift production activities from craftwork to industrial production. In fact, in our view, what really matters in the context of ‘survival clusters’ is for the latter to be able to leap from artisanal production to industrial production. We also recognise that in a globalised context shaken up by the entry of very powerful economies such as China, India and southeast Asia, one
84
SME Cluster Development
may cast some doubt on the relevance of the above-described sequence. Nonetheless, we would emphasise the validity and novelty of this overall analysis, that is, the need to promote the gradual development of SMEs clusters, so that policy targets feasible stages one at a time and, through a feasible trajectory of growth, pushes these clusters to move from craftwork to industrial production. A second relevant aspect of this proposed approach to cluster development is based on important streams within SME cluster literature. Indeed, this analysis identifies the influence that three general types of factors (i.e., ‘collective efficiency’, ‘social embeddedness’ and ‘policy inducement’) had on this development process. On this basis, we argue that these types of factors work concurrently to upgrade cluster development, which can have significant meaning for less successful clusters in developing countries. It could persuade policy-makers to overcome the traditional fragmentation of support and to organise an ‘eclectic approach’ to cluster development that should contribute to create a more competitive industrial atmosphere.
2. The analysis of development stages The first objective of this chapter is to establish a methodological approach that can have important effects on policy-making for clusters in developing countries. This can benefit naturally from the historic analysis of industrial district development and attempts. This ‘stage approach’ that we propose here, does not intend to be prescriptive, but rather to coordinate the value of models of SME cluster development on a temporal basis (i.e., a specific stage is relevant for those clusters that are at a lessdeveloped stage), therefore to provide policy-makers with a more flexible theoretical platform. This idea is based upon Brusco’s (1990) identification of four historic models that synthesised the debate on local production systems in Italy from the 1950s onwards. The first of these concerns the process of agglomeration of small workshops, specialised in traditional manufacturing activities, which started a few years before the Second World War in the Third Italy. The second model relates to the 1950s and the early 1960s, in which a few large enterprises took the lead in local production systems, concentrating manufacturing activity, but also spurring the shift from craft to industrial production (i.e., production through scale economies for mass consumption). The third model refers to the crisis of the Fordist system, which promoted an intense creation of dynamic SMEs capable of aggregating and
A Stage and Eclectic Approach
85
jointly producing and marketing their products abroad. This started at the end of the 1960s and went on until the end of the 1980s. A fourth model has been discussed in the past 10–15 years. It refers to a globalised environment in which competition forces SMEs to shift up the technological frontier to avoid the ‘low-road competition’ of newcomers (Pyke and Sengenberger, 1990; Kaplinsky and Readman, 2001; Guerrieri and Pietrobelli, 2004). In our view, these four models closely represent the historic sequence of the stages of development of industrial districts. The analysis of stages is a useful exercise, even in the context of developing country clusters, since it prevents local policy-makers from crossing too many stages at once during too ambitious development efforts. This particularly occurs with developing country clusters that, for some researchers, do not show much potential for growth, such as the so-called ‘survival clusters’ and ‘satellite clusters’ (Markusen, 1996; Altenburg and Meyer-Stamer; 1999; Knorringa, 2002). When policy-makers in developing countries attempt to upgrade craft type of agglomerations directly into the most advanced model of industrial district development, failure is often the likely outcome. This work identifies the different stages of development of successful Italian industrial districts and, in doing so, proposes feasible and practical short-term development steps for less dynamic clusters in developing countries, as well as the policies that can produce effective results in such a timeframe. As mentioned in the introduction to this chapter, this analysis does not suggest that every SME cluster has to necessarily pass through all of the above-mentioned stages of growth. It rather intends to reaffirm that the development of clusters is a stage process, whose steps and trajectory need to be analysed in detail, and policies thought through in terms of feasibility, in order not to lead the programme to failure.
3. Summing up the eclectic analytical framework The second objective of this chapter is to establish an ‘eclectic theoretical framework’ that helps to interpret the growth process of Italian industrial districts by combining a set of hypothetical key determinants of development. Briefly, this section summarises the above-mentioned theoretical approaches to industrial district development that have been described in more detail in Chapter 4. The identification of their relevant key features35 helps set up our specific theoretical framework. An old economic approach emphasises the relevance of the natural resource endowment as well as transportation cost to the markets
86
SME Cluster Development
(Losch, 1940). The more recent analysis of ‘centrifugal’ and ‘centripetal forces’ revives the strength of this old economic approach to the analysis of the forces that stimulate firms to agglomerate in general, but also to create geographical clusters (Krugman, 1998). However, this approach does not deal with the sectoral kind of clustering, but just with geographical clustering. Thus, this stream of economic literature deals with the generic concentration of all types of industries. This consideration makes this approach less relevant to our purposes (i.e., analysing sectoral and geographical concentrations of small firms) and will not be dealt with in detail in this work. Another approach emphasises the ‘spontaneous’ development of Italian IDs. The free interplay of market forces catalysed the districts’ growth by pulling together rich endowments in labour and natural resources, and by benefiting from the dense social and institutional capital that supported the local ‘collective efficiency’ and related competitiveness (Best, 1990; Brusco, 1990; Schmitz and Musick, 1994). If the industrial district development process depends, however, on these region-specific conditions, it becomes hard to replicate elsewhere (e.g., in developing country clusters). This spontaneous process cannot be substituted by any policy, which can only lightly support spontaneity, through the provision of ‘real services’ (i.e., business development services) at the local level. In this way, this approach could facilitate the subsequent development in both Italian industrial districts and other SME clusters (Brusco, 1990; Schmitz and Musick, 1994). A further important approach takes a different stance by emphasising the developmental role of national and international policy frameworks (i.e., the ‘policy inducement’ of development). Specifically, its theorists observe the history of the Third Italy and stress the role of national policies in shaping a dynamic economic environment. Throughout the twentieth century, various laws for the promotion of SMEs were implemented and these supported their competitive upgrading (Bertini, 1998; Bianchi and Parrilli, 2002a). This approach has implications for the policy perspective that transcends the limits of the local level and anchors the development of SME systems to national and international policy frameworks (Arrighetti and Seravalli, 1997; Bianchi, 1998; Cowling and Sugden, 1999). This ‘policy inducement’ can be complemented through a new stream of the literature that emphasises the role of global value chains and the powerful actors that deal with the cluster from outside, such as large distribution chains and whatever kind of other important actors that can direct the development of chain and clusters towards its own objectives (e.g., universities, state enterprises in gas, in weapons, etc.). This approach
A Stage and Eclectic Approach
87
still recognises the important role that the concrete governance system assumes in the specific context of clusters (Humphrey and Schmitz, 2004; Pietrobelli and Rabellotti, 2004). These two approaches (‘spontaneous development’/‘collective efficiency’ and ‘policy inducement’), which emphasise very important aspects in the development of industrial districts, should be looked at together. ‘Spontaneous growth’ represents the endogenous dynamism of the local context, which, through its rich economic and social basis, advances along a notable process of development. ‘Policy inducement’ (at the international, governmental and local level) represent the legal framework that national governments and international organizations/communities (e.g., the European Union, and the World Trade Organization) provide to support cluster development. ‘Spontaneity’ guarantees a ‘bottom-up’ movement to development (e.g., local people’s initiatives creating new market opportunities through the organization of local trade fairs; local business incubators boosting technological upgrading), while ‘policies’ guarantee a ‘top-down’ action which may benefit firms through laws and programmes set up by enlightened authorities. Combining these two approaches may help to achieve an important outcome. Although this is quite an important theoretical consideration, deeper insight is required. These two kinds of factors alone do not explain industrial district development, because, for example, the situation in most of southern Italy questions both approaches. In respect of the first approach, the region exhibits several spontaneously created small firm clusters bordering on success but never quite making it (Amin, 1994). In contrast to the second approach, SME support initiatives have been implemented throughout the country but did not help regions to create successful clusters as in central-northern Italy. Simultaneously, in the latter region, very different degrees of policy support produced similar levels of success, such as in the case of Emilia-Romagna versus Veneto and Lombardy (Belussi, 1999; Pietrobelli and Rabellotti, 2002). Thus, a further perspective is needed to explain why certain SME clusters developed and others did not. We call it the ‘social-embeddedness approach’. This is endorsed by ‘spontaneous approach’ theorists, but with different emphases and prospects. In fact, they consider the ‘socialembeddedness approach’ to be just one of the many features of the ‘spontaneous approach’ and to be a hard-to-influence feature for which no special policy can be designed. In contrast, we would like to consider it separately since it refers to an area of human behaviour not driven solely by economic motivation and which therefore needs to be managed apart (see section 4.3 in Chapter 4). For instance, the dynamism of
88
SME Cluster Development
local agents does not only depend on economic reasons (e.g., profit, risk, etc.), but also relies on non-economic elements, such as the social value attributed to work in one’s own firm, the value attributed to managing an independent activity, and to being one’s own boss, and so on (Hjalager, 1993; Kantis, 2004; Aranguren et al., 2005). Augmenting these forms of social capital cannot be achieved through economic policies alone, but by way of a wider array of approaches and policies, not least social (Bianchi and Labory, 2004; Aranguren et al., 2005). Furthermore, we would like to systematise this ‘social approach’ to SME cluster development by combining the traditional concept of ‘trust’ with another that is less frequently treated: ‘self-realisation’. Some theorists emphasise the concept of ‘trust’ as a fluid that helps to reduce transaction costs and ease cooperation and ‘collective efficiency’ (Bagnasco, 1988; Becattini, 1990; 2000a; Lorenz, 1992; Trigilia, 1992; Dei Ottati, 1994; Nadvi, 1999b; Schmitz, 1999). Trust is a key aspect of the necessary ‘social cohesion’, which strengthens the capacity of a district to work as a system. This ‘social cohesion’ promotes cooperative attitudes and experiences (e.g., cooperatives, credit consortia, and business associations). A more complete socio-economic approach would also focus on factors that encourage the dynamic capacity of the local system within the wider marketplace. This is based upon the spirit of ‘self-realisation’. In the context of successful industrial districts and clusters, this spirit can be seen through the outstanding push towards entrepreneurship, innovation, firm creation and spin-offs (Becattini, 2000b; Kantis et al., 2002; Kantis, 2004). In this sense, studies indicate the importance of institutions such as ‘sharecropping’ as historic bases for the creation of this individual dynamism (Brusco, 1982; Bagnasco, 1988).36 The important history of arts and corporations in the Third Italy is also said to have provided the necessary stimuli for the continuous trend of small firm creation (Platteau, 1994; Putnam, 1993). The key point here is that ‘social cohesion’ and ‘self-realisation’ are not to be taken in isolation from one another, but rather as interdependent factors that jointly produce a ‘positive sum game’ for the local system as a whole. ‘Social cohesion’ alone is more likely to generate a socially comfortable but economically static society. ‘Self-realisation’ alone is more likely to create a dualistic society, in which a small minority of firms are able to enter the international markets and cross the technology frontier, while the larger element of local production system remains linked to very traditional production and markets (see Chapter 4).37 From a development strategy perspective, the traditional social approach to clustering (linked to the ‘spontaneous approach’) has not
A Stage and Eclectic Approach
89
yet produced the necessary policy results. It has either isolated the Third Italy as a region-specific experience, limiting the discussion about possible policies applicable to other contexts, or it has maintained a very localised policy approach, focusing on the creation and diffusion of ‘collective (economic) goods’, such as those linked to technological knowhow, trade promotion, business skills, workers specialisation and so on (Trigilia, 1992). In contrast, we highlight the need to take a strategic policy approach that includes social features (i.e., ‘social cohesion’ and ‘self-realisation’) as explicit key determinants for development. This would target direct development actions at the social context (e.g., applied schooling programmes for young people, local social forums, etc.) instead of leaving this to simple trickle-down effects from economic policies. This policy level, which we would name ‘from inside’ (i.e., ‘social embeddedness’), is to be seen as interdependent and complementary with the afore-mentioned ‘bottom-up’ (i.e., ‘collective efficiency’) and ‘top-down’ (i.e., ‘policy-inducement’) approaches in the generation of more effective processes of growth. Although the absence of one of these factorial levels will not prevent the formation of industrial districts and SME clusters, it is likely nonetheless to reduce their competitive potential. An ‘eclectic approach’ to SME cluster development that takes into account these levels, linked interdependently, may help to better explain the growth of these systems and to promote it elsewhere through appropriate policies.38
4. The stages of development of the Italian industrial districts 4.1 The passage from rural life to artisanal clusters (1920s–early 1950s) This section sketches the passage from rural economies to artisanal clusters. Starting from a condition of dispersed rural life, people started gathering to live in urban centres. This happened in a very particular way, different from other kinds of urban agglomeration. Many rural inhabitants moved to towns and decided to work in a specific sector, which after a while defined the productive and social features of that town. This cluster condition is very relevant here, since it resembles very much that of ‘survival clusters’ that are often visible in developing countries (Altenburg and Meyer-Stamer, 1999; Knorringa, 2002). Despite being very basic types of agglomerations, these were interesting starting positions, capable of benefiting from their early basic features
90
SME Cluster Development
(i.e., geographical and sectoral concentration of firms) and later of upgrading into the future industrial districts. There are inevitable differences among Italian IDs in their first stage and some kinds of survival clusters in developing countries (e.g., the sharecropping tradition of rural Italy). The question, however, is whether these differences create definitive gaps between these areas of the world, or whether they represent only region-specific aspects that do not prevent other local system from growing competitively over time, not least the above-mentioned ‘survival clusters’, at least the most dynamic among them. We support this second interpretation on the basis of a five-year period of direct observations and applied research for universities and international organisations in what are defined ‘survival clusters’ in the context of Latin America (Barahona et al.; 1998; 1999; Parrilli, 1998a; 1998b; 1999a; 2001; 2002a; 2002b; 2003; Narváez and Parrilli, 2000; Bianchi and Parrilli, 2002a; Artola and Parrilli, 2006). This experience allowed us to observe several key features of industrial district competitiveness (e.g., joint actions, external economies, active entrepreneurship) at work all of the time in specific ‘survival clusters’ (i.e., the most dynamic among them), which make me take a positive view on the development perspectives of these types of SME clusters (on these considerations, see the Central American case studies in Chapter 6). This first stage differs from the development of the industrial triangle of northwest of Italy (Milan-Genoa-Turin), because the people that arrived in the towns of the industrial districts came from local rural areas and not from the South (Becattini, 1990; Murat and Paba, 2001). Economic factors operated in two directions in stage I, one acting generally to encourage agglomeration and manufacturing development, and another that favoured the growth of specialised production. In the first case, the classic phenomenon of the (expected) wage gap among Table 5.1:
Stage I: the 1920s to the early 1950s
Main trend in stage I
Economic factors
Social factors
Policy factors
Urbanisation and craft agglomeration
Urban-rural wage gap, natural endowment, local market, craft skills
Family links in town, entrepreneurial spirit, tradition of craft work and corporations in towns
Statute for small firms and craftwork, basic infrastructures, public employment
Source: Author’s compilation.
A Stage and Eclectic Approach
91
rural and urban activities encouraged rural people to leave the countryside to search for a job in town (Harris and Todaro, 1970). The actual existence of a local market – despite the rather reduced size of towns in this area – guaranteed that the rural migrants would find enough consumers for their products. In the second case, specialised manufacturing production arose thanks to the local (or easily accessible) natural resource endowment (Losch, 1940). The abundance of wood, for example, promoted the development of furniture industries in Brianza and Appennini; the caves in the TuscoEmilian hills promoted the development of the tile industry in Sassuolo; and cattle farming in the Padany countryside helped to create a strong dairy industry in Parma. Another important element was the craft skills that the population had practiced for centuries in the production of food, shoes, clothes and furniture in both rural and urban areas. Evidence of ‘social cohesion’ and ‘self-realisation’ can be seen throughout this urban development. The migration of thousands of farmers was promoted by family linkages with those people who had already moved to town and who helped newcomers to settle down. Simultaneously, the rural family indirectly assisted their migrating relatives by providing a security net in case they returned ‘home’ (Bagnasco, 1988). With regard to ‘self-realisation’, these migrants participated actively in urban life. They avoided unemployment by creating micro-enterprises that started a process of accumulation. They benefited from both the tradition of skilled craftwork rooted in the history of medieval arts (Putnam, 1993; Platteau, 1994) and from the traditional sharecropping practice, which had accustomed the peasants to working hard, effectively (generating a high relative productivity) and to taking risks to reap the maximum possible harvest (Brusco, 1982; Bagnasco, 1988; Bellandi, 2001). The third general type of factor at work was ‘policy inducement’, which history demonstrates was relevant at that time too. In Italy, from the 1930s onwards, the state organised specific policies for the legal protection of micro, craft and cooperative enterprises, such as the Law of Failure in 1942, which gave craft enterprises an important advantage. This was ‘the advantage of not being subject to failure (which) is something that is not present in any other industrialised country’ (Brusco and Paba, 1997: 324; see also Arrighetti and Seravalli, 1997: 358). In the following years, these provisions were strengthened through the general law for the promotion of SMEs, approved in 1952, and Law 860/1956 for the support of craft enterprises (Bianchi, 1992; Bertini, 1998; Bianchi and Parrilli, 2002a). The institutional environment often supported these policies. Local governments and public institutions convinced the people that adequate
92
SME Cluster Development
facilities and incentives would be offered to help them adjust to urban life. For instance, the availability of electricity, telephones, water, sewage, transport, housing and markets were certainly important stimuli for the people to urbanise and raise their standard of living. At the same time, the incentive represented by employment potential within the public sector also had particular significance (Murat and Paba, 2001). These aspects were mediated by the high responsiveness of political leaders (at the demand of local businesses) within regions and clusters in the Third Italy, which has characterised their history from the Middle Ages (Putnam, 1993). On the whole, the economic (in this stage less the ‘collective efficiency’ and more the traditional factors of production), social (‘socialembeddedness’) and policy (‘policy-inducement’) spheres constituted powerful and coordinated factors in this first phase of local development. 4.2 The passage from craft production to industrial concentration (1950s–1960s) The second stage started with the growth of large factories that would later constitute the basis for a specialised industrial know-how that promoted a process of SME spin-offs. These large firms became leaders in the local development process (Brusco, 1990; Bellandi, 2001). This resulted in a generalised shift of local production towards a process of industrialisation, which can be observed through the dramatic increase in manufacturing and industrial output over those years (Barca, 1997; Cazzola, 2000; Cohen and Federico, 2001). The most interesting characteristic of this process is the structural innovation that these firms brought to the local production context. Up to that point, output was artisanal, with small craft production oriented Table 5.2:
Stage II: the 1950s to the 1960s
Main trend in stage II
Economic factors
Social factors
Policy factors
Industrialisation and massconsumption
Growth of larger firms, abundant workforce with craft skills
Social cohesion within the districts, strong entrepreneurship from within the region
European Recovery Plan, National champions and first laws to finance SME development
Source: Author’s compilation.
A Stage and Eclectic Approach
93
to the local market, as opposed to the new trend of industrialisation involving large volumes of standardised goods for mass consumption (Piore and Sabel, 1984; Womack et al., 1990). A mix of economic resources, governmental policies and social strengths facilitated this local upgrading from craft agglomerations, which were no more than ‘survival clusters’, to new industrial concentrations. The entrepreneurs of the new large firms were often locally based. This local embeddedness of capital eased the assimilation of a new and abundant workforce and lowered the intensity of conflict between the newcomers and the existing industry workers, common at the time in the Industrial Triangle and which led to the breakdown of the Fordist model (Locke, 1995). In this initial phase of industrialisation, large firms benefited greatly from the plentiful supply of labour that accepted to work for an attractive wage (compared to their rural alternative). At the same time, their craft skills allowed these large firms to achieve high productivity, which helped to sell their goods in the wider national market. At the social level, the mechanisms of ‘social cohesion’ and ‘selfrealisation’ were at work too. Good social relations were maintained thanks to the small size of the ID towns and because their large firms often employed just several hundred people (against thousands in the large factories of Milan and Turin). All of this facilitated an easier and more ‘human’ management of working relations (Becattini, 1990; Locke, 1995). During this stage, there was little evidence of individual entrepreneurship emanating from ‘self-realisation’. In fact, over this period, the people became very sensitive to higher wages and the related social security benefits offered by large firms. Moreover, large firms offered them very good opportunities to learn technical know-how and new production practices that they used directly 10–15 years later. That is why, in this phase, many artisans preferred working for large firms rather than continuing to work in craft enterprises (Locke, 1995; Bertini, 1997; Bardi, 2000). In terms of ‘policy inducement’, the state organised an industrial policy based on renewed support from the European Recovery Program (i.e., Marshall Plan). It oriented its incentives to stimulate the growth of public enterprises and ‘national champions’.39 General and sectoral state policies subsidised the basic industries of iron, steel and chemicals, as well as the manufacturing of automobiles and electric goods (Bianchi, 2002b). Notwithstanding this, the government approved the first laws for the promotion of small and medium firms, thus making them recognised in the institutional arena (e.g., from banks). From the institutional point of view, increasing public sector employment and the provision of basic infrastructures continued to stimulate the migration of rural population to towns.
94
SME Cluster Development
Although this second development stage of the future industrial districts was very short (10–15 years), it was critical in that the economic system shifted from artisanal production to industrialisation. State policies and migration helped a few local and national entrepreneurs to create their own large industries in specific localities in the Third Italy, creating leaders of local development which would catalyse further changes and favour the passage to the following development stages. The SMEs maintained a more marginal role; but their social milieu started structuring for the boom of the 1960s and the 1970s. 4.3 The passage from industrial concentration to industrial districts (1960s–1980s) The third stage represents a more traditional configuration of industrial districts. It concentrates many SMEs that are able to work together through a detailed division and specialisation of labour and to sell their finished goods in the international market. The catalyst for such a structural change was the crisis of the Fordist production system that culminated in the late 1960s. Overproduction generated serious supply and demand imbalances, while economic and social conflicts among capitalists and workers led the system into a confused situation in which large cities and firms lost their leadership positions. In this context, a new feature of production came about in specific geographical regions: dense networks of SMEs capable of working together and offsetting the scale and scope advantages that large enterprises had thus far enjoyed. ‘Joint actions’, ‘external economies’ and flexibility constituted the new competitive forces of small firms, capable to respond to the new type of demand for more personalised and innovative goods (Best, 1990; Schmitz, 1992). While large firms entered a deep crisis, their workers – within the industrial districts – left and set up their own small firms. The birth rate Table 5.3:
Stage III: the 1960s to the 1980s
Main trend in stage III
Economic factors
Social factors
Policy factors
Increased production and internationalisation of SMEs
Crisis of Fordism, collective efficiency among SMEs. EC Common Market.
Preference for independent and family work. Trust among people
Sabatini and Ossola laws, local governments’ and Business associations services
Source: Author’s compilation.
A Stage and Eclectic Approach
95
of SMEs within the districts increased from 27 percent in 1951 to an astonishing 99 percent in 1961. Employment within SMEs increased from 59 percent in 1951 to 65 percent in 1961 and 73 percent in 1971, with a parallel decline in large firms’ share (Brusco and Paba, 1997: 288, 292). Often this occurred through collaboration between new entrepreneurs and their former employers, who sold or loaned machinery to specialised workers in order to externalise production, and reduce fixed costs and risks. As part of the more general process of outsourcing, the wider market controlled by ‘surviving’ large firms guaranteed a rather secure basis for small firms’ output and helped them grow undisturbed (Lazerson, 1990). This solution permitted the local production system to cope with the crisis, open up new markets and experiment with new organisational models. This collective response to the crisis of mass production cannot be explained without reference to the social factors of development, which explain why workers did not surrender to the crisis, but instead fought to find a common solution. In this environment, where more flexible practices were needed, ‘self-realisation’ and ‘social cohesion’ were crucial, through incentives for subcontractors, for example (ibid.). ‘Self-realisation’ is exhibited in the entrepreneurial spirit of the people, based upon the socialised preference for independent activity. The desire prevailed to be one’s own master, to be able to decide the daily timetable, the commercial relationships, the workers, the type of investments, and so on, even though this might temporarily imply earning less and facing higher risks (Hjalager, 1993; Kantis, 2004). The crisis of the Fordist system facilitated the re-emergence of this entrepreneurial spirit and pushed many people to undertake an intense process of small firm creation. Evidence of ‘social cohesion’ in this stage is clear because most of the people in the industrial districts had known each other for decades. These localities were thus characterised by a trustful atmosphere which eased the start-up of SMEs. This trust did not need to be based upon ethnic roots, as occurred in early phases of growth in many developing country clusters (Knorringa, 1999; Nadvi, 1999b; Schmitz, 1999), simply because industrial district societies were homogeneous at that time in terms of their culture, traditions and economic practices (Becattini, 1990). In industrial districts, trust depended more on the positive repetition of production and trade exchanges among local actors, which allowed them to know each other better, diffuse information about each other’s reputation (Lorenz, 1992; Dei Ottati, 1994; Platteau, 1994; De Propris, 2005) and recognise the ‘cost of non-cooperating’ (Schmitz, 1999; Axelrod, 1984). As a substantial factor of ‘social cohesion’, the value of the family cannot be underestimated. The SMEs succeeded because entrepreneurs
96
SME Cluster Development
had been supported by their families through long working hours, shared responsibility, extended skills and the capacity to control the production and marketing cycle (Bagnasco, 1988; Becattini, 1990). The family is probably the factor of ‘social cohesion’ that buttressed the development of these local production systems, sometimes in a more hidden and implicit way (stage II), and other times in a more explicit and active way such as through this stage III. To enable the local economies to reach the third stage of development, these local production systems were also supported by government policy, through important laws for SMEs development, for example. An illustration is the 1329/1965 Law, also known as the ‘Sabatini Law’, concerning the technological upgrading of firms and the 227/1977 Law, or the ‘Ossola Law’ on export promotion (Bertini, 1998). Thanks to Sabatini, more than 200,000 firms were able to buy new machinery on a leased basis and to upgrade their technological assets over a period of 30 years at an average value of US$ 50,000 per firm. Thanks to Ossola, small firms increased their capacity to export (ibid.). These laws were approved as part of the conscious effort by the state and national and/or local societies to help SMEs to upgrade and join the international market competitively. In important industrial district regions, the legal and policy system was complemented by efficient public institutions, which constituted the organisational framework that supported the implementation of laws and regulations and promoted the realisation of this development stage. Often, local institutions played the role of public service agents that promoted the deep division and specialisation of labour which is typical of industrial districts.40 These institutions were important not only for the policy tools they represent, but also for the daily support they gave to the local production system (Best, 1990; Brusco, 1990; Bellini, 2005). In the Third Italy, there are several examples of such important contributions to industrial district success, although stronger in some regions (e.g., Emilia-Romagna and Tuscany) and weaker in others (e.g., Veneto, where all support activities were organised privately by individual enterprises or business associations and Lombardy, where state institutions preferred to encourage local governments and business associations) (Belussi, 1999; Pietrobelli and Rabellotti, 2002). 4.4 The passage from traditional IDs to new competitive IDs (1980s–2000s) This stage covers new challenges that local production systems have been facing in the new globalised market, in which they compete with other production systems, often led by large competitive firms. In this
A Stage and Eclectic Approach Table 5.4:
97
Stage IV: the 1980s onwards
Main trend in Economic factors stage IV
Social factors
Policy factors
New global competition
New education/ career models, new types of social cohesion/linkages
Law 44/86 on firms’ creation, Law 46/82 on innovation, wider institutional networks
Knowledge and innovation, groups of SMEs, multinational networks
Source: Author’s compilation.
new context, other competitors, such as continental China and other East Asian and eastern European countries can compete in the global market for the same consumers. From an economic point of view, the new competitive environment is wider in scope and constantly changing, because it addresses several growing consumers segments with particular tastes and preferences. Thus, new competitors cannot alone, rely on price since consumers expect more and more innovative and personalised goods and services (Pyke and Sengenberger, 1990; Cooke, 1996; Kaplinsky and Readman, 2001). As knowledge creation lies at the centre of this new competitive arena, the flow of this knowledge among the chain’s economic agents assumes a strategic role. In particular, individual employees are subjects of knowledge, especially in its tacit forms (e.g., transfer of know-how), which accompanies them to their next position. Thus, although ‘distance is no (more) important when transmitting information, still it is important when we refer to the transmission of knowledge and tacit knowledge, which now are the strongest base for industrial development and competitiveness’ (Audretsch, 1998: 21–2; Cooke and Wills, 1999). In this new era of globalisation, the continuing importance of physical clustering is also based on other aspects of knowledge that assume a growing importance in the context of high-tech industries. These refer to firms’ efforts to generate their own innovations (e.g., through R&D investment) and to the intangible capital accumulated in the cluster that favours incremental innovations. It also depends on the local capacity to appropriate benefits from the innovative activity (Bianchi and Labory, 2004; Guerrieri and Pietrobelli, 2004). Complementarily, an easier diffusion of codified knowledge through the ever-growing installed base system of information and communication technologies is producing big changes in the configuration of local
98
SME Cluster Development
production systems, in terms of both markets and production. Necessary production facilities (i.e., factories) can be set up far from the industrial districts – for example, the experience of many plants and joint ventures in eastern Europe during the past decade by ID furniture and leather enterprises. Simultaneously, leading firms tend also to adhere to ‘multinational networks’, for instance in optical and footwear industrial districts (Gilly and Torre, 1998; Storper, 1998). If the first type of operation tends to take place in search of cost advantages, such as cheap labour, the second is more typically focused on closeness to top innovative productive and market networks that add new competitive advantages to the firms and clusters that exploit them (Cooke and Wills, 1999; Guerrieri and Pietrobelli, 2004; Rabellotti, 2004). These new situations have motivated a stream of theorists to study the role played by global value chains and their governance system. They recognise the importance of exploring the linkages existing between local producers and other agents from the supply and distribution chains. The value of goods is added through the various links of the chain to different extents depending on the kind of agents involved (e.g., SMEs, top brand companies, and so on). This highlights the importance of governance within global value chains, which studies the ways the agents relate to each other in a variety of systems (i.e., hierarchy, quasi-hierarchy, network, and market) and defines the effective opportunities and threats that they face in the market (Gereffi and Korzeniewicz, 1994; Humphrey and Schmitz, 2004; Rabellotti, 2004; Sacchetti and Sugden, 2005). In this sense, if local production systems (i.e., IDs) maintain their importance, linked to the generation and transmission of tacit forms of knowledge and to the special characteristics of innovation (i.e., opportunity, appropriability and cumulativeness), global value chains acquire increasing weight, linked to the dynamics of market power that determine the changing assignation of product values among the different economic agents (Humphrey and Schmitz, 2004; Pietrobelli and Rabellotti, 2004). As a result of these changes and relations in the global and local environment, industrial districts exhibit growing interrelationships and joint ventures between firms and institutions. The case of high-tech spin-offs is an example of this new type of networking (e.g., between local institutions, universities, business incubators and firms). Although, it mainly occurs on an individual or small group basis within high-tech sectors (e.g., sea explorations, health industry, naval engineering: Baroncelli, 2001; European Union, 2001), there are growing efforts made by European institutions (e.g., European Business Network – EBN), state agencies (e.g., Sviluppo Italia), local institutions and business associations to apply
A Stage and Eclectic Approach
99
this approach in a more systematic way in local production systems and IDs (Bianchi and Parrilli, 2004; Kantis, 2004). Another recent strategy of industrial districts is the formation of ‘groups’ of firms, which demonstrate the importance of scaling back where necessary to a relevant size of operation. This aspect is not very important in production scenarios, in which economies of scale have been ensured by subcontracting in IDs for decades. In contrast, this new form of grouping refers to associated practices among a limited number of local leading firms in aspects of marketing, distribution, and research and development. It would be the case, for example, of the constitution of promotion consortia (Brioschi and Cainelli, 2001). In these sectors, new kinds of grouping create new economies of scope and agglomeration, strengthening their competitive edges against competitors from East Asia and other countries with low labour costs. The case of the tile industrial district in Sassuolo is the clearest example of this trend (Russo, 1989). In terms of ‘policy inducement’, policies and laws are also at work in stage IV. These instruments represent the framework that defines the environment for SME operations and the incentive systems that supports their dynamism. Streams of thought exist that analyse the important power relations within and outside of the local production system due to the existence of key groups of interests that play the development game from opposed perspectives, a situation which can create severe constraints to SME cluster development (Cowling and Sugden, 1997; Bianchi, 1998; Guerrieri and Pietrobelli, 2004; Humphrey and Schmitz, 2004; Rabellotti, 2004; Sacchetti and Sugden, 2005). Some of these theorists stress the need for a wide approach to policymaking that includes several aspects of competitiveness, such as institutional support and competition policy, public goods, services and infrastructures, innovation, territorial agglomeration and small firms. All of these aspects and their related policies can work efficiently when they are integrated within a systemic framework and vision, such as that applied in the formation of the European Union (Bianchi, 1998). Others emphasise the importance of upgrading SME (cluster) management to international standards (e.g., by using ICT), as well as the capabilities and capacities that guarantee access to high value-added productions. This may help prevent industrial districts from losing power to the strongest actors within global value chains (Guerrieri and Pietrobelli, 2004; Humphrey and Schmitz, 2004). During this stage, the government was active in introducing and implementing industrial policies. In the case of the Third Italy, various legal provisions have promoted the role of SMEs and clusters in the
100
SME Cluster Development
local and the national economy. In particular, two types of regulations have been approved, which tend to spur further dynamism within local production systems – laws to support the creation of new firms, and laws for the promotion of innovation through special financial assistance and other technical public sector help (Bertini, 1998: 327).41 These legal tools have been applied in the region, but not as extensively as the most successful policy instruments of the past decades. This aspect points to limitations of these instruments that need to be further investigated. The efficient impact of policy requires a specific institutional environment, which, in this stage, is represented by the creation of the abovementioned wider public–private networks. It promotes ‘higher-road’ types of relations among economic agents, as in the case of the spin-off of small high-tech enterprises from universities, independent and applied research centres, and local governments (Cooke, 1996; Baroncelli, 2001; European Union, 2001; Bianchi and Parrilli, 2004). It represents a good example of the institutional effort to generate wider agglomeration and scope economies at the local level in search of higher value-added production. In the past ten years, new agencies have been working to upgrade these efforts to systematic forms of support to local production systems and industrial districts (for instance, Spinner and AlmaCube at the University of Bologna) (Bianchi and Parrilli, 2002a; 2004). The social aspect of development is probably the weakest at present. The strength of human capital depends not only on knowledge and skill, but also on motivation and social linkages. In the context of SME systems, the spirit of ‘self-realisation’ remains a key factor, since all the strategic responsibilities and decisions are taken by the entrepreneur. But when he/she becomes old, he/she would like to shift these responsibilities to other family members. Today, some IDs face a problem because the younger generations are less willing to accept these responsibilities. They opt for managerial roles within big enterprises and large bureaucracies (e.g., universities, the state, public enterprises) and leave their parents’ activity, rather than extending their own and the firm’s capabilities into newer fields such as electronics and ICTs, marketing, etc.42 This weakness helps explain why several industrial districts are under pressure and change configuration, for instance, by increasing the size of leading enterprises and creating more selected ‘groups’ of SMEs (Brioschi and Cainelli, 2001). In terms of ‘social cohesion’, SMEs still need to organise ‘joint actions’ as a way to compete against larger firms. But the globalisation of market dynamics and new waves of immigration (i.e., from eastern Europe and North Africa) have changed the context of ‘social cohesion’. Transaction
A Stage and Eclectic Approach
101
costs are rising as the social fabric has been changing and economic agents do not know each other. For instance, more formal contracts are needed. New formulae for ‘social cohesion’ are therefore needed to coagulate the more heterogeneous population and their SMEs around complementary objectives (Bianchi, 2000). These represent the new challenge for industrial district economic systems and a new borderline between competitive and declining IDs. These forms of cooperation are not generated through economic actions aiming to achieve higher competitiveness, which tend to be short-term oriented. There is a need for some deeper and longer-term linkages, which can only be guaranteed at the social level. What are the social features of present industrial districts that can help maintain the previously ingrained socio-economic atmosphere? It is a crucial issue upon which to base a discussion of the journey into new competitive stages.
5. Two policy keys for ‘survival clusters’ in developing countries This study does not pretend to be exhaustive about the lessons that can be extracted from Italy’s industrial district benchmarking practices. It rather analyses two key aspects in the trajectory of development of Italian IDs and, on those bases, singles out two policy suggestions for a specific (but large in number) type of developing country cluster – the above-mentioned ‘survival clusters’. Let us analyse them in detail. 5.1 Policy key I: the importance of upgrading through stages This analysis specifies the several stages of development through which Italian industrial districts have passed before arriving at their present position. Identifying the sequence of development stages seems a relevant prerequisite for structuring appropriate policy schemes for the development of SME clusters (in particular, the most dynamic ‘survival clusters’) in developing countries. In particular, a large number of clusters have been defined as ‘survival clusters’, which means agglomerations of independent uncomplementary craft enterprises producing a limited variety of individual products of rather low quality for local markets. These types of clusters are said not to have visible growth process and/or potential (Altenburg and Meyer-Stamer, 1999; Knorringa, 2002), which make them less interesting from both a research and policy perspective. We would rather like to offer a different and more positive view on (at least some of) these clusters. The following considerations can be useful also for other categories
102
SME Cluster Development
of SME clusters in developing countries, such as the ‘satellites’ of large companies working outside the cluster and, to a lesser extent, ‘hub and spoke’ configurations, in which local SMEs can be too dependent on the core large firm to be able to develop autonomously over time (Markusen, 1996; Knorringa, 2002; Guerrieri and Pietrobelli, 2004). ‘Survival’ and ‘satellite’ SME scenarios are often negatively assessed as having little development prospects, especially when one benchmarks their experiences with other clusters without taking into account their sequence of development up to their present competitive capacities. However, this work focuses on industrial district development from the simplest historic stage (i.e., craft agglomeration) to the most complex (i.e., new competitive IDs) in order to emphasise that stages are essential steps in a trajectory of growth (Humphrey, 1995; Knorringa, 2002; Guerrieri and Pietrobelli, 2004). The first lesson stresses that identifying the actual development stage in each cluster helps one understand the effective potential and better plan support policies and actions. For instance, current prospects for economic growth within ‘survival clusters’ point to the need to leave behind the craft-type mode of production and to pass to a true phase of industrialisation, characterised by production in larger series of higher quality goods for wider markets. While in the case of Italian IDs, large firms offered support, SMEs and SME clusters in developing countries run the risks of ending up in a satellite condition of exploitation or marginalisation (Hymer, 1972; Portes et al., 1989; Cowling and Sugden, 1997; 1999; Knorringa, 2002). For this reason, governmental bodies, policy-makers and international agencies tend to directly target the most advanced stage of industrial districts, very often with failure. Failure is therefore the result of short-cuts taken by institutions and firms attempting to reach the end result of competitive and sustainable industrialisation in their SME clusters. Too many bottlenecks and market failures in financing and access to higher technology and wider markets indicate that some development steps are missing, with particular reference to stage II – not for the importance of relying on large firms, but rather for the key passage from craft production to industrialisation. The Italian experience nevertheless indicates that SME systems can successfully pass through this stage by adopting a particular approach to large firms. These industrial districts were able to reap significant benefits from large firms, in particular from their capacity to invest in new and costly technology and from their access to wider markets – both of which represent harder targets for SMEs alone (Lazerson, 1990; Fauri, 1996; Bertini, 1997). While working with large firms, prospective SME entrepreneurs were also able to start a learning process in aspects in
A Stage and Eclectic Approach
103
which large companies are best performers (e.g., division and specialisation of labour, cost-cutting practices, high marketing and innovation profile, and access to credit). Of course, there are other paths to help SME clusters to reach the same outcome (e.g., industrialisation). These can be based upon other forms of leadership, such as the entry of small high-tech firms (Baroncelli, 2000; European Union, 2001; Bianchi and Parrilli, 2002a), the use of public procurement (Tendler and Amorim, 1996), the creation of business incubators (Schmitz and Musick, 1994; Kantis, 2004), and so on. But what is most important is the relevance of a step that is too often neglected in (policy for) the most dynamic ‘survival clusters’: achieving industrialisation. The analysis of stages is a useful exercise for developing country clusters since it prevents local policy-makers from crossing too many stages at once, which makes the development process often too ambitious and failure very likely to occur (e.g., when governmental bodies plan to support the export capacity of SME clusters focusing on traditional micro enterprises). The result of this work and its related policy consequences do not intend to reach prescriptive conclusions, but rather to consider models of SME cluster development on a temporal basis (i.e., each cluster needs to anticipate its most likely next phase, rather than jumping to the most advanced stage of development), and therefore offer policy-makers a more flexible theoretical platform. Adopting this kind of strategy may be helpful also in the most dynamic among the ‘survival clusters’; it is likely to be the only way to build up a local capacity to progressively offset other booming production systems (e.g., Guang Dong clusters in China) that may jeopardise the future prospects of development of less dynamic clusters in Latin America and in other developing regions. 5.2 Policy key II: the relevance of an eclectic approach From the Italian experience, this chapter singles out a second valuable element to help SME clusters in developing countries to upgrade. In Italy, the construction of competitive industrial districts has not only resulted from a balanced micro- and macro-economic process, but from a more complex process that took into account a set of economic forces (i.e., bottom-up), blended with social forces (i.e., from inside) and with targeted governmental policies (i.e., top-down). For instance, a lack of spontaneous economic forces (for instance, ‘joint actions’, ‘external economies’, and so on) would have reduced the competitive advantages of IDs and their capacity to offset all kinds of market failure and large firm power (Schmitz, 1992; Nadvi, 1994; Rabellotti, 1995). The lack of ‘self-realisation’ and ‘social cohesion’, which, in the context of SME clusters, boost entrepreneurial spirit and cooperative actions, would
104
SME Cluster Development
have prevented, for example, the close relationships between SMEs and large firms that led to the widespread phenomena of enterprise spin-offs (Brusco, 1982; Becattini, 1990; 2000a; Platteau, 1994). The same can be said about ‘policy inducement’. Without government tools such as the law for the technological restructuring of SMEs, the problem of accessing investment credit would hardly have been solved (Cooke, 1996; Bertini, 1998). The lack of a law to promote export consortia would have probably reduced the SMEs’ speed of insertion in international markets (Bianchi and Parrilli, 2002a). On the whole, these disadvantages would not have impeded the formation of competitive IDs, but would have more likely reduced the speed and overall success of their development. This discussion is important, because policy-makers and development agencies in developing countries too often compare their SME clusters to the top competitive clusters in the Third Italy and Silicon Valley, among others. But they do not take into account the special social, economic and policy features of those production systems and the several phases they have passed through. Thus, such an abstract comparison leads them to consider their own clusters to be of static and ‘survival’ type (Altenburg and Meyer-Stamer, 1999). In this sense, this second consideration highlights the importance of creating a development framework in which specific economic, social and policy aspects are combined through an ‘eclectic approach’. This means taking into account aspects that are usually left out of policy programmes (i.e., social factors) and targeting specific policies (e.g., through education schemes targeting ‘self-realisation’ as a driver to entrepreneurship; through local social forums easing relationships and supporting cooperation between the local agents). These social factors would then be harmoniously combined with the system’s economic and government policy factors to construct a more complete and integrated development process that is likely to make the SME cluster more competitive in the global market. In the next chapter, this work explores these hypotheses in three different contexts, a cluster in an industrialised country and two clusters in developing countries. On the basis of the literature on clusters in developing countries, the latter two would be defined as ‘survival clusters’. Through this empirical exploration, we want to identify the strengths and limits of cluster development in the case of ‘survival clusters’, but also to describe some elements of a visible dynamism, which, in specific cases (the most dynamic ‘survival clusters’), and under specific conditions, could be improved a lot by taking a ‘stage and eclectic approach’ to SME cluster development.
6 Empirical Explorations in Survival Clusters in Central America and in Competitive Italian Industrial Districts
1. Introduction Theoretical and empirical research on the clustering of firms shows that this form of networking is an important means in promoting the competitiveness of small and medium enterprises in the global market. In the specific context of developing countries, among other positive effects, the clustering of small firms can also help avoid typical economic and societal divisions that hold back the escape from underdevelopment (Parrilli, 2004a; see also Chapters 3, 4 and 5). In this chapter, the analysis focuses on the theoretical framework that has been developed in Chapter 4 and formalised in Chapter 5, and its potential effects for analysis and policymaking in the context of ‘survival clusters’, which are quite common in developing countries, and, within that category, of the most dynamic. The two hypotheses presented in sections 2 and 3 of Chapter 5 provide the foundation for the approach taken here. The first suggests that SME clusters pass through various stages in a trajectory of growth. Analyses made on the history of industrial districts in Italy open up the possibility that SME clusters pass through stages in a process of social and economic upgrading. This hypothesis introduces possibilities for the development of the most dynamic group of ‘survival’ (Altenburg and Meyer-Stamer, 1999; Knorringa, 2002) and ‘satellite’ clusters (Markusen, 1996; Knorringa, 2002; 2005; Guerrieri and Pietrobelli, 2004) in developing countries, which can have the effect of dynamising policy support within those contexts. The approach represents a change with respect to existing literature, which tends to be more pessimistic about these kinds of clusters. The second hypothesis refers to different but complementary factors that contribute to the development of clusters. For this purpose, this work explores the drivers emphasised by the main theoretical and policy 105
106
SME Cluster Development
approaches to SME cluster development. The integration of these forces through an ‘eclectic approach’ might help better explain why many clusters have not grown as desired, and might also help plan more effective support policies and actions. This second aspect also introduces a modification with respect to the main policy approaches to SME cluster development, which tend to be less comprehensive, and for this reason, might be less effective. In the second section of this chapter, the subject of ‘survival clusters’ is positioned within the wide literature on industrial districts and SME clusters. Section 3 empirically investigates the impact of ‘collective efficiency’, ‘social embeddedness’ and ‘policy inducement’ on the growth process of clusters. The evidence analysed comes from two furniture clusters in Central America (Sarchí in Costa Rica and Masaya in Nicaragua) and the Italian furniture district of Forlì in Emilia-Romagna. A final section of conclusions is then presented.
2. ‘Survival clusters’ in developing countries The academic literature on SME clusters essentially focuses on very successful clusters in developed (Marshall, 1918; Brusco, 1982; Piore and Sabel, 1984; Best, 1990; Pyke and Sengenserger, 1990; Becattini, 2000a) and developing countries (Schmitz, 1992; Pedersen et al., 1994; Rabellotti, 1995; Knorringa, 1999; Nadvi and Schmitz, 1999; van Dijk and Sandee, 2002). More recently, new development trends are being explored, such as the establishment of trans-local and trans-national networks of SMEs (Gilly and Torre, 1998; Storper, 1998; Guerrieri and Pietrobelli, 2004; Rabellotti, 2004) and urban trans-sectoral networks and clusters (Audretsch, 1998; Krugman, 1998; Scott, 1998; Fujita et al., 2001). The current tendency in research on clusters is to focus on the new competitive and technological frontiers of the most advanced types of clusters. This research effort is motivated by the need to find a developmental answer to the situation of the many national economies whose foundation is based upon traditionally strong local production systems (e.g., Italy, France), and which are struggling to maintain their shares of the globalised market (Cooke, 1996; Audretsch, 1998; Bianchi, 1998; Bell and Albu, 1999; Brioschi and Cainelli, 2001; Guerrieri and Pietrobelli, 2004; Pietrobelli and Rabellotti, 2004; 2005). This focus, however, leaves unanswered the issue of other world regions and clusters not so close to the frontiers of technology. In fact, the role in development played by many ‘survival’ clusters is ignored from both a research and policy perspective. These clusters can be defined as
Empirical Explorations
107
local systems composed of many craft producers working independently from one another (i.e., no division and specialisation of labour), producing individual products of rather low quality (due to little use of machinery) for the low-income segment of local consumers. Generally, these local firms do not grow, but tend to simply turn over their capital, offering the owner’s family the means simply to subsist (Altenburg and Meyer-Stamer, 1999; Knorringa, 2002; 2005). They certainly represent part of the traditional sector of SMEs that lags behind with respect to the modern sector in Latin America (see chapters 2 and 3). These local survival systems are often neglected, regardless of their number and importance in the economic life of developing countries and despite the many support operations that numerous development agencies undertake for SMEs in those countries. For this reason, it is thought that focusing on these cases is worthwhile, in particular referring to the group of survival clusters that show the potential to grow over time.43 The work here bases its overall argument on this issue, by expanding the 1970’s slogan about small enterprises: ‘Small is Beautiful’ (Schumacher, 1973). Introducing this concept into the core of its analysis, this work hypothesises that, under specific conditions, ‘clustering is beautiful’. This approach highlights the healthy, but often small-founded steps that potentially competitive clusters also make. The chapter duly justifies the setting up of a more comprehensive and stronger policy support for such clusters. This general perspective modifies the standard observation that SME clusters in the world are not similarly competitive, and some are more so than others. Some theorists would argue that, among SME clusters, some have the capacity to grow, while others do not. Such a view embodies the thought that many clusters in developing countries tend to be ‘survival clusters’, whose economic performance tends to stagnate (Amin, 1994; Altenburg and Meyer-Stamer, 1999; Knorringa, 2002; 2005).44
3. Empirical evidence from three clusters 3.1 The context This section analyses data from three SME clusters45 in three different countries: the furniture clusters of Masaya, in Nicaragua, Sarchí in Costa Rica, and Forlì in Italy. Considering standard performance indicators (e.g., sales, income, exports and fixed assets), it becomes evident that these clusters – as well as their countries – are not equally competitive. They represent different situations and development prospects.
108
SME Cluster Development
Table 6.1:
Mode of some basic features in Forlì, Masaya and Sarchí Monthly sales
Fixed assets
Employees (avg. number)
Age of the firm (years)
Forlì, Italy
40–200,000
1,100,000
19
21
Sarchí, Costa Rica
3–6,000
30,000
8
16
Masaya, Nicaragua
2–3,500
10,000
8
13
Note: in the case of Forlì, data are in euros; data in Sarchí and Masaya are in US dollars. Source: Author’s survey, 2002–03.
The analysis here, however, is not interested in studying standard performance indicators, but rather in understanding the development process these clusters have gone through, whether they can grow further, and what future steps/stages they can effectively target. A simplified observation can be made that the Italian furniture cluster of Forlì ranks between what, in our ‘stage approach’ are the traditional industrial districts and the new competitive industrial districts (Brusco, 1990: 8–12; Parrilli, 2004b; see also Chapter 5). The two Central American clusters, currently attempting to upgrade to industrial production, represent something close to the craft mode of production, which attempts to shift to a more competitive production organisation. In this sense, they represent examples of the most dynamic cases among the abovementioned ‘survival clusters’. This preliminary consideration does not prove that all SME clusters are going to move to more advanced competitive stages. Discussing the hypothesis that they might develop is the first objective here, which should help reveal the real dynamism of most types of cluster, modifying the idea that ‘survival’ clusters in developing countries do not grow (Altenburg and Meyer-Stamer, 1999; Knorringa, 2002). The second objective is to analyse the development of the three clusters through an ‘eclectic approach’ that integrates three types of development factors representing the main approaches to SME cluster development discussed in Chapters 4 and 5 (i.e., ‘collective efficiency’, ‘policy inducement’ and ‘social embeddedness’). These aspects should help explain their different levels of competitiveness. The analysis is based upon a survey of 95 case studies of micro, small and medium enterprises that were realised between the last few months of 2002 and January 2003: 30 in Masaya, 33 in Sarchí – where only firms with less than 30 workers operate – and 32 in Forlì. These firms were identified
Empirical Explorations
109
through a random (Central America) and stratified (Forlì) selection46 based upon the universe of local enterprises presented by the database of important institutions (BCN, 1996; CNA, 2002; Murillo 2002). In these three clusters, the universe of enterprises includes 130–150 firms, which means that the sample of selected enterprises in any of them is relevant. As mentioned earlier in this section, the information presented does not focus specifically on performance indicators, but rather on those factors identified by the main approaches to SME cluster development (i.e., ‘external economies’ and ‘joint actions’, ‘social cohesion’ and ‘selfrealisation’, ‘national’ and ‘local SME policies’). 3.2 The district of Forlì, Italy This cluster is specialised in furniture production and encompasses 240 enterprises for a total of about 3,200 employees (Bertini, 1997; Bardi, 2000). Half of the firms focus on core production (i.e., upholstery), while the rest focus on the supply of parts, such as wooden structures, metal nets, gum bumpers and leather. It is a well recognised and competitive cluster, one of the main upholstery production areas in Italy, and has a mature export market, especially to France and Germany. The district’s overall output is about ∈360 million, 40–60 percent of which is usually exported (Bertini, 1997; Bardi, 2000; Camera di Commercio, 2005). Following the proposed stage approach, the history of this cluster is traced. Before the First World War, which is also before industrialisation took place in the area, no furniture production was evident. Just a decade after the First World War, however, one thousand workshops were operating and employing about 2,000 workers (Fauri, 1996: 5). In the 1920s and the 1930s, plenty of small workshops started up in Forlì and its surrounding small towns (e.g., Cesena, Meldola, Castrocaro). Production was organised through craft enterprises that each employed an average of two workers and sold their furniture in local markets. It was in the 1950s that a big jump was made, from basic craft agglomeration to industrial production, when a local entrepreneur set up a modern factory with around 60 employees. Production was thus industrialised, which catalysed further developments by pushing other entrepreneurs to make similar technology investments and upgrading the production capacity of the cluster (Fauri, 1996; Bertini, 1997). After about 10–15 years, the cluster entered the next development stage. This evolution occurred in the early 1970s, when the Fordist system crisis hit local medium- and large-sized firms, leading to a closing down or restructuring of many operations. Many small firms were created upon the initiative of specialised workers leaving the larger firms. Their
110
SME Cluster Development
former employers often helped by lending or selling them good quality machinery in order to externalise production and reduce costs (Fauri, 1996; Bertini, 1997). A second event, the participation of local artisans at the international furniture trade fair in Milan in the early 1970s, pushed the cluster to upgrade from craft-type agglomeration led by a few large firms, to a true industrial district. Producers got to know French traders, who appreciated their skills, visited the cluster and decided to open up a trading channel for upholstery production (Bertini, 1997; Bardi, 2000). Within a few years, a dynamic industrial district was operating and exporting tens of millions of euros to the European market (Fauri, 1996; Belussi and Bertini, 1998; Bardi, 2000). In recent years the cluster has been under pressure from growing competition from the southern Italian cluster in Matera (led by the large company Natuzzi), but also from production in eastern Europe and China. At the end of the 1990s, a wave of public and private efforts was made to overcome this problem. Firms started targeting higher-income consumer segments in order to differentiate themselves from the ‘low-road’ type of competition implemented in less-developed competing regions (Bertini, 1997; CSIL, 1997; Belussi and Bertini, 1998). Some interesting outcomes can be observed (Camera di Commercio, 2004). Figure 6.1 indicates the growth rates of this cluster over the ten years 1994–2004. Coming from the expansion linked to the exchange rate devaluation in 1992, the cluster underwent a severe crisis in 1998–99, which led to negative growth rates (⫺17 percent). Soon after, the local
20 15 10 5 0 ⫺5 ⫺10
Figure 6.1:
Growth rates in the furniture industry in Forlì
Source: Camera di Commercio, 2005: 69.
2004
2003
2002
2001
2000
1999
1998
1997
1996
1995
⫺20
1994
⫺15
Empirical Explorations
111
industry started a recovery, which led to a positive growth rate in 2002. In 2004, the volume of physical output increased by 3 percent, with a further 5.5 percent increase in orders from abroad, employment increased by 4.1 percent, despite a curious 4 percent decline in employers and managers and a still important export market, where 40 percent of sales were accounted for (Camera di Commercio, 2005: 62–7). These positive growth rates, however, are not sufficiently steady and do not represent a permanent solution to the competitive problems of this cluster as indicated by the slowdown that occurred in 2003–04. The following qualitative analysis may help to clarify some of the main bottlenecks faced by the district. Among the forces that pushed the cluster along a trajectory of growth and development, ‘collective efficiency’ was one of the most important. Right from the early stages, business associations supported the collective actions of craftsmen. In the 1960s and the 1970s, most firms were already members of craft and industry associations (e.g., CNA, Confartigianato) and almost 100 percent are so today (our survey). These associations perform a double function. They are political organisations that discuss relevant issues with public authorities and obtain their commitment for public support (Capecchi, 1990; Putnam, 1993; our survey). They also provide important services to SMEs (e.g., accountancy and tax advice, information about environmental protection, etc.) (Brusco, 1990; Bellini, 2005). Two promotional consortia have also been created by groups of leading local enterprises (in one case, a rather old consortium which is renewing its efforts after a long ‘lethargy’) with the objective of exploring new international markets such as the United States and China. This interesting story has also a downside, however. Consensus about how best to support the district is not easily reached, and internal dissatisfaction is rife. Many entrepreneurs complain about the scarce support they actually receive from their associations, limited to the delivery of few basic services at a market price. For these reasons, local entrepreneurs seem nowadays not to rely too much upon associations, but mostly upon themselves and therefore they work alone. In terms of ‘external economies’, Forlì benefited for years from the abundance of ‘clients’ and ‘labour’. International traders were responsible for the boom in the 1970s (Fauri, 1996; Bertini, 1997). At present, this flow is shrinking, since traders are often looking for cheaper bargains elsewhere. Between the 1920s and the 1960s, ‘labour’ was abundant due to the intense migration from the countryside to the town (Fauri, 1996). Throughout the next period to the early 1980s, it was the main driver
112
SME Cluster Development
behind the expansion of furniture production. Although ‘labour’ has more recently come from heavy immigration from northern Africa, it is not the skilled labour actually needed. In fact, specialised workers search for better paid jobs in other sectors (e.g., metal mechanics). Other ‘external economies’ were previously weaker than they are today. The flows of ‘innovation’ and ‘information’ were scarce within the craft agglomeration in the 1930s and the 1940s, but also in the phase of large firm-led industrialisation and in the 1970s and the 1980s’ IDs, because local production was characterised by middle to low quality standards and cheap prices. Global competition today does not leave much scope for continuing this kind of production, which is why local producers show an increasing sensitivity to these externalities and the capacity to embrace them. The different weights given today to these externalities represent the changing priorities of the cluster. Due to the increased cost of labour, firms can no longer compete on price alone, and therefore have to discover new competitive advantages through knowledge and innovation (Pyke and Sengenberger, 1990; Cooke, 1996; Kaplinsky and Readman, 2001). This is why local producers are very sensitive to the ‘flow of information’ and ‘innovation’ that helps them to upgrade production (e.g., in the use of new materials and machinery) and marketing practices (e.g., selling through web pages). The second relevant approach to SME clustering emphasises the ‘social factors’ of development: ‘self-realisation’ and ‘social cohesion’ (see Chapters 4 and 5). With respect to the first, a changing trajectory can be observed. Table 6.2 illustrates an extremely high level of firm creation post-First World War, which then decreased in the 1950s and the 1960s (followed by further expansion in the 1970s and the 1980s). While most workshops continued their operations and involved new family members in their daily activity, many other people searched for employment in large firms (Bertini, 1997). In this way, individuals’ spirit of initiative focused Table 6.2: of Forlì
Number of firms and workers in wood and furniture in the province
No. of enterprises No. of employees
1911
1927
1937
1951
1961
1971
1999
Few Few
955 2,016
1,007 1,886
1,140 2,632
1,297 5,694
762 4,015
688 5,800
Note: The province of Forlì includes a wider territory than just the cluster of Forlì; this explains why the number of employees and firms here differ from the level presented at the beginning of 3.2 for the cluster. Source: ISTAT, III, IV, V Censimento generale dell’industria e del commercio, in Fauri (1996).
Empirical Explorations
113
on learning new production practices and earning a better income, both of which became useful means of production in the following phase of industrial district development. In fact, as soon as the large firms entered a phase of crisis (late 1960s– early 1970s), many workers decided to leave and set up their own workshops (ibid.). The interest to be one’s own master, take every important decision and maintain closer relations with workers, often family members, were strong stimuli for people to create their own enterprises. This is supported by the fact that 93 percent of the firms existing today and that started up in the late 1970s to the early 1980s, were spin-offs (see Table 6.4). This aspect is changing nowadays. New perspectives have opened up to young people, who are less willing to undertake entrepreneurial initiatives in this sector or take over from their parents at the head of the firm. This is creating problems within the cluster, since the entrepreneurial push has weakened. For example, the spin-offs generated by workers (previously employed in current enterprises) are few and the rate of firm creation is low (see Table 6.4). On the basis of the information collected through this survey, it is realised that the above-mentioned new trend (see Table 6.4) is not Table 6.3:
Producers’ perceptions of external economies in Forlì, 2002 Forlì, Italy
Good clients inflow Good information flow Good innovation diffusion Abundance of skilled workers
3rd 1st 1st 4th
Source: Author’s survey on the basis of entrepreneurs’ perceptions. Notes: 1st is the external economy producers perceive as the strongest, 4th is the external economy that producers perceive as the weakest.
Table 6.4:
Enterprises created by workers in Forlì Forlì, Italy
% of current enterprises created over time by workers from firms within the cluster
93.7
% of enterprises created by workers from current enterprises
25
Source: Author’s survey, November 2002–January, 2003.
114
SME Cluster Development Table 6.5:
‘Innovative’ producers during 2000–02 (%) Forlì, Italy
Product innovation Process innovation Market innovation Average innovation rate
78.1 50 62.5 63.6
Source: Author’s survey, November 2002–January 2003.
transforming Forlì into a static cluster; it just represents a new phase for this mature cluster. The present saturation of the European market and higher barriers to entry (e.g., much heavier start-up costs than those of 30 years ago, when small workshops were created in the backyards of family houses with little or no machinery), also explain this decreasing dynamism. That is why, nowadays, firms attempt to transform from traditional to innovative enterprises that avoid price competition by investing in sophisticated technologies in order to offer unique and well-remunerated products to high-income consumers in global markets (Pyke and Sengenberger, 1990; Cooke, 1996; Kaplinsky and Readman, 2001). The new entrepreneurial tendency to create joint-ownership firms among two to four partners (instead of traditional sole-trader or family businesses), represents another aspect of this competitive strategy of pulling together specialised knowledge and skills.47 With respect to ‘social cohesion’, mixed perceptions come about. For instance, producer participation in social networks (e.g., hobbies, religious groups), which are supposed to strengthen ‘social cohesion’ within the cluster, is nowadays quite low. Only 25 percent participate and this is mainly in sporting activities. However, there are other indicators of the presence of social ‘glue’ within the cluster. One of these is the duration of work relationship between producers and their suppliers, clients and workers (see Table 6.6). The duration and selectivity in work relationships is significant, which confirms the value that these entrepreneurs attribute to maintaining close relationships with people from the locality (e.g., entrepreneurs with workers; suppliers with producers, etc.). These relationships, based on mutual understanding and trust, have the general effect of lowering transaction costs and of increasing the competitiveness of local production (Lorenz, 1992, Dei Ottati, 1994; De Propris, 2005). The third type of factor refers to ‘policy inducement’ of development. Forlì benefits from the long-running national legislative support for
Empirical Explorations Table 6.6:
115
Duration of work relationships with suppliers, clients and workers Suppliers
Number (average) Years (mode) Years (average)
11 10 9
Workers 19 8 6.9
Clients 171 7 6.2
Source: Author’s survey, November 2002–January 2003.
SMEs, starting from the early 1930s, at the time of the craft agglomeration, with the instigation of the statute of craft enterprises (Arrighetti and Serravalli, 1997; Bianchi and Parrilli, 2002a). This support continued in the 1950s with new laws for the promotion of SMEs, later strengthened through the 1965 Sabatini Law for financing technological upgrading of SMEs through leasing mechanisms supported by a state guarantee fund. In the 1970s, the Ossola Law was approved to provide guarantees for exporting firms in their internationalisation process. In the 1980s and the 1990s, laws on firm creation (Law 44/1986 and 95/1995), innovation (Law 81/1991) and regional planning and investment (Law 341/1996 on ‘territorial pacts’) supported the move from traditional industrial districts to new competitive IDs (Bianchi, 1992; Bertini, 1998; Bianchi and Parrilli, 2002a). These national initiatives have been complemented by significant support at the local level through the above-mentioned instruments as well as through promoting the provision of business development services from the mid 1970s through to the early 1980s, and by way of the creation of several ‘industrial areas’ around the town in the past 10 years. In fact, the cluster of Forlì is localised in Emilia-Romagna, a region that from the mid 1970s supported the enterprise system through a set of regional development agencies that deal with regional development (ERVET), innovation (ASTER), information (DEMOCENTER), textile development (CITER), etc. (Bianchi and Parrilli, 2002a). Moreover, the survey shows that about ten industrial areas can now be found in a diameter of about 15 kilometres around Forlì. The smallest areas include between 50 and 100 firms operating in several sectors, while the largest includes more than 300 firms. In synthesis, the history of Forlì shows two relevant aspects. First, the cluster passed through different development stages that permitted a gradual upgrading from lower to higher levels of competitiveness. Second, this analysis shows the important role played by the factors taken into account in the ‘eclectic approach’ in promoting the shift from one stage to the next. ‘National’ and ‘local policies’ have been at work
116
SME Cluster Development
throughout this time, and have been supporting the growing role of SMEs in Forlì and across the country. ‘Joint actions’ and ‘external economies’ have been evident too, promoting the economies of agglomeration that help to make these SMEs more competitive. Social strengths (i.e., entrepreneurship through ‘self-realisation’ and cooperation through ‘social cohesion’) have also promoted the passage from one stage to the next, although in different ways over time (e.g., creating new SMEs in the 1930s and then again in the 1970s; learning know-how in the 1950s–1960s and developing innovation capabilities in the 1990s). The recent growth in competition from other areas in Italy and elsewhere spurred the need to discuss within the cluster what development strategies were needed. Today, apart from the problem of entrepreneurial turnover in some firms, significant changes are occurring in the cluster, with special reference to the entrepreneurial attitude towards business. In fact, together with the traditional membership of business associations (although criticised), which characterises one of the main strengths of the industrial districts, many firms are also investing in new technology, in both production and marketing (for instance, computerised-numericalcontrol (CNC) machinery, web pages). Producers are keener to reap the ‘external economies’ linked to innovation and information spillovers. New support policies have also been set up at both the ‘governmental’ and ‘local level’. In this sense, Forlì does not look like a static cluster, but one that is restructuring to raise its competitiveness and to remain a leading upholstery exporter in global markets. This analysis may represent a point of reference for the potentially most dynamic clusters among the survival’s category in developing countries. In fact, it portrays the story of a cluster with humble artisan origins, which progressed to be a more industrial style SME cluster, became in more recent decades a typical industrial district, and today is struggling to be even more innovative and competitive. This is a growth trajectory that could be taken as a possible benchmark by many SME clusters in developing countries. We can now turn to see whether basic conditions exist in less developed countries to allow a group of dynamic ‘survival clusters’ to advance along a similar growth trajectory. 3.3 The ‘survival clusters’ of Sarchí, Costa Rica, and Masaya, Nicaragua Both of these clusters are quite new, since furniture production was only undertaken on a significant scale in these areas in the 1980s. The craft agglomeration that can be seen today in these clusters is very similar to the situation of Forlì in its beginning, half a century ago.
Empirical Explorations
117
Sarchí is a small town (25,000 inhabitants) in the central mountains of Costa Rica, 50 kilometres from the capital, San José, and 30 kilometres from Alajuela and Heredia, the country’s other main towns. It is an old cluster with traditions in oxcart production. It was only in the 1980s that certain artisans started focusing on furniture due to the fall in demand for oxcarts, which became more of a decorative art, while demand for furniture grew from domestic consumers (Pérez-Sáinz, 1994). Little by little, the number of micro and small enterprises reached 120, producing all sorts of furniture, mainly for the national market, with very limited export sales (Pérez-Sáinz, 1994; Murillo, 2002). Masaya is a small Nicaraguan town of about 100,000 people. Typically, the people work in craft activities linked to the production of shoes, clothes, hammocks and furniture. This town is situated 30 kilometres from the capital, Managua. Just like Sarchí, this is a new cluster, since few workshops produced furniture for local consumers before the 1980s, when many more workshops started up and benefited from the support of the CONAPI, the government body which controlled the imports of materials and the exports of final products (Parrilli, 1999a; see also Chapter 4). With the election of liberal governments in the 1990s, CONAPI was replaced by a new governmental institution, the Governmental Programme for the Support of Micro Enterprises (PAMIC), which offered a more marketfriendly support to small firms. The number of SMEs rose to the current 120 micro and small firms (BCN, 1996). Nowadays, the cluster produces mainly for the local market, although output bound for the national and international markets has grown in recent years (e.g., towards Costa Rica, El Salvador, and the United States). Under the ‘collective efficiency’ approach, a few elements can be observed about these two clusters. In Sarchí, two ‘joint actions’ are being implemented, the first of which concerns a long-standing sales cooperative that is enjoying growing success (Pérez-Sáinz, 1994). The other refers to a new municipal production committee, which in cooperation with the local Health Ministry department, studies new ways to promote local production (e.g., through the attraction of tourists to the local rainforest reserve) and the relocation of enterprises to industrial areas to prevent pollution problems for the population (i.e., dust and noise; see Murillo, 2002). Informal exchanges are very common among producers: 70 percent are used to lending each other machinery, raw materials and workers in an atmosphere of mutual support. These relationships ease daily operations and help reduce overall transaction costs. In Masaya, the situation is more difficult with respect to cooperation. Only 30 percent of entrepreneurs experience any informal type of
118
SME Cluster Development
cooperation, and no business associations exist. This can be explained by the recent history of Nicaragua. The military conflicts from the 1970s up to the early 1990s, motivated certain distrust among people (Mendoza, 1990; Barahona et al., 1998; Parrilli, 1998b; 2002a). Moreover, the cooperative system developed by the socialist government in the 1980s fell into crisis with market liberalisation in the early 1990s. The producers concerned started a struggle between themselves for control over the means of production, some with more success than others. This whole story left a bad feeling among many producers in the country and makes it harder now to develop ‘joint actions’ (Parrilli, 1998b; 1999a; 1999b). Nonetheless, something seems to be changing. Growing efforts to organise public meetings among the economic, social and political agents can be noticed within the municipality (Henriquez, 2002). These efforts have led to the creation of a municipal production committee that involves representatives of the SMEs as well as local NGOs, international agencies (i.e., UNIDO) and the local authorities (i.e., the mayor). This committee is supposed to support producers’ requests for more adequate services for local production (e.g., market promotion). Considering ‘external economies’, Sarchí seems to be benefiting from a variety of spillovers, in terms of new ‘customers’ to the area (which led producers to start a recent wave of investment in retails shops), an abundance of ‘labour’ and a flow of ‘innovation’ and ‘information’ (visible, e.g., in the widespread use of newer materials and new techniques, such as certified wooden boards, or more standardised inputs, such as melamine boards). Our analysis, represented through a Likert scale (from 5 to 1) indicates that the ‘clients’ inflow’ and the abundance of ‘labour’ are 5 4 3 2 1 0 Clients inflow
Skilled Flow of Diffusion of workers information innovation Sarchí
Others
Masaya
Figure 6.2: Likert scale (5–1) of producers’ perceptions on the sources of external economies Source: Author’s survey, August 2002.
Empirical Explorations
119
rated the highest (respectively 5 and 4), which means that producers are very conscious of the opportunity to reap these external advantages from being located close to several other similar enterprises. Small manufacturers value relatively high also the free flow of ‘innovation’ and ‘information’ (respectively 3 and 2, especially in Sarchí). In general, putting to one side a standard performance analysis (based on sales, exports, income, etc.), Sarchí seems like a dynamic cluster in which local producers try to capture all the possible spillovers from each others’ activities. In Masaya, ‘labour’ is abundant. This is due to national recognition of this locality as a cluster for furniture production (INCAE, 1997; Barahona et al., 1998). The manufacturing activity has the capacity to create growing employment opportunities. The ‘flow of clients’ has also been increasing in the past five to seven years, along with increasing stability in the country, leading to greater commercial opportunities for producers.48 In contrast, entrepreneurs here consider the ‘flow of innovation’ and ‘information’ to be quite poor and often complain about the local tendency to imitate each other quickly, which pushes them to be more secretive. Our interpretation suggests that innovations in these clusters are not qualitatively outstanding. All furniture makers adopt similar techniques, materials, machinery and often the same clients. This situation seems to explain why it is increasingly less relevant to embrace (incremental) innovations here. In contrast, the recentness of this cluster explains the attention that is paid to ‘labour’ and to ‘clients/traders’ (a similar trend was observed in Forlì in its early stages of growth). In social terms, the tendency to form new enterprises can be considered as an interesting indicator of the local spirit of ‘self-realisation’. In a much greater sense than in Forlì, SMEs in both Sarchí and Masaya have spun-off a large number of micro and small firms over the past few years. These aspects are consistent with a strong local socio-economic dynamism, which might be based upon two main reasons: on the one hand, the need to create new income opportunities in rather poor countries (it would be more the case of Masaya than Sarchí) (Tokman, 1992; Aranguren et al., 2005); on the other hand, the desire workers manifest to set up their own business, manage their own daily activity independently and succeed as entrepreneurs (Brusco, 1982; Hjalager, 1993; Aranguren et al., 2005; see also Table 6.7). The relative youth of these clusters promotes firm creation and spinoffs, because know-how is accumulating and a large number of people (e.g., workers, young people) want to try out their ability to succeed in the main local specialty production. They feel that there is a large market out there and that clients constantly visit the locality. Thus, they worry
120
SME Cluster Development
Table 6.7:
Enterprises created by workers in Sarchí and Masaya Sarchí, Costa Rica
Masaya, Nicaragua
% of current enterprises created over time by workers from firms within the cluster
75.7
80
% of enterprises created by workers from current enterprises
75.7
53.3
Source: Author’s survey, August 2002.
little about the costs of setting up an enterprise, since they will do it at the lowest cost by buying some basic machinery and/or contracting out the stages of production that are costly to perform in-house. In the case of Sarchí, this strong tendency to firm creation is also rooted in the more stable political and economic environment of Costa Rica, which makes it easier to understand and calculate the risk and return on investment. In both Sarchí and Masaya, this tendency is also linked to their recognised positions as the leading furniture clusters in their countries. Innovation is another important indicator of the push to ‘self-realisation’ in clusters (see Chapter 4). Of course, innovation here does not refer to original inventions, but rather to the adoption and/or imitation of standard international practices related to this specific sector (Parrilli, 2002a; Romjin, 2002). Figure 6.3 indicates the innovation efforts undertaken by producers (e.g., investments in product, process and market innovations) independently from the cost of these efforts, which often depends on the availability and cost of financial resources in the local market. In Sarchí there are signs of firms undertaking innovation through industrial specialisation (e.g., specific products), technology investments (e.g., new techniques and materials) and market orientation (e.g., environmental certificates for production inputs). In Masaya, 49 percent of the producers innovate. In this case, it mainly means adopting better production and market practices. It is very common for producers to imitate newer products observed in the market, by adding certain personal aspects that introduce a form of product differentiation. Of course, this differentiation is neither obvious to the ordinary consumer nor does it permit the producer to avoid price competition (Romjin, 2002). In marketing, innovation refers mainly to the search for new clients, possibly foreign traders, in search of which producers increasingly participate in international fairs. A higher international interest in local production is evident, as indicated by the growing number of exporters in the past five to six years.49
Empirical Explorations
121
Product innovation 80 60 40 20 Average innovation
Process innovation
0
Market innovation Sarchí Figure 6.3:
Masaya
Innovative producers over two years (%)
Source: Author’s survey, August 2002.
Table 6.8:
Duration of work relationships with suppliers, clients and workers Suppliers
Workers
Clients
Number (average)
4
8
42
Sarchí
Years (mode) Years (average) Number (average)
4 4 6.3
3.9 3.9 8
4 3.7 39
Masaya
Years (mode) Years (average)
3 3.4
3.5 3.5
2.8 3.1
Source: Author’s survey, August 2002.
The simplicity of the innovative effort in Masaya and in Sarchí depends mostly on their severely restrained access to investment funds. Annual interest rates in Central America are still over 20 percent, and high guarantees are required, which reduce sharply the possibility to apply for long-term loans for investment (BCN, 2002; BCCR, 2002). In terms of ‘social cohesion’, there are more extra-economic spaces (e.g., sport, religious groups, ecology, schools) than in Forlì. A significant 55 percent of producers participate in these activities, which spur local ‘social cohesion’ (less in Masaya). This indication supports the hypothesis that these clusters are well endowed with elements that can spur ‘joint actions’ and collective planning for development. As Table 6.8 shows, the duration of work relationships within these clusters is relevant too.
122
SME Cluster Development
In this survey, Sarchí and Masaya exhibit shorter-term, but still significant, relationships than Forlì (respectively 4 years and 3–3.5 years on average). These lower numbers are due to the age of these clusters and of most enterprises therein (13 years in Masaya and 16 in Sarchí against 21 in Forlì). In the case of Masaya it is also due to the civil conflicts of the 1970s and the 1980s and the tensions following liberalisation in the 1990s (Mendoza, 1990; Parrilli, 1998b; 1999b). In terms of ‘policy inducement’, public policy has not been supportive until now. In the case of Costa Rica, the first national policy for SME promotion was only approved in 2002 (Gazeta Oficial, 2002). Since then, the related law has been going through a process of institutionalisation and implementation, with the creation of the appropriate guarantee funds, technological funds and other policy instruments.50 Up to now, all industrial and development policies have been directed to sector development (e.g., in high-tech sectors) and export, independent of the existence of clusters and SMEs (Aguilar et al., 1998). At the local level, the most significant impact of public policy in Sarchí is in terms of higher costs of doing business. The local health ministry office requires an improvement in work conditions to control emissions from workshops, while the municipal authority worries about the payment of production licenses (Murillo, 2002). These pressures push the producers into relocating to industrial areas outside of the town. Individual investments are taking place, while municipal and state support has not yet materialised. Talks held within the local production committee represent a step towards the institutional consolidation of furniture production in Sarchí. In Nicaragua, although a law for SMEs has been under discussion since the mid 1990s, the interested parties (i.e., four SME associations and international agencies such as UNIDO and the German GTZ) have not been able to convince Congress to approve the law (Bouchier, 1998; Parrilli, 2003; see also Chapter 4). In recent years, a growing number of international agencies and local NGOs have started to support SME and cluster development. However, they do not share a uniform approach and common development plan, hence the weak results (see Chapter 4). At the local level, many NGOs and governmental organisations are working with groups of firms in Masaya. These institutions tend to work in a fragmented way, each supporting a group of producers through the delivery of business development services (e.g., training, technical assistance and credit). They do not though seem to worry about pulling together local entrepreneurs in sector and local collective initiatives (Parrilli, 1999a).51
Empirical Explorations
123
There are signals nevertheless of a growing consciousness of this need. In fact, a few years ago the local authority promoted the restructuring of an old castle into a market for all types of craftworks, which has become a point of reference for tourists. More recently, the mayor and UNIDO have been promoting the creation of a local production committee composed of producers’ representatives, international agencies (i.e., UNIDO), governmental institutions (e.g., INPYME, programme for the delivery of business development services – PROSEDE) and international and local NGOs (e.g., Nitlapan, and Tierres des Hommes). In synthesis, this analysis leaves aside standard performance indicators that would show Sarchí and Masaya to be two typical ‘survival clusters’ with little or no capacity to grow. In contrast, it shows elements of interesting dynamism in these localities. The Sarchí cluster cannot be compared to the current day cluster of Forlì, since its enterprises are all micro and small enterprises producing mainly for the small national market. It may, however, be compared to Forlì in its first stage of development and might be even more dynamic than that cluster. In fact, Sarchí exhibits several interesting characteristics, such as the new wave of investment in retail shops (which occurred in Forlì in the 1980s after the cluster had already opened up the export market) and the various attempts at innovation through the usage of materials and techniques more in line with international standards (e.g., environmental certificates, use of plantations). Government policy also shows significant changes towards a more supportive approach at the national level (e.g., the new law for SMEs) as well as the recent local producers’ committee set up to solve critical issues for local production (Murillo, 2002). The situation in Masaya is less clear. Some ‘joint actions’ are taking place through the support of international agencies and local NGOs. Labour abundance and client inflow constitute significant ‘external economies’. The entrepreneurial spirit has been growing, while ‘social cohesion’ still suffers from the long civil conflict of the 1970s and the 1980s (Mendoza, 1990; Parrilli, 1999a). ‘Governmental policy’ is more significant at the local level, but rather poor at the national level (Bouchier, 1998; MIFIC, 2001b; Parrilli, 2003). On the whole, Masaya also shows new interesting elements of dynamism, which confirm the hypothesis that this ‘survival cluster’ is developing and that ‘clustering is beautiful’ even for this less competitive kind of cluster. Simultaneously, Masaya manifests weaknesses that need to be addressed to create a purposeful environment to push this cluster to higher development levels. This may find positive results through the adoption of an ‘eclectic policy approach’ that pulls together different types of complementary public support actions.
124
SME Cluster Development
4. Concluding remarks for policy-making Two types of conclusions can be drawn from this analysis. This study does not try to ignore the substantial competitive differences that appear among SME clusters. Forlì represents a successful cluster, while Sarchí and Masaya do not achieve significantly comparable success in terms of competitiveness, production and exports. Nevertheless, this aspect is not the objective of this work, which rather emphasises the dynamic aspects of growth in these different contexts. Two main conclusions can be extracted from the analysis. The first is that the selected ‘survival clusters’ also show several elements of dynamism and development. Competitive differences might not represent definitive gaps but rather different stages of development that the clusters may be passing through in their growth trajectory. The study of these three SME clusters clarifies that development is a complex process that implies upgrading through ‘stages’. Each is passing through a specific stage. Sarchí and Masaya represent the craft type of agglomeration attempting to shift to industrial production, while Forlì is transforming from a traditional to a new competitive industrial district. Other passages between stages lay ahead, which need to be identified in each specific case in order to better plan the policy support and feasible steps that can be expected from producers and the whole cluster in the short term. The common negative perception surrounding the whole category of ‘survival clusters’ may be re-addressed to allow developing countries to reap more benefits from the presence of this kind of pre-competitive SME agglomeration, at least in the case of the most dynamic cases among the ‘survival cluster’ category (i.e., embryonic clusters). From an industrial policy perspective, this position stresses that more efforts can be made to identify feasible development steps for these kinds of clusters (Humphrey, 1995; Knorringa, 2002; 2005; Guerrieri and Pietrobelli, 2004). This is more likely to motivate all local forces to join together (e.g., entrepreneurs, private and public agencies) and create higher economies of agglomeration, which help them competitively enter global markets. The second conclusion addresses the determinants of growth at the cluster level. This empirical exploration seems to be in line with the hypothesis that a thorough approach to development includes several types of factors extracted from the main bodies of literature on clustering: ‘collective efficiency’, ‘social embeddedness’ and ‘policy inducement’. This approach can explain the diversified historic development of SME clusters and their present competitiveness. It can also do justice to the complexity of the development process, which takes place on the
Empirical Explorations
125
basis of the dynamic and reciprocal interrelations among these types of factors and their cumulative effect upon local production systems. When only a few of these factors work simultaneously, the outcome is more likely to be less than optimal, which can be seen in ‘survival clusters’. When all of these determinants work simultaneously, a better outcome is more likely, which is what happened with the successful experience of Italian industrial districts (Parrilli, 2004b; see Chapter 5).
7 Conclusions
This volume analyses specific issues related to two topics: SME development and SME cluster development in the case of developing countries, in particular Latin America. These two parts are certainly interconnected because they focus on the development of specific local social and economic agents: the SMEs. However, the two parts are also independent from one another, due to their own peculiarities. In fact, Part I (SME development) focuses on the fracture that is arising within several Latin American national production systems between a competitive modern sector and a traditional small-scale enterprise sector. This fracture is based upon the inadequate working of the market, which raises market failures in finance, technology and market access. These badly hit micro and small traditional firms, while not affecting the large national modern companies, which benefit from a wider access to international technology and services. Overall, this fracture is creating problems in the whole development process of these countries. A solution is, therefore, investigated, identified and proposed, especially in the final section of Chapter 3. This solution is consistent with the discussion held in Part II. In fact, the systemic policy approach does not neglect the important results achieved through neoliberal policies to stabilise the economy in several developing countries. But, at the same time, it shows that those policies have not been sufficient to push the national production system to grow in a balanced and cohesive way. In contrast, tendencies to split the enterprise system have been growing over the past few years, which now call for a more proactive role of the state and of international organisations in the market in order to close that fracture and ease the development process of SMEs and SME clusters. Part II focuses on the development of the less dynamic kinds of SME clusters, known as ‘survival clusters’. These are very common local 126
Conclusions
127
production systems in developing countries and are often left on their own because it is considered that they cannot have significant potential to develop and grow. It is important to discuss their conditions and their potential, which in some cases can be higher and in other cases lower (see footnote 1 in Chapter 1), in order to find a way to contribute to their upgrading process and the development of their own countries. The overall theme of cluster policies is well linked to Part I of this volume, since they represent one of the most outstanding examples of what has been called ‘linkage policies’. These aim at closing the fracture that has been arising within the national production systems of many developing countries. The coherence between Part I and Part II is also based upon the utilisation of an integrated approach to the development of national production systems, which turns into what is called the ‘eclectic approach’ in the context of cluster development.
1. SME development In Part I, there is an analysis of SMEs in Latin America, which can be of some interest also in other developing countries. In this context, we identify a special feature that is affecting SME development nowadays, which is the growing fracture within local, national and regional production systems between a tiny modern sector and a large traditional sector. Large national and foreign companies and a few small and medium enterprises compose the first sector. These are capable to work with advanced technology and skilled labour to produce innovative goods and services for the national high-income segment of the population as well as for international markets. The micro enterprises as well as the majority of small and medium firms represent the second sector of firms. Due to market failures and accessibility problems to finance, technology and markets, these enterprises work with traditional technology and unskilled labour to produce non-standardised and even low-quality goods and services for the local and national market (Bianchi et al., 2000; Bianchi and Parrilli, 2002a; 2002b; Parrilli, 2004a; see chapters 2 and 3). This ex-post kind of fracture, which is triggered by the working of the market, does create a problem to the specific local and national production systems; this represents a problem for the whole Latin American region and, more in general, for developing countries. When a large majority of traditional enterprises are separated from a minority of modern firms and progress too slowly, the national development process is likely to end up being self-constrained. In this context, the good performance in managing the national (macro) economy is likely to lose significance. In fact, the
128
SME Cluster Development
GDP growth does not show possible skewed income distribution, which is often the case in developing countries; it may represent the success achieved by the modern sector and its dynamic capacity to respond to the challenges imposed by globalisation, but it will not explain the (often negative) result achieved by the traditional sector in the globalised market. In terms of human and social development (in particular, in terms of skills and capacities of the economically active population), the growing opportunities offered by globalisation (e.g., increasing flow of capital, goods and services, human resources, increased markets) will be exploited by a tiny part of the population, since just a tiny part of it will be earning enough to support its own development plans. For instance, as Lewis identified (1973), when the economy is divided between modern and traditional sectors, the population employed in the traditional sector earns a subsistence income, which only allows them to pay for the basic education of their children. This weakness is likely to give them small opportunities to improve their socio-economic conditions. The overall social mobility will be constrained. This will reinforce the tendency to split the Latin American economy in an (ex-post) dual environment and weaken the future development prospects of the country. This tendency can be defined as an ex-post type of dualism, which is different from the ex-ante type of dualism proposed by the structuralism and the dependence theory in the 1960s and the 1970s (Cardoso and Faletto, 1972; see section 2 in Chapter 3). In fact, in this analysis it is the market that produces this kind of division, while in the former approach it was society that pre-defined the market opportunities and the related economic outcomes (Frank, 1970; see section 2 in Chapter 3). The analysis of selected countries (Argentina, Chile and Nicaragua) highlights the process of growing fragmentation within their economies. This outcome tends to be connected, among other aspects, to the weakening of the manufacturing sector that is due to the reduced importance attributed to industrial development policies (Bianchi, 2002a; see Chapter 3). Argentina was much stronger 20 years ago, when the manufacturing sector was at its highest peak (30 percent of GDP), before falling up to the present terms (16 percent of GDP) (Bianchi, 2002a). The retreat from the application of industrial development policies is likely to be one of the factors that explain the declining weight of industry within the economic structure of the country. Together with other variables, such as the quick market liberalisation, the unconditioned privatisation, the lack of industrial development policies pushed Argentina to lose the capacity to cope with significant exogenous pressures coming from the international market (e.g., the growing expansion of China’ and southeast Asian countries’
Conclusions
129
exports into the Latin American markets). This is what occurred at the beginning of 2000, when a strong financial crisis impacted the country (Bianchi, 2002a; see Chapter 3). A similar argument can be proposed about Chile that, since 1998, faces a less promising scenario than in previous decades. The modern enterprise sector has been increasing its sales despite its employment reduction, while the firms in the traditional sector have been reducing their overall sales despite the increase in employment (INE-Gobierno-BancoEstado, 2001; Bianchi and Parrilli, 2002a; Parrilli, 2004a; see Chapter 2). A substitution process is taking place between capital and labour in modern firms; this process permits them to improve their productivity and insertion in the international markets, whereas micro and small traditional firms are not able to undertake this kind of upgrading (i.e., increase in productivity) and remain less and less competitive in the same national market (Parrilli, 2004a). Also in this case, the manufacturing sector has been shrinking from the highest 27 percent of GDP in 1972 to the present 16 percent, leaving the overall production system in a weaker position in front of unexpected exogenous shocks, which lead the system to destructuring even further in a sort of vicious circle. This is happening with the massive flow of imported goods from China and East Asia, which has also been facilitated by the unilateral opening of markets that Chile adopted from the 1970s onwards (Stumpo, 2000; Bianchi and Parrilli, 2002a). The case of Chile is studied in detail in Chapter 2. In this chapter, evidence is presented about the type of policies that this country has been implementing over the past few decades. In particular, it is shown that from the early 1990s onwards the main government development agency – CORFO – created a set of support programmes for enterprise development and specifically for SMEs (see Table 2.2 in Chapter 2). Through these programmes, CORFO responds to the demand coming from different groups of interest within the market and, in this way, this organisation adopts a policy approach that is closer to international agreement on liberalisation and market-pulled economies (Williamson, 1990). Nevertheless, it was found that this action does not represent a systemic action. In fact, CORFO has helped only a fraction of enterprises within the whole national production system: about 15,000 from a total of 500,000 enterprises and a few hundred firms for each individual support programme. These data indicate that the possible trickle-down effects are very likely to be reduced and the impact of this policy approach tends to be limited, especially in the manufacturing sector (CORFO, 1999; Parrilli, 2004a; see Chapter 2).
130
SME Cluster Development
Other countries show a similar trend. It is the case of Nicaragua, which in the 1970s was one of the leading Central American economies, thanks to its exports of agro industrial products, such as coffee, cotton and meat, among others (Laenen, 1988; Reyes, 1993). In the past 25 years, due to internal conflicts and the lack of industrial development policies, among other reasons, Nicaragua has been losing ground in its own market and in the region. Also in this case, the manufacturing sector was reaching about 30 percent in the late 1970s (Laenen, 1988) and still 27 percent in the early 1990s (Reyes, 1993), but through the policies applied by the liberal governments in the past 10–15 years it shrunk up to the present 14–15 percent of GDP (BCN, 2002). Also in this case, we observe a country that is rich in terms of natural resources (such as sea and lakes, forestry, farmland, mining, among others), that cannot benefit significantly from them and cannot improve its conditions (GDP per capita has been stagnant in the past 20 years; BCN, 2002) and the competitiveness of its SMEs (see Table 4.3 in Chapter 4). Table 4.3 shows low levels of value added generated by SMEs in the mid 1990s. These levels are maintained up to now (Parrilli, 2002b), apart from some specific successful case – that is, dairy production in recent years (Artola and Parrilli, 2006). On the whole, Nicaragua copes with the same kind of problems that Argentina and Chile face today. The firms in the traditional sector are unable to improve their competitiveness levels and are marginalised within their own economy and the regional market, whereas a few large, medium and small enterprises, often linked to international groups (e.g., Nestlé in coffee production and Parmalat in dairy products) are capable to modernise, produce and export. An ex-post dual economy is formed, which reduces the possibility of the country to advance to higher development levels as a whole (see Chapter 3). The last few governments have been targeting policies to make the country as it was in the 1960s and the 1970s, when it was known as the Central American granary (i.e., specialised in agricultural production for all the region; see Parrilli, 1999b). This constitutes a strategic failure, which depends upon the lack of industrial development policies that help the country to diversify the economy and promote structural change (Parrilli, 1999a; 2002b; 2003; Peres and Stumpo, 2000; 2002). The case of Costa Rica seems to show the opposite trend with respect to the three former countries. Up to 1998, the growth trend of large and small firms was similar (Aguilar et al., 1998; Wong Chacon, 2002). Firms in the manufacturing sector were increasing in number and participation in GDP (Aguilar et al., 1998). Overall, Costa Rica is one of the countries
Conclusions
131
that suffered least through the impact of the world crisis in Latin America (see Chapter 3). Exogenous reasons matter, such as the significant support received for years from the US development agency (ibid.), but also endogenous capacities matter. Among these, the government strategic approach never dismissed the importance of an industrial development strategy oriented to promoting the role of both manufacturing and SMEs. Defining the overall priority (i.e., exporting higher value added productions to the international market), promoting the strengthening of the economic agents (e.g., the Chamber of Industry and Commerce, the few large national companies, the small and medium firms, the universities and other relevant public and private institutions), unifying their efforts and combining it with the transnational corporations eased this process of systemic growth (Aguilar et al., 1998; see Chapter 3). Also the approval of the first national law for SME promotion is another indication of the strategic approach taken by government in economic development (Gazeta Oficial, 2002). Relevant lessons can be learned and transferred from this to the above-mentioned countries and other developing countries. On the whole, the four cases presented in Chapter 2 and Chapter 3, show one of the new key problems that SMEs and their production systems are facing in these days of increasing globalisation. The final section of Chapter 3 singles out some policy lessons that can be learned from the European experience in general (and from some specific experiences, such as Italy’s in particular) and applied, with the appropriate adaptations to the varied contexts, cultures and idiosyncrasies of Latin America and other developing regions and countries. Within the proposed systemic approach “high-tech policies” are proposed and blended with “structural policies” and “linkage policies” (see Chapter 3). Through the first type of policies, the objective is to continue shifting outward the production technology frontier and the capacities of each national production system to generate innovations and high value added production, leaving aside the ‘low-road’ type of competition and development, which can hardly face the strong competition of Asian countries (Pyke and Sengenberger, 1990; Cooke, 1996; Kaplinsky and Readman, 2001; Guerrieri and Pietrobelli, 2004). The proposed ‘structural policies’ are targeted to help traditional SMEs to increase productivity levels and improve competitiveness in the open markets. This is one of the ways of making them able to continue producing and offsetting the competition of large firms from other areas of the world (e.g., China and East Asia). Upgrading productivity does not mean working close to the frontier of technology, but adopting international standard technologies and techniques (overcoming
132
SME Cluster Development
obsolete technologies). Very interesting programmes have been organised in order to create such capacities in national production systems in developing countries. The Mise-à-Niveau applied in Tunisia and Morocco with the support of UNIDO and the European Union, represents one of the most interesting cases of structural policies targeted to the traditional enterprise sector (Ministère de Economie de Tunisie, 2000; Di Tommaso et al., 2001). Other policies and programmes can help producing similar kinds of outcomes (e.g., financial lines for the restructuring of traditional firms; training of workers, etc.: Bianchi and Parrilli, 2002a; 2002b). These policies are likely to produce the above-mentioned aggregate result (the growth and development of the national production system) when a third kind of policy is applied. This refers to ‘linkage policies’, which try to pull together modern and traditional sectors and enterprises in order to connect them in a coordinated system, where a deep division and specialisation of labour spreads around the system and makes it more competitive. These policies and these outcomes (division and specialisation of labour) are also likely to distribute more homogeneously wealth and incomes generated through the economic activity (Bianchi et al., 2000; Bianchi and Parrilli, 2002b; Parrilli, 2004a). This systemic approach does not pretend to be the panacea to the abovementioned problem of a fracture within the national production system in developing countries. Nonetheless, it could be one of the possible strategies that could be applied (after appropriate adaptations to the local idiosyncrasies) in an exploratory way to SMEs in Latin America and other developing contexts (and, of course, modified in a trial and error way up to the best possible outcome).
2. SME cluster development Part II of this volume moves from the recognition of the fundamental role that a broad spectrum of policies at the national (and international) level is needed as a means to recompose the national production system and that, within these, cluster policies become an essential instrument both at the national and local level. Part II focuses on the topic of SME (‘survival’) cluster development in Latin America, with possible lessons for developing countries in general. Chapter 4 first goes through the main streams of literature on SME clustering and discusses their relevance for development purposes. These are the ‘collective efficiency approach’ (Schmitz, 1992; 1995; Nadvi, 1994; Pedersen et al., 1994; Rabellotti, 1995; 1999; Humphrey and Schmitz, 1996; van Dijk and Sandee, 2002; among others), the ‘policy-inducement approach’ (Capecchi, 1990; Cooke, 1996;
Conclusions
133
Bianchi, 1998; Cowling and Sugden, 1999; among others) and the ‘social embeddedness approach’ (Bagnasco, 1988; Becattini, 1990; Lorenz, 1992; Platteau, 1994; among others). In particular, with reference to this latter approach, is proposed a formulation that includes two aspects: ‘social cohesion’ and ‘self-realisation’. In our view, the first represents an extension of the well-studied aspect of trust, because – to the concept of trust – we add the values that promote the cohesion of an economic group and that, therefore, raise its productivity (Dasgupta, 2002; De Propris, 2005). We also emphasise the role of ‘self-realisation’ as a key aspect of the local dynamism, since it is capable of spurring entrepreneurship, innovation, firm creation, productive and commercial experiments, and so on (Brusco, 1982; Becattini, 2000b; Parrilli, 2004b). Our hypothesis is that these three kinds of approaches (‘collective efficiency’, ‘social embeddedness’ and ‘policy inducement’) are likely to be very relevant in the context of developing countries’ clusters and, in particular, of the most dynamic clusters among the whole category of survival agglomerations. Through this analysis, we have also pointed out that policy-making and development strategies are likely to benefit from an approach that takes those three approaches into account and blends them into a comprehensive theoretical and development policy structure (in Chapter 5 formalised as the ‘eclectic approach’ to SME cluster development). In Chapter 4, we also analyse the case of Nicaragua in order to study the policies and programmes that have been applied in recent years and the actual and potential benefits of adopting specific aspects linked to the above-mentioned approaches to SME cluster development. The case of Nicaragua shows that a set of support actions (more than policies) have been implemented from the early 1990s onwards, but with little effect (Parrilli, 1999a; 2002b; 2003). This failure may be connected to several factors, but we highlight, in particular, the lack of coherence and coordination among the different agencies that have been working for SME (cluster) development in terms of both support programmes and theoretical development frameworks (ibid.). Each of the three main approaches that are identified seem to be useful to construct a systemic approach that could be applied in an exploratory way to the development of specific kinds of ‘survival clusters’ in Nicaragua and other Latin American countries. In Chapter 5, we formalise our theoretical and policy approach to SME cluster development, which we call the ‘stage and eclectic’ approach. Within this concept, we emphasise the importance of two aspects. The
134
SME Cluster Development
importance of ‘stages’ is linked to the need of identifying the characteristics of each cluster and its effective potential to grow, which cannot be independent from the present development stage (Brusco, 1990; Humphrey, 1995; Knorringa, 2002; 2005, Pietrobelli and Rabellotti, 2004; 2005). Targeting too high objectives in the short term (e.g., targeting immediately the industrial district model) is more likely to drive local policy-makers, institutions and economic agents – in a word, the ‘cluster’ – to fail. In contrast, targeting feasible and progressive stages of development for dynamic ‘survival clusters’ can help these local production systems respond to the new challenges represented by globalisation and to face the threatening entry of new competitive production systems (i.e., China) in the world market. The response is neither evident nor easy, but upgrading in clusters, which leads them to shift from craft to industrial production as a first step, is the only possible answer for these local production systems (Pietrobelli and Rabellotti, 2004; 2005). However, the importance of an ‘eclectic approach’ is emphasised and linked to the need of considering the relevance of several different determinants of development. In our hypothesis, these determinants are the ones that the main streams of literature on SME cluster development (i.e., ‘collective efficiency’, ‘social embeddedness’ and ‘policy inducement’) identified over time. Pulling together these three streams and their general determinants (‘joint actions’ and ‘external economies’ within ‘collective efficiency’, ‘self-realisation’ and ‘social cohesion’ within ‘social embeddedness’ and ‘national’ and ‘local policies’ within ‘policy inducement’), could help to shape an approach that is likely to respond to the need of several SME clusters in developing countries, especially a portion of those that are considered as less dynamic: the ‘survival clusters’ (Altenburg and Meyer-Stamer, 1999; Knorringa, 2002). After this theoretical formulation, we benefit from an earlier analysis of Brusco (1990), who identified various types of local production system in the recent Italian history. Taking his contribution into account, we hypothesise that his typology represents a sequence of development stages that the general type of (Italian) industrial district made over time. Going through the history of Italian industrial districts, we explore whether our conceptual framework could resemble the trajectory of development of those specific and successful territorial cases (i.e., Italian IDs). From a detailed analysis of the related literature, we find out that the above mentioned theoretical framework and the related hypothesis (i.e., sequence) is consistent with the development trajectory of these local production systems. Chapter 5 ends identifying two basic lessons coming from that trajectory: the potential relevance for developing countries’ clusters (i.e., their
Conclusions
135
policy-makers and economic agents) of targeting one feasible ‘stage’ at a time and, simultaneously, the importance of taking an ‘eclectic policy approach’. In Chapter 6, this theoretical and policy framework is explored in three specific clusters. Two of these are from Central America: one is the furniture cluster of Masaya, in Nicaragua, and the other is the furniture cluster of Sarchí, in Costa Rica. The third cluster is selected from the region of some of the most successful clusters (i.e., industrial districts) in Emilia-Romagna, Italy. This is the furniture cluster of Forlì. From the methodological point of view, we refrain from analysing in detail the standard indicators of performance, such as sales, exports, income, investments, etc. (see Table 6.1 in Chapter 6). This kind of analysis would, indeed, show too big and static differences between competitive industrial districts from industrialised countries (such as Forlì: Bertini, 1997; Bardi, 2000) and survival clusters from developing countries (such as Masaya and Sarchí; see Pérez-Sáinz, 1994; Barahona et al., 1998; Murillo, 2002; Parrilli, 2002a). Thus, these latter would be considered as very static realities and dismissed as having little potential to develop. In contrast, the study of the development trajectory of these clusters in the past 10–20 years pushes us to take a different approach (Pérez-Sáinz, 1994; Barahona et al., 1998; 1999; Parrilli, 1998b; Narváez and Parrilli, 2000; Murillo, 2002; Parrilli, 2002a). In this sense, we prefer focusing on a qualitative analysis that tries to identify the aspects of the local dynamism. In particular, we focus on the analysis of indicators that refer to the above-mentioned main determinants of cluster development (i.e., ‘collective efficiency’, ‘social embeddedness’ and ‘policy inducement’) that are considered in the ‘eclectic approach’. Although this empirical exploration does not allow us to test and fully confirm the two hypotheses proposed in Chapter 5, elements appear that show a qualified consistency with a dynamic profile in the economic activity of these clusters. This dynamism has limitations that refer to the specific stage of development in which these more dynamic kinds of ‘survival clusters’ are situated (which very much resemble the early stages – stage I – of development of Italian industrial districts; see Brusco, 1990; and Chapter 5), and that define the possible step forward in these territorial agglomerations (but not the only possible steps). On the whole, this work confirms the plausibility of our proposed approach, which could be tested more specifically in future research and development projects that we could undertake in Latin America.
Notes 1 This more dynamic typology of survival clusters could resemble to embryonic clusters. They are survival clusters in general terms, due to the fact that their accumulation process is not clearly visible, especially in quantitative terms. Nevertheless, they have the potential to develop over time once specific conditions are in place. The analysis of these conditions is a substantial part of this volume. 2 For example, in these areas, the rates of unemployment are typically very low and the competitive performance (e.g., production, export, income per capita) very high. 3 This approach was affirmed in the 1960s and the 1970s by the structuralism theorists. 4 This concept (‘traditional SMEs’) refers to the way the SMEs conduct their production and market operations. In general, they are micro, small and medium firms mainly (although some micro enterprises and SMEs can be very high-tech oriented) operating in traditional manufacturing sectors (e.g., textiles, shoes, furniture, tiles, etc.). 5 In contrast to the case of developing countries, in industrialised countries SMEs have often been able to solve this problem (i.e., market failures) through collective actions (e.g., the creation of credit consortia, production cooperatives, export consortia) supported by state policies. These actions permit them cope with market failures. 6 There are economists, such as Becattini, Bellandi, Schmitz, Nadvi and Humphrey, that share relevant parts of these approaches. However, the analytical and policy conclusions we extract on the basis of this work (that are synthesised in the so-called ‘eclectic approach’) are peculiar to the author and represent an innovative aspect within the literature. 7 The United Nations Development Programme (UNDP) statistics linked to the Human Development Report (2002) rank these countries from among the most industrialised (Italy: 19), to intermediate (Costa Rica: 38) and low levels of development (Nicaragua: 120). 8 This chapter has been published first in Review of International Political Economy (www.tandf.co.uk) 11(5), 2004, pp. 905–25, with the title ‘Integrating the national industrial system: the new challenge for Chile’. It is based upon the results of consultancy work conducted for the Ministry of Economy and the Inter-American Development Bank (IADB) on ‘Obstacles and Opportunities to Invest for the Promotion of the Small and Medium-sized Enterprises in Chile’ (Bianchi and Parrilli, 2002a). Through this work, the author interviewed almost one hundred major government representatives, business associations, small and medium enterprises, local governments, universities, NGOs and other service enterprises. 9 In the actual debate in Latin America, there are arguments that prefer using another concept of ‘dualism’, such as found in the documents of the Economic Commission for Latin America and the Caribbean of the UN (ECLAC) and 136
Notes
10
11
12
13
14
15
16
17
18 19
137
others. This view prefers focusing on the concept of ‘heterogeneity’ in order to explain the different types of reactions that the various categories of enterprises and economic agents have within an open economy (Bianchi and Di Tommaso, 1998; Ocampo, 2001; Parrilli, 2000b). In our view, this active trade policy is still quite compatible with a general neoliberal approach, since it tends to create an impressive number of agreements on reciprocal tariff reductions with the country’s trading partners (i.e., with all the Latin American countries, the United States, East Asia and the European Union). Still, one of the complaints made by interviewed entrepreneurs was that CORFO targeted its activities mainly to the most relevant groups and firms, neglecting to work with less competitive groups of enterprises (Bianchi and Parrilli, 2002a). UF (Unidad de Fomento) is the unit adopted by Chilean government to estimate its social and policy expenditure. In 2002, its value was about $25 per unit. Of course, the recent economic slowdown in Chile is also linked to world economic problems that arose in the late 1990s. The fact that Chile remains one of the best performing countries makes this type of analysis particularly important, since it can generate meaningful insights for less dynamic developing economies. In this paper, the processing of natural resources is considered ‘modern’ because of the relatively sophisticated technology adopted by the large enterprises that usually control them (e.g., copper, pulp and paper). Nonetheless, it is difficult to call it a science-based sector that generates growing innovation and competitiveness. We should not forget that this institutional and economic development has also been possible thanks to the sound macroeconomic policies that European countries applied according to the European framework and agreements, which led to the Stabilisation Pact and the creation of the common currency. This stage appears quite far away in Latin America, where the effects of industrial destructuring add to those of the monetary instability, which followed the crisis of the unilateral opening models. Within the present national statistics large firms include firms with more than 100 workers; SMEs include firms with between 5 and 100 workers, and micro enterprises include firms with less than five workers (Banco Central de Nicaragua-GTZ-MEDE, 1998). Nonetheless, the survey of the manufacturing sector was realised in 1995 on the basis of a different classification, which considered large firms those with more than 30 workers; SMEs those firms that had between 4 and 30 workers and micro enterprises those firms that had between one and three workers (BCN, 1998; in Parrilli, 2002b). In this case, we need to clarify that there are no available data to distinguish between the part of the economically active population that shifted to other economic sectors (i.e., services and trade) and the part that increased the informal economy. For a critical review of this approach see also Singh (1992). This concept was elaborated by Arnaldo Bagnasco (1977), who described the economy of central northern of Italy as being centered on geographical agglomerations of SMEs.
138
Notes
20 For details on this analysis see next chapter. On the different models of local production systems in the recent Italian history see also Brusco (1990). 21 The cooperative system can be broadly classified in different typologies: ‘production cooperatives’, in which the associates share the means of production for realising goods and services; ‘consumption cooperatives’, in which the associates share the facilities for selling goods (supermarkets and the like); and ‘credit cooperatives’, in which the associates share common funds to access more easily to formal credit. 22 Starting from a different approach of analysis, other schools of thought (e.g., economic geography and regional studies) arrive at similar conclusions. Indeed, they also focus on the importance of regions in the new global market and on a parallel need for policies at the intra-regional level (Garofoli, 1992; Pecqueur and Rui Silva, 1992; Audretsch, 1998; Scott, 1998). 23 An interesting example of creation of trust comes from the promotion of a dynamic furniture cluster in Ceará, the north-eastern state of Brazil, out of public procurement contracts and technological assistance from the state organisation SEBRAE (Tendler and Amorim, 1996). The proactive role of public institutions has created a form of trust that resembles the systemic type. 24 As in the case of ‘self-realisation’, also in the case of ‘social cohesion’ the use of economic concepts only (such as ‘joint action’) does not explain why local societies tend to cluster and work together through time even in the absence of economic benefits. For instance, entrepreneurs tend to associate to local and national institutions independently from the benefits they can reap from those. 25 In this sense, the case discussed by Schmitz (1995b) about Sinos Valley and the joint actions arising on the basis of ‘the cost of non-cooperating’ could have a short-term duration or, in case they last, they could also be based upon the ancient common cultural roots. 26 It is interesting to note that a recent study of the Inter-American Development Bank focused on the main motivations for creating small businesses in Latin America and East Asia. The aspect of ‘self-realisation’ comes out as the main motivation amongst the entrepreneurs in these areas (Kantis, 2004). 27 In recent years, foreign penetration of the Nicaraguan market has increased dramatically, and is quite evident in the case of shoes and furniture. Production for local consumption is about $3 million and for export about $0.5 million (furniture) and $2.5 million (shoes). The country imports however $9.5 million (furniture) and $13.5 million (shoes) of these final goods (BCN, in Narváez and Parrilli, 2000). 28 Among the most active organizations working for the promotion of small industries in Nicaragua are the United Nations Industrial Development Organization (UNIDO), the Inter-American Development Bank and the governmental agencies for development of Germany (GTZ), Switzerland (COSUDE) and Holland. 29 A different position is manifested in the rural area, where small producers still count with plenty of effective cooperative organizations. 30 An interesting joint project on innovation between the World Bank and the Ministry of Agriculture discusses a number of dynamic entrepreneurial and commercial ideas in the wood and furniture value chains (MAGFOR, 1999). 31 The former Minister of Industry and Trade, Norman Caldera, confirmed the existence of this problem and its negative economic effects for the national
Notes
32
33 34
35
36
37
38
39 40
41
139
industrial sector during a conference at the Princess Hotel, in Managua, 11 July 2000. This kind of practice (attributing prizes) is widely practiced nowadays in Latin America and also in Nicaragua. Nonetheless, it is not referred to such a systemic development framework and, therefore, it is not fully understood (by producers) in its possible spillovers and very few producers effectively participate in these potentially important initiatives. This chapter has been first published with the same title in the journal European Planning Studies, Carfax Publishers, 12(8), December 2004, pp. 1115–31. I would like to add that the dependence on (a few) large enterprises can be considered as a less optimal condition in respect of Italian traditional and new competitive IDs because it tends to create rather hierarchical relations and a weaker endogenous development in the cluster, which is very much dependent upon those large enterprises (Sacchetti and Sugden, 2003; Humphrey and Schmitz, 2004). Moreover, large firms can leave the cluster whenever better conditions arise outside, leaving small firms without a market. These reasons explain why we strategically prefer focusing on the Italianate trajectory as a better organisational solution for SME clusters. Some theorists within the relevant literature share different streams. Nevertheless, the following distinction and the relative labels are useful to address the target of this work: showing the importance to take an integrated/eclectic approach to SME cluster development. On the importance of individual dynamism for the development of small firms (not taken in clusters) there are other important contributions, such as those discussed by Hjalager (1993), Cowling (1999), and Kantis (2004). Grabher (1993) presents a complementary view to that discussed above. This view emphasises the ‘dilemma’ between ‘too little embeddedness’ and ‘too much embeddedness’ within the local social tissue. ‘Too little embeddedness’ weakens the local system because too little social tissue supports the production system (e.g., little trust and cooperation); this situation resembles what we envisage as ‘self-realisation without social cohesion’. ‘Too much embeddedness’ weakens the system because it may prevent new external forces to introduce richness (e.g., knowledge) to the local production system; this situation resembles what we envisage as ‘social cohesion without self-realisation’. Other aspects can also be taken into account, such as geographic localisation, size of urbanisation, historic specificities, etc. All of these factors matter and are likely to explain part of the development of local production systems. But these seem to be rather specific and, in a quantitative type of analysis, would be more likely to be considered dummy variables. These explain some processes, but not in all cases. Public investments grew about 350 percent versus 140 percent for private investment during 1954–62 (Cohen and Federico, 2001: 100). Although In Brusco’s analysis, local institutions found their role through the ‘industrial districts Mark II’ (i.e., the following step, from the late 1980s onwards), in our view they are part of that first identification of the industrial districts as such, especially in the notorious cases of Emilia-Romagna and Tuscany. There, their constitution and expansion dates back to the 1970s and the 1980s. These are Law 44/1986, then amended by Law 95/1995, on Firms Creation and Law 46/1982, then amended by Law 317/1991, on Innovation.
140
Notes
42 Author’s interviews of entrepreneurs in the furniture cluster in Forlì, Emilia-Romagna, January 2003. 43 Based upon the analysis of ‘survival clusters’ in other developing regions (i.e., Africa and Asia), we could differentiate two typologies of clusters within the general category of ‘survival clusters’: on the one hand, those less dynamic clusters, where producers (often self-employed people in the service sector), have accepted a working solution that they are willing to leave as soon as they obtain a better job offer; on the other, those clusters where the producers are rooted in specific local traditions (often in manufacturing) and show a clear preference for maintaining a kind of work that is historically related to the local production specialisation. 44 This overall argument could be used, to some extent, for another type of less competitive cluster: the satellite cluster described by Markusen (1996) in her study of the US economy, and which also exists in developing countries (Knorringa, 2002; Guerrieri and Pietrobelli, 2004). 45 Usually the concept of ‘cluster’ and ‘industrial district’ differ: ‘cluster’ is taken as a general concept: a geographical agglomeration of firms specialised in the same sector (Schmitz, 1995a). ‘Clusters’ may be more or less successful forms of agglomerations of firms that may include also large companies that lead subcontracting chains; in specific cases the division and specialisation of labour is very low and public support minimal (the case of ‘survival clusters’ in developing countries: Altenburg and Meyer-Stamer, 1999). ‘Industrial districts’ are, in our view, a specific kind of cluster; they are successful clusters based on small and medium enterprises only, which are capable of intensive division and specialisation of labour. These firms receive significant support from public institutions (Becattini, 1990). The case studies would therefore represent ‘clusters’ in the case of Masaya and Sarchí, and ‘industrial district’ in the case of Forlì. Notwithstanding this important characterization, we often refer to Forlì as a ‘cluster’ (in general terms). The concept of ‘clustering’ is interpreted here as the process of firms agglomeration. It is a concept that we use as a synonymous for the coming together of a large number of firms in both IDs and clusters. 46 The stratified selection is due to the need to take into account the presence of a reduced number of medium enterprises in the cluster (up to 75 workers). This was not necessary in the case of the Central American clusters, because all firms there were micro and small-sized firms (with less than 30 workers) at the time of the survey. 47 The same trend can be observed, for example, in the recently studied case of the jewelry district in Arezzo, Tuscany (see Lazzeretti, 2004). 48 These considerations are based upon several interviews with external observers and institutions working in the area in 2002. Among them are Rafael Henriquez, UNIDO officer, German Cerda, technician of the institute of research and development Nitlapan, based within the Central American University in Managua, and Carlos Ivan Hueck, the mayor of the locality. 49 An interesting indication of this has been reported by producers and officers working in the forestry development project ‘PROFOR’, led by the Ministry of Agriculture and Forestry in cooperation with the World Bank and targeting innovation within the furniture production chain. They participated to international trade fairs for furniture and wooden products in Koln in 1998
Notes
141
and Valencia in 1999 and reported this kind of open appreciation and interest from international traders. 50 Presentation of the Minister of Economy, Vilma Villalobos, at the Central American School in Industrial Development and SME Policy, held at the University of Costa Rica, San José, 4 August 2003. 51 Notwithstanding this difficulty, the Ministry of Industry and Commerce (MIFIC) has been promoting the meeting among the private sector and the national and international organisations that promote SME development. This led to the organisation of the first National SME Congress (MIFIC, 2001a) and the elaboration of the governmental plan to support SMEs (MIFIC, 2001b).
Bibliography AESMEC (1998), ‘The Role of SME: Asian and European Experiences; Proceedings of the 1998 Asia-Europe SME Conference, Naples, 28–30 May. Agosin M.R. (1999), Comercio y crecimiento en Chile, Revista de la Cepal, No. 68, agosto. Aguilar J., Cordero J., and Trejos S. (1998), Estrategia tecnologica y desempeño exportador de la pequeña y mediana empresa: el caso de Costa Rica, Serie Divulgacion Economica No. 31 (Instituto de Investigaciones en Ciencias Economicas, Universidad de Costa Rica, San José). Agurto S. and Guido A. (2002), Calidad del empleo y movimientos migratorios de la población nicaragüense, El Observador Económico, Julio–Agosto (FIDEG, Managua). Alarcon C. and Stumpo G. (2001), Las pequeñas y medianas empresas en Chile, Paper de la CEPAL (CEPAL, Santiago). Altenburg T. and Meyer-Stamer J. (1999), How to promote clusters: policy experiences from Latin America, World Development, 27(9), September. Amin A. (1994), ‘The difficult transition from informal economy to Marshallian industrial district’, in UNCTAD, Technological Dynamism in Industrial Districts (UNCTAD, Geneva). Amsden A. (1994), Why Isn’t the Whole World Experimenting With the East Asian Model to Develop? Review of the East Asian Miracle, World Development, 22(4). Aranguren M.J., Audretsch D. and Callejon M. (2005), ‘Empresarialidad: contribucion al crecimiento y al bienestar economico’, in Parrilli, Bianchi and Sugden (2005); see below. Arrighetti A. and Seravalli G. (1997), ‘Istituzioni e dualismo dimensionale dell’industria italiana’, in F. Barca (ed.), Storia del capitalismo italiano dal dopoguerra ad oggi (Donzelli Editore, Roma). Artola N. and Parrilli M.D. (2006), ‘The Development of the Dairy Cluster in Boaco and Chontales, Nicaragua’, in C. Pietrobelli and R. Rabellotti (eds), Upgrading in Clusters and Value Chains in Latin America: The Role of Policies (InterAmerican Development Bank, Washington, DC). Asheim B. (1994), ‘Industrial districts, inter-firm co-operation and endogenous technological development: the experience of developed countries’, in UNCTAD, Technological Dynamism in Industrial Districts (UNCTAD, Geneva). Audretsch D.B. (1998), Agglomeration and the Location of Innovative Activity, Oxford Review of Economic Policy, 14(2), summer. Axelrod R. (1984), The Evolution of Cooperation (Basic Books, New York). BCC (Banco Central de Chile) (2002), Indicadores económicos (BCC, Santiago), at ⬍www.bcentral.cl⬎ . BCCR (Banco Central de Costa Rica) (2002), Indicadores economicos (BCCR, San José de Costa Rica), at ⬍www.bccr.fi.cr⬎. BCIE (Banco CentroAmericano de Integración Económica) (1999), Informe de proyecto, mimeo, Managua. BCN (Banco Central de Nicaragua) Instituto Nacional de Estadística – Ministerio de Economía (1996). Directorio nacional de las industrias manufactureras (BCN, Managua). 142
Bibliography
143
BCN (Banco Central de Nicaragua) (2000, 2003), Informe Annual (BCN, Managua). BID (Banco Interamericano de Desarrollo) (1999), Servicios para el Desarrollo Empresarial, Proyecto conjunto con GTZ-INDE, mimeo, Managua. Bagnasco A. (1977), Le Tre Italie (Le Monnier, Firenze). Bagnasco A. (1988), Il mercato come costruzione sociale (Il Mulino, Bologna). Bagnasco A. (1999), Tracce di comunitá (Il Mulino, Bologna). Balassa B., Bueno G., Kuczynksy P. and Simonsen M.H. (1986), Toward Renewed Economic Growth in Latin America (Institute for International Economics, Washington, D.C). Banco Central de Nicaragua (BCN)-GTZ-MEDE (1998), Directorio de las Empresas (Direccion de Contabilidad, Managua). Barahona T., Parrilli M.D. and Narváez E. (1998), La cadena maderera: en busca de los eslabones perdidos, Cuaderno de Investigación No. 6, Nitlapán (UCA, Managua). Barahona T., Parrilli M.D., Narváez E. and Mairena R. (1999), La cadena maderera en las Segovias, Cuaderno de Investigación No. 9, ADESO-Nitlapán (UCA, Managua). Barca F. (1997), ‘Compromesso senza riforme del capitalismo italiano’, in F. Barca (ed.), Storia del capitalismo italiano dal dopoguerra ad oggi (Donzelli Editore, Roma). Bardi A. (2000), ‘Il settore del mobile imbottito: una proposta di politica industriale’, in Economia e società regionale, No. 1, (Angeli, Milano). Baroncelli, A. (2001), Percorsi imprenditoriali generati nell’Universitá (CLUEB, Bologna). Becattini G. (1990), ‘The Marshallian Industrial District as a Socio-economic Notion’, in F. Pyke and G. Becattini and W. Sengenberger, Industrial Districts and Interfirm Cooperation, ILO, Geneva. Becattini G. (2000a), Il distretto industriale (Rosenberg & Sellier, Torino). Becattini G. (2000b), ‘La fioritura della piccola impresa ed il ritorno dei distretti industriali’, in G. Becattini (ed.), Il distretto industriale (Rosenberg & Sellier, Torino). Bell M. and Albu M. (1999), Knowledge systems and technological dynamism in industrial clusters in developing countries, World Development, 27(9), September. Bellandi M. (2001), Local development and embedded large firms, Entrepreneurship and Regional Development 13(1). Bellandi M. (2005), ‘Pequeñas empresas y distritos industriales’, in Parrilli, Bianchi and Sugden (2005); see below. Bellini N. (1996), Regional economic policies and the non-linearity of history, European Planning Studies, 4(1). Bellini N. (2005), ‘Servicios de apoyo a las empresas: un mapa conceptual y algunas practicas interesantes’, in Parrilli, Bianchi and Sugden (2005); see below. Belussi F. (1999), Policies for the Development of Knowledge-intensive Local Production Systems, Cambridge Journal of Economics, 23(6). Belussi F. and Bertini S. (1998), ‘Evoluzione, apprendimento e modelli di acquisizione delle conoscenze: due sistemi locali a confronto nel settore del mobile imbottito’, in Economia e Societá Regionale, No. 2 (Franco Angeli, Milano). Bertini S. (1997), Il settore del mobile imbottito a Forlì (IRES, Progetto Adapt-Learning Region, Bologna). Bertini S. (1998), SMEs, industrial development and competitiveness in Italy, paper presented at AESMEC Conference, Naples, May. Bessant J. and Kaplinsky R. (1995), Industrial Restructuring: Facilitating Organisational Change at the Firm Level, World Development, 23(1). Best M. (1990), The New Competition (Pinter, London).
144
Bibliography
Bianchi P. (1992), ‘Industrial Strategies and Structural Policies’, in K. Cowling and R. Sugden (eds), Current issues in Industrial Economic Strategy (Manchester University Press, Manchester). Bianchi P. (1998), Industrial Policy and Economic Integration (Routledge, London). Bianchi P. (2000), ‘Policies for Small and Medium-sized Enterprises (SMEs)’, in W. Elsner and J. Groenewegen (eds), Industrial Policies After 2000 (Kluwer Academic, London). Bianchi P. (2002a), Che cosa impariamo dalla crisi argentina, L’Industria, Vol. 1 (Il Mulino, Bologna). Bianchi P. (2002b), La Storia dell’Industria italiana dall’unitá ad oggi (Il Mulino, Bologna). Bianchi P. and Di Tommaso M. (1998), The Role of SMEs in a changing global scenario: towards a new policy approach, paper presented to AESMEC Conference, Naples, May. Bianchi P., Di Tommaso M. and Rubini L. (2000), Le api audaci (Franco Angeli, Milano). Bianchi P. and Labory S. (eds) (2004), The Economic Importance of Intangible Assets, (Ashgate, London). Bianchi P., Labory S., Paci D. and Parrilli M.D. (2006), ‘Small and Medium-sized Enterprise Policies in Europe, Asia and Latin America’, in P. Bianchi and S. Labory (eds), Handbook of Industrial Policy (Edward Elgar, Cheltenham). Bianchi P. and Parrilli M.D. (2002a), Obstaculos y oportunidades de inversion publica para el desarrollo de las pequeñas y medianas empresas en Chile (Inter-American Development Bank, Santiago). Bianchi P. and Parrilli M.D. (2002b), Small and Medium-Sized Enterprises: A Comparative Approach to Latin America and the European Union, Research Paper No. 26 (University of Ferrara). Bianchi P. and Parrilli M.D. (2004), ‘La experiencia italiana de fomento a la empresarialidad’, in H. Kantis (ed.), Desarrollo Emprendendor: America Latina y la Experiencia Internacional (Inter-American Development Bank, Washington, DC). Bisang R., Burachik G. and Katz J. (1995), Hacia un nuevo modelo de organizacion industrial. El sector manufacturero argentino en los años 90 (CEPAL/Alianza Editorial, Buenos Aires). Bouchier S. (1998), Las MIPYMEs en Centroamérica (GTZ, San Salvador). Brioschi F. and Cainelli G. (2001), Diffusione e caratteristiche dei gruppi di piccole e medie imprese nelle aree distrettuali: Dove va il distretto industriale?, (Fondazione Giordano dell’Amore, Giuffré, Milano). Brusco S. (1982), The Emilian Model: Productive Decentralistion and Social Integration, Cambridge Journal of Economics, 6, pp. 167–84. Brusco S. (1990), ‘The Idea of Industrial Districts’, in F. Pyke, G. Becattini and W. Sengenberger (eds), Industrial Districts and Interfirm Cooperation (ILO Geneva). Brusco S. and Paba S. (1997), ‘Per una storia dei distretti industriali italiani dal secondo dopoguerra agli anni novanta’, in F. Barca (ed.), Storia del Capitalismo Italiano (Donzelli, Roma). Caddy J. (1998), The Changing Role of SMEs in Japan, paper presented to AESMEC Conference, Naples, May. Camera di Commercio (2005), Rapporto Sull’economià 2004 (Ufficio Studi, Forli). Capaldo G., Lo Storto C. and Marino A. (1998), SMEs in Europe: taxonomies and enterprises policies, paper presented to AESMEC Conference, Naples, May.
Bibliography
145
Capecchi V. (1990), A History of Flexible Specialisation and Industrial Districts in Emilia Romagna, in F. Pyke and W. Sengenberger (eds), Industrial Districts and Interfirm Cooperation (ILO, Geneva). Cardoso F. and Faletto E. (1972), Dependencia y desarrollo en America Latina (Editores Siglo XXI, Mexico). Cazzola F. (2000), L’Italia contadina (Editori Riuniti, Roma). Ceglie G. and Dini M. (1999), El desarrollo de clusters y redes de Pymes en los paises en desarrollo: el caso del Nicaragua, Revista Encuentro, 31(50). CEPAL (2001), Anuario Estadistico de América Latina y el Caribe, LC/G.2151-P/B (United Nations, New York). CEPAL (2003), Anuario Estadistico de América Latina y el Caribe, LC/G.2190-P/B (United Nations, New York). Chenery H. and Syrquin R. (1980), ‘A Comparative Analysis of Industrial Growth’, in R.C. Matthews (ed.), Economic Growth and Resources, Vol. 2 (Trends and Factors) (St Martin’s Press, New York). Chudnovsky D., Kosakoff B. and Lopez A. (1999), Las multinacionales latinoamericanas: sus estrategias en un mundo globalizado (Fundo de Cultura Economica, Buenos Aires). Cimoli M. and Katz J. (2001), Structural Reforms, Technological Gaps and Economic Development: A Latin American Perspective, Druid Nelson and Winter Conference, Aalborg, 12–15 June. Cipolla C.M. (1996), Uomini, tecniche, economie (6th edn, Feltrinelli, Milan). CNA (Confederazione Nazionale degli Artigiani) (2002), Database (Forlì). Cohen J. and Federico G. (2001), The growth of the Italian economy 1820–1960, Cambridge University Press, Cambridge. Cooke P. (1996), Building a twenty-first century regional economy in Emilia-Romagna, European Planning Studies, 4(1). Cooke P. and Wills D. (1999), Small firms, social capital and the enhancement of business performance through innovation programmes, Small Business Economics, 13, September, pp. 219–34. CORFO (1999), Informe Anual (CORFO, Santiago). Cornia G.A., Jolly R. and Stewart F. (1989), Per un aggiustamento dal volto umano (Franco Angeli, Milan). Cowling K. and Sugden R. (1997), Beyond Capitalism (Pinter, London). Cowling K. and Sugden R. (1999), The wealth of the localities, New Political Economy, 4(3). Cowling M. (1999), ‘The Entrepreneurial Society in Practice’, in K. Cowling (ed.), Industrial Policy in Europe (Routledge, London). Cristaller W. (1972), ‘How I Discovered the Theory of Central Places’, in P. English and R. Mayfield (eds), Man, Space and Environment (Oxford University Press, Oxford). CSIL (Centro Studi Industria Leggera) (1997), Analisi e proposte per lo sviluppo del mobile imbottito come sistema territoriale (Comune di Forlì, Provincia di Forlì, CNA, Confartigianato, Assoindustria, Milano). Dasgupta P. (2002), ‘Social Capital and Economic Performance: Analytics’, in E. Ostrom and T.K. Ahn (eds), Social Capital: A Reader (Edward Elgar, Cheltenham). De Franco M., Sevilla R., Downing H. and Sandino A. (1996), Caracterización y propuesta de estrategia para el sector semi-industrial de productos lácteos en Nicaragua, Vols I and II (Consultores en Gerencia y Economía, Managua).
146
Bibliography
De Propris L. (2005), ‘Confianza y capital social como factores intangibles del desarrollo’, in Parrilli, Bianchi and Sugden (2005); see below. De Soto H. (1989), El otro sendero (ILDIS, Lima). Dei Ottati G. (1994), Trust and the Interlinking of Credit, Cambridge Journal of Economics, 18. Di Tommaso M., Rubini L. and Lanzoni E. (2001), Support to SMEs in the Arab Region: The Case of Tunisia (UNIDO, Bologna). Dini M. and Stumpo G. (2002), Análisis de la política de fomento a las pequeñas y medianas empresas en Chile, LC/L.1838-P/E (CEPAL, Santiago). Dore R. (1983), Goodwill and the Spirit of the Market, British Journal of Sociology, 34. Dornbusch R. (1996), The Latin Triangle, MIT Paper (Boston, MIT). European Commission (1998), The competitiveness of European enterprises in the face of globalisation. How it can be encouraged, Commission, COM (1998) 718 final (Brussels, European Commission). European Commission (2002), Produttivitá: la parola chiave per la competitivitá delle economie e delle imprese europee, Commission, COM (2002) 262 definitivo (Brussels, European Commission). European Union (2001), Corporate and research-based Spin offs: proceedings from the experts workshop, Brussels, 18 January. Falabella G. (2001), ‘Desarrollo diverso y proyecto país’, in O. Muñoz, Más allá del bosque (FLACSO-Chile, Santiago), pp. 235–67. Fauri F. (1996), Lo sviluppo industriale della provincia di Forlì-Cesena (1926–1996) (Associazione degli Industriali di Forlì e Cesena, MDM Grafiche, Forlì). Federconfidi (1997), Confidi e congiuntura economica: rilevazioni 1984–94 (Franco Angeli, Milano). Ffrench-Davis R. (2002), El impacto de las exportaciones sobre el crecimiento en Chile, Revista de la Cepal, No. 76, pp. 143–60. FNI (Financiera Nicaraguense de Inversiones) (1999), Informe de resultados, Report (FNI, Managua). Fischer R. (2001), Trade liberalisation, development and government policy in Chile, paper presented at the 11th Bilateral Economic Development Seminar, Tokyo (2000), Universidad de Chile, Santiago. Frank G.A. (1970), Il sottosviluppo capitalista (Jaca Book, Milano). Frank G.A. (1992), 1492 and Latin America at the Margin of World System History, paper presented to 32nd Annual Meeting of the International Studies Association, Atlanta, 1–4 April. Fujita M., Krugman P. and Venables A. (2001), The Spatial Economy (MIT Press, Cambridge, MA). FUNDES (2002), Programa de mejora de las condiciones del entorno empresarial de las Pymes (FUNDES, Santiago). Furtado J., Garcia R. and Suzigan W. (2005), ‘Pequeñas y medianas empresas en sistemas locales de producción en Brazil: encadenamientos locales y globales’, in Parrilli, Bianchi and Sugden (2005); see below. García J. (1999), Propuesta de politicas para el desarrollo de las PYMEs (DGPYMEMinisterio de Fomento Industria y Comercio, Managua). Garofoli G. (1992) (ed.), Endogenous development and Southern Europe (Avebury, Aldershot). Gazeta Oficial (2002), Ley de Fortalecimiento de las Pequeñas y Medianas Empresas, Asamblea Legislativa, Gazeta Oficial No. 94, 17 May, San José.
Bibliography
147
Gereffi G. and Korzeniewicz M. (eds) (1994), Commodity Chains and Global Capitalism (Greenwood Press, Westport, CT). Ghani E. and Stewart F. (1991), How significant are externalities for development?, World Development, 19(6), pp. 569–93. Gilly J.P. and Torre A. (1998), Introduzione, in Prossimità: dinamica industriale e territorio: Studi francesi, L’Industria, 19(3), luglio-settembre. Grabher G. (1993), ‘Rediscovering the social in the economics of interfirm relations’, in G. Grabher (ed.), The Embedded Firm (Routledge, London). Granovetter M. (1985), Economic action and social structure: the problem of embeddedness, American Journal of Sociology, 91(3), November, pp. 481–510. Guerrieri P. and Pietrobelli C. (2004), Industrial districts’ evolution and technological regimes: Italy and Taiwan, Technovation, 24(11), November, pp. 899–914. Harris J.R. and Todaro M. (1970), Migration, unemployment and development, in the American Economic Review, 60(1), March, pp. 126–42. Hart J. (1977), Informal income opportunities and urban employment in Ghana, Journal of Modern African Studies, 11(1). Henriquez R. (2002), Informe de la primera reunión de la comisión de producción en el Municipio de Masaya, UNIDO officer’s report, July, mimeo, Managua. Heymann D. and Leijonhufvud A. (1992), High Inflation (Oxford University Press, New York). Hirschman A.O. (1984), ‘A Dissenter’s Confession: The Strategy of Economic Development Revisited’, in G. Meier and D. Seers (eds), The Pioneers of Development (Oxford University Press, New York). Hjalager A. (1993), Why not Entrepreneurs?, Entrepreneurship and Regional Development, 1, pp. 85–97. Humphrey J. and Schmitz H. (1995), Principles for Promoting Clusters and Networks of SMEs, Discussion Paper No. 1 (UNIDO, Vienna). Humphrey J. and Schmitz H. (1996), The Triple C Approach to Local Industrial Policy, World Development, 24(12), December, pp. 1859–77. Humphrey J. and Schmitz, H. (1998), Trust and Cooperation, IDS Discussion Paper No. 150, University of Sussex, Brighton. Humphrey J. and Schmitz H. (2004), ‘Chain Governance and Upgrading: Taking Stock’, in H. Schmitz (ed.), Local Enterprises in the Global Economy: Issues of Governance and Upgrading (Edward Elgar, Cheltenham). Humphrey J. (1995), Industrial Reorganization in Developing Countries: From Models to Trajectories, World Development, 23(1), pp. 149–62. Hymer S. (1972), ‘The Multinational Corporation and the Law of Uneven Development’, in J. Bhagwati (ed.), Economics and the World Order (Macmillan, London). INCAE (1997), Agenda de competitividad para Centroamerica (INCAE, Managua). INE (Instituto Nacional de Estadistica) (1999), Encuesta de Cuentas Nacionales (INE, Santiago). INE-Gobierno-BancoEstado (2001), Primera encuesta de la microempresa y tercera de las pequeñas y medianas empresas manufactureras (INE, Santiago). Innocenti A. and Labory S. (2002), The advantages of outsourcing in terms of information management, Quaderni del Dipartimento no 370 (Università di Siena). INPYME (2000), Programa de intervención integral 2000–2005, available at www.inpyme.gob.ni (Gobierno de Nicaragua, Managua). Kantis H. (ed.) (2004), Desarrollo Emprendendor: America Latina y la Experiencia Internacional (Inter-American Development Bank, Washington, DC).
148
Bibliography
Kantis H., Ishida M. and Komori M. (2002), Entrepreneurship in Emerging Economies: The Creation and Development of New Firms in Latin America and East Asia (IADB, Washington, DC). Kaplinsky R. and Readman J. (2001), Integrating SMEs in Global Value Chains: Spreading the Gains from Globalisation, IDS Bulletin, 32(3), University of Sussex, Arighton. Knorringa P. (1999), Agra: an Old Cluster Facing the New Competition, World Development, 27(9) September. Knorringa P. (2002), ‘Cluster Trajectories and the Likelihood of Endogenous Upgrading’, in M.P. van Dijk and H. Sandee (eds), Innovation and Small Enterprises in the Third World (Edward Elgar, Cheltenham). Knorringa P. (2005), ‘Trayectorias de Clusters Urbanos y Oportunidades de Mejora Endógena en Países en Desarrollo’, Parrilli, Bianchi and Sugden (2005); see below. Kosakoff B. (1993), El desafio de la competitividad: la industria argentina en la transformacion (CEPAL/Alianza Editorial, Buenos Aires). Kosakoff B. (2000), Corporate Strategies Under Structural Adjustment in Argentina (Macmillan, London). Krueger A. (1983), Trade and employment in developing countries (Chicago University Press, Chicago). Krugman P. (1998), What’s New About the New Economic Geography?, Oxford Review of Economic Policy, 14(2), Summer. Laenen A. (1988), Dinámica y transformación de la pequeña industria en Nicaragua (CEDLA Latin America Studies, San José). Lazerson M. (1990), ‘Knitwear Industry in Modena’, in F. Pyke, G. Becattini and W. Sengenberger, Industrial Districts and Interfirm Cooperation (ILO, Geneva). Lazzaretti L. (2004), The industrial districts as a community of organisations: the emergent jewellery district of Arezzo (1947–2001), paper presented at EUNIP Conference, 13–15 December, Birmingham. Lewis W.A. (1973), ‘Sviluppo economico con disponibilita illimitata di manodopera’, in B. Jossa (ed.), Economia del sottosviluppo (Il Mulino, Bologna). Locke R. (1995), The Remaking of the Italian Economy (Cornell University Press, Ithaca, NY). Lorenz K. (1992), ‘Trust, Community and Cooperation’, in M. Storper and A.J. Scott (eds), Pathways to Industrialisation and Regional Development (Routledge, London). Losch A. (1940), Die Raumliche Ordnung der Wirtschaft, Fischer, Jena; published in 1954 as The Economics of Location, Yale University Press, New Haven, CT. Maddison A. (1991), Monitoring the World Economy 1820–1992 (OECD, Paris). Maddison A. (2001), The World Economy: A Millennial Perspective (OECD, Paris). MAGFOR (Ministerio de Agricultura, Ganaderia y Forestal) (1999), Proyecto de Fomento Forestal sobre la Innovación, Proyecto PROFOR, MAGFOR-Banco Mundial, mimeo, Managua. Markusen A. (1996), Sticky Places in Slippery Space: A Typology of Industrial Districts, Economic Geography, 72. Marshall A. (1916), Principles of Economics (Macmillan, London). Marshall A. (1918), Industry and Trade (Macmillan, London). McCormick D. (1999), African enterprise clusters and industrialization: theory and reality, World Development, 27(9).
Bibliography
149
Meller P., Ryan R. and Solimano A. (1996), Growth, Equity and the Environment in Chile: Issues and Evidence, World Development, 24(2), pp. 255–72. Mendoza R. (1990), Los costos del verticalismo: campesinos sin rostro, Revista Envío, 107, Septiembre. Mendoza R. (2002), La paradoja del café: el gran negocio mundial y la gran crisis campesina Nitlapán (Oxfam, Christian Aid, Cafod, Universidad Centroamericana Press, Managua). Miller L. and Bianchi P. (1998), ‘Las Pymes en el marco de los procesos de integración’, in L. Miller and P. Bianchi (eds), Innovación y territorio (LindavistaEditJus, Mexico). Ministère de Economie de Tunisie (2000), La Mise à Niveau (Tunis). Ministerio de Economía (2002), Estatuto de la pequeña empresa (Santiago de Chile). Ministerio de Hacienda (2001), Presupuesto 2002 (Dirección de Presupuestos, Santiago de Chile). MIFIC (Ministerio de Industria, Fomento y Comercio) (2001a), Memoria del I Congreso Nacional PYMEs (MIFIC, Managua). MIFIC (Ministerio de Industria, Fomento y Comercio) (2001b), Política de fomento a las PYMEs (MIFIC, Managua). Moori-Konig V., Ferraro C.A. and Yoguel G. (2002), ‘Situacion y politica para las pequeñas y medianas empresas industriales en Argentina’, in W. Peres and G. Stumpo (eds), Las pequeñas y medianas industrias en America Latina (CEPAL/Siglo XXI, Mexico). Murat M. and Paba S. (2001), ‘Flussi migratori e modelli di sviluppo industriale. L’esperienza italiana dal dopoguerra agli anni ’90’, in G. Lunghini, F. Silva and R. Targetti, Politiche pubbliche per il lavoro (Il Mulino, Bologna). Murillo I. (2002), Diagnóstico de las empresas del sector muebles en Sarchí, Ministerio de Salud de Costa Rica, mimeo, Sarchí. Nadvi K. (1994), ‘Industrial Districts Experiences in Developing Countries’, in UNCTAD, Technological Dynamism in Industrial Districts (UNCTAD, Geneva). Nadvi K. (1999a), Collective Efficiency and Collective Failure: The Response of the Sialkot Surgical Instrument Cluster to Global Quality Pressures, World Development, 27(9). Nadvi K. (1999b), Shifting Ties: Social Networks in the Surgical Instrument Cluster of Sialkot, Pakistan, Development and Change, 30. Nadvi K. and Schmitz H. (eds) (1999), Clustering and Industrialization: Introduction, World Development, 27(9). Narváez E. (2001), Competitividad industrial: tipologia de productores de calzado en Nicaragua, Cuadernos de Investigacion No. 13, Nitlapán (Universidad Centroamericana, Managua). Narváez E. and Parrilli M.D. (2000), El mercado nacional de los muebles en Nicaragua (Proyecto Nicamueble Finlandia-Chile-Nicaragua-INPYME, Managua). Noteboom B. (2002), Trust (Edward Elgar, Cheltenham). Observatoire des PME Europeennes (2000), Sixieme rapport. Rapport de synthese (Commission Europeenne, Brussels). Observatory of European SMEs (2002/1), Highlights from the 2001 Survey (European Commission, Brussels). Observatory of European SMEs (2002/2), SMEs in Europe, Including a Glance at EU Candidate Countries (European Commission, Brussels). Observatory of European SMEs (2003), Highlights from the 2003 Observatory (European Commission, Brussels), p. 8.
150
Bibliography
Ocampo A.J. (2001), Re-thinking the Development Agenda (ECLAC, Santiago). O’Donnell G. (1982), El estado burocratico autoritario (Editorial Belgrano, Buenos Aires). Parrilli M.D. (1998a), Building up Commodity Chains for Competitiveness in Nicaragua: the Cases of Coffee and Sesame Seeds, Pensamiento Propio, 3(6) (CRIES, Managua). Parrilli M.D. (1998b), Competencia destructiva hacia adentro vrs competencia constructiva hacia fuera, Revista Encuentro, 30(47). Parrilli M.D. (1999a), Apostando a un nuevo actor del desarrollo: las PYMEs industriales en Nicaragua, Revista Encuentro, 31(50), December. Parrilli M.D. (1999b), A necessary turnaround: betting on cities and towns, Revista Envio No. 204, March. Parrilli M.D. (2001), Analisis Subsectorial en la rama cuero/calzado de Nicaragua en El 2000, Cuadernos de Investigacion No. 14, Nitlapán (Universidad Centroamericana, Managua). Parrilli M.D. (2002a), ‘Innovation and Competitiveness Within the Small Furniture Industry in Nicaragua’, in M.P van Dijk and H. Sandee, Innovation and Small Enterprises in the Third World (Edward Elgar, Cheltenham). Parrilli M.D. (2002b), ‘Las pequeñas y medianas industrias en Nicaragua’, in W. Peres and G. Stumpo, Las pequeñas y medianas industrias en America Latina (ECLAC/Siglo XXI Editores, Mexico City). Parrilli M.D. (2003), La competitividad de las MIPYMEs en Nicaragua, in R. Hernandez, La competitividad de las MIPYMEs en Centroamerica (ECLAC-GTZ, Mexico City). Parrilli M.D. (2004a), Integrating the Industrial System: The New Challenge for Chile, Review of International Political Economy, 11(5). Parrilli M.D. (2004b), A Stage and Eclectic Approach to Industrial Districts Development: Two Policy Keys for Survival Clusters in Developing Countries, European Planning Studies, 12(8), December. Parrilli M.D. (2004c), ‘Le piccole e medie imprese latinoamericane nell’era della globalizzazione: approcci di politica per lo sviluppo industriale a confronto’, in G. Prodi, Politiche industriali in una economia aperta (Franco Angeli, Milan). Parrilli M.D. (2005), Industrializzazione, sviluppo ed istituzioni internazionali (TECOM, University of Ferrara). Parrilli M.D., Bianchi P. and Sugden R. (2005), Alta tecnologia, productividad y redes: un enfoque sistémico para el desarrollo de las pequeñas y medianas empresas (Colegio de Tlaxcala Press, Mexico). Pecqueur B. and Rui Silva M. (1992), ‘Territory and Economic Development: The Example of Diffuse Industrialisation’, in G. Garofoli (ed.), Endogenous Development in Southern Europe (A vebury, Aldersheft). Pedersen P.O., Sverrisson A. and, van Dijk M.P. (eds) (1994), Flexible Specialization: The Dynamis of Small-Scale Industries in the South (ITP Publications, London). Peres W. and Stumpo G. (2000), Small and medium-sized manufacturing enterprises in Latin America, World Development, 28(9), pp. 1643–55. Elsevier-North Holland. Peres W. and Stumpo G. (2002), Pequeñas y medianas empresas industriales en América Latina y el Caribe (CEPAL/Editorial Siglo XXI, Santiago). Perez-Aleman P. (2000), Learning, adjustment and economic development: transforming firms, the State and associations in Chile, World Development, 28(1), pp. 41–55. Pérez-Sáinz J. P. (1994), Sarchí y el Capital Social (FLACSO, San José, Costa Rica).
Bibliography
151
Pérez-Sáinz J.P. (1995), Globalización y neoinformalidad en América Latina, Nueva Sociedad 135, January. Perry G. and Serven L. (2002), La anatomia de una crisis multiple: que tenia Argentina de especial y que podemos aprender de ella, Desarrollo Economico, 42(167), pp. 323–75. Pietrobelli C. and Rabellotti R. (2002), Business development services in Italy. An empirical analysis of three regional experiences: Emilia-Romagna, Lombardia and Veneto, Serie Desarrollo Productivo (CEPAL-ECLAC, Santiago del Chile). Pietrobelli C. and Rabellotti R. (eds) (2004), Upgrading in Cluster and Value Chains in Latin America: The Role of Policies (Inter-American Development Bank, Washington, DC). Pietrobelli C. and Rabellotti R. (2005), ‘Un Enfoque Sectorial de Políticas para Clusters y Cadenas de Valor en América Latina’, in Parrilli, Bianchi and Sugden (2005); see above. Piore M. and Sabel C. (1984), The Second Industrial Divide (Basic Books, New York). Platteau J.P. (1994), Behind the market, where real societies exist, Journal of Development Studies, 30(3), April, pp. 533–77. Porter M. (1990), The Competitive Advantage of Nations (Free Press, New York). Portes A., Benton M. and Castells L. (1989), The Informal Economy: Studies in Advanced and Less Developed Countries (Johns Hopkins University Press, Baltimore). Prebisch R. (1982), Hacia una dinámica del desarrollo, La Obra de Prebisch en la CEPAL (Fondo de Cultura Económica, México). Prebisch R. (1984), ‘Five Stages in My Thinking on Development’, in G. Meyer and D. Seers (eds), Pioneers in Development (Washington, DC: World Bank/Oxford University Press). Putnam R. (1993), Making Democracy Work (Princeton University Press, Princeton, NJ). Pyke F. and Sengenberger W. (1990), ‘Introduction’, in F. Pyke, G. Becattini and W. Sengenberger (eds), Industrial Districts and Interfirm Cooperation (ILO Geneva). Pyke F. and Sengenberger W. (eds) (1992), Industrial districts and local economic regeneration (ILO, Geneva). Rabellotti R. (1995), Is There an Industrial District Model? Footwear Districts in Italy and Mexico Compared, World Development, 23(1), pp. 29–41. Rabellotti R. (1999), Recovery of a Mexican Cluster: Devaluation Bonanta or Collective Efficiency?, World Development 27(9), September. Rabellotti R. (2004), ‘How Globalization Affects Italian Industrial Districts: The Case of Brenta’, in H. Schmitz (ed.), Local Enterprises in the Global Economy: Issues of Governance and Upgrading (Edward Elgar, Cheltenham). Reinhardt N. and Peres, W. (2000), Latin America’s New Economic Model, World Development, 28(9), pp. 1543–66. Reyes O. (1993), La agroindustria en Nicaragua (ESECA-UNAN, Managua). Rocha J.L. (1998), Superintendencia: el oscuro objeto del pleito? Revista Envío, No. 196, July. Rocha J.L. (1999), San Francisco Libre: voluntad a prueba de desiertos, Revista Envío, No. 202, January. Rodrik D. (1994), ‘King Kong Meets Godzilla: The World Bank and The East Asian Miracle, in A. Fishlow et al. (eds), Miracle or Design? Lessons from the East Asian Experience, Overseas Development Council, Policy Essay No. 11 (ODC, Washington, DC). Rogoff K. (2004), The sisters at 60, The Economist, 24 July.
152
Bibliography
Romjin H. (2002), ‘Acquisition of Technological Capability in Small Firms in Developing Countries’, in M.P. van Dijk and H. Sandee (eds), Innovation and Small Firms in the Third World (Edward Elgar, Cheltenham). Russo M. (1989), ‘Technical Change and the Industrial Districts: The Role of Inter-firm Relations in the Growth and Transformation of Ceramic Production in Italy’, in E. Goodman (ed.), Small Firms and Industrial Districts in Italy (Routledge, London). Sacchetti S. and Sugden R. (2003), The Governance of Networks and Economic Power: The Nature and Impact of Subcontracting Relations, Journal of Economic Surveys, 17(5), December, pp. 669–92. Sacchetti S. and Sugden R. (2005), ‘Redes y Conocimiento en las Empresas Farmacéuticas después de la Revolución Biotecnológica: Implicaciones para los Países en Desarrollo’, in Parrilli, Bianchi and Sugden (2005); see above. Sachs J.D. and Warner A.M. (1995), Natural Resource Abundance and Economic Growth, Technical Report No. 5398 (NBER, Cambridge, MA). Saez S. and Valdez J.G. (1999), Chile y su política comercial lateral, Revista de la CEPAL, 67, April, pp. 81–94. Santiago. Sako M. (1992), Prices, Quality and Trust: Inter-Firm Relations in Britain and Japan (Cambridge University Press, Cambridge). Scapini J.C. (1998), Desarrollo productivo de la industria manufacturera, Estadística y Economía, Segundo Semestre, Santiago. Schmitz H. (1992), On the Clustering of Small Firms, IDS Bulletin, 23(3), pp. 64–9. Schmitz H. (1995a), Collective efficiency: growth path for small-scale industry, Journal of Development Studies, 31(4), pp. 529–66. Schmitz H. (1995b), Tale of a Supercluster, World Development, 23(1), January, pp. 9–28. Schmitz H. (1999), From ascribed to earned trust in exporting clusters, Journal of International Economics, 48(1), pp. 139–50. Schmitz H. and Musick B. (1994), Industrial Districts in Europe: Policy Lessons for Developing Countries, World Development, 22(6). Schumacher E.F. (1973), Small is Beautiful (Blond & Briggs, London). Schumpeter J.A. (1934), The Theory of Economic Development: An Inquiry into Profits, Capital, Credit, Interest, and the Business Cycle (Boston, Harvard University Press). Scott A.J. (1998), Regions and the World Economy: The Coming Shape of Global Production, Competition, and Political Order (Oxford: Oxford University Press). Singh A. (1992), ‘Industrial Policy in the Third World in the 1990s: Alternative Perspectives’, in K. Cowling and R. Sugden (eds), Current Issues in Industrial Economic Strategy (Manchester: Manchester University Press). Solórzano O. (2000), Implementación de reformas de políticas económicas en Nicaragua, UNCTAD, Project Paper, mimeo, Managua. Speer, John G. (1997), ‘The Urban Informal Economic Sector’, in T.W. Walker (ed.), Nicaragua Without Illusions: Regime Transition and Structural Adjustment in the 1990s (Wilmington, DE: Scholarly Resources). Stiglitz, J. (1998), Towards a New paradigm for Development, 9th Raul Prebisch Lecture (UNCTAD, Geneva). Stiglitz J. (2001), In un mondo imperfetto (Donzelli Editori, Roma). Storper M. (1998), Innovazione come azione collettiva, L’Industria, 19(3), July–September. Stumpo, G. (2000), ‘I sogni del giaguaro: riforme economiche, ristrutturazione produttiva e strategie imprenditoriali in Cile’, in A. Goldstein (ed.), America
Bibliography
153
Latina 2000: dalla stabilizzazione alla crescita (Istituto per gli Studi di Politica Internazionale, Milan). Sugden R. and Wilson J. (2002), Economic Development in the Shadow of the Consensus: A Strategic Decision-Making Approach, Contributions to Political Economy, 21, pp. 111–34. Sylos Labini P. (2000), Sottosviluppo: una strategia di riforme (Editori Laterza, Roma and Bari). Tendler J. and Amorim M. (1996), Small firms and their helpers: lessons on demand, World Development, 24, pp. 407–26. Tokman, V. (1992), Beyond Regulation (PREALC-OIT, New York and Geneva). Trigilia C. (1992), ‘Italian Industrial Districts: Neither Myth Nor Interlude’, F. Pyke and W. Sengenberger (eds), Industrial Districts and Local Economic Regeneration (ILO, Geneva). UNDP (2000–2002), Human Development Report (UNDP, New York). UNIDO (1994), Proyecto de creación de redes de empresas; Project Report (UNIDO, Managua). Van Dijk M.P. and Rabellotti R. (eds) (1997), Enterprise Clusters and Networks in Developing Countries (Frank Cass, London). Van Dijk M.P. and Sandee H. (eds) (2002), Innovation and Small Enterprises in the Third World (Edward Elgar, Cheltenham). Williamson J. (1990), ‘The Progress of Policy Reform in Latin America’, in J. Williamson (ed), Latin American Adjustment: How Much Has Happened (Institute for International Economics, Washington, DC). Wade R. (1990), Governing the Market: Economic Theory and the Role of Government in East Asian Industrialization, Princeton University Press, Princeton. Womack J.P., Jones D.T. and Roos D. (1990), La macchina che ha cambiato il mondo (Rizzoli, Milano). Wong Chacon M. (2002), ‘Situacion actual y politica de apoyo a las pequeñas y medianas empresas industriales en Costa Rica’, in W. Peres and G. Stumpo (eds), Las pequeñas y medianas industrias en America Latina (CEPAL-Chile/Siglo XXI, Mexico). World Bank (1993), The East Asian Miracle: Economic Growth and Public Policy (New York, Oxford University Press). World Bank (2002), Review of Small Business Activities (World Bank, Washington, DC).
This page intentionally left blank
Index academic spin-offs 51–2 see also high-technology spin-offs agglomerations advantages 52 large firm-led 83, 84 simple 83, 84 agro industry, Nicaragua 73, 130 Argentina 37–40, 128–9 de-industrialisation 38–9 employment concentration 39–40 industrial development policies 128 manufacturing sector 38 privatisation 39 production system fracture 37 sales concentration 39–40 BCIE 72, 77 see also Central American Integration Bank (BCIE) Brazil 138 Brusco, S. 83, 134 business associations 111 business incubator programmes 79–80 Caldera, Norman 77, 138–9 Central American Integration Bank (BCIE) 72, 77 centrifugal forces 86 centripetal forces 86 Chile 13–29, 40–3, 129 active trade policies 18 competitiveness 43 conceptual framework of study 14–16 de-industrialisation 25 dualism 42–3 employment concentration 41 enterprises: reaction to crisis 21–4; recent evolution 21–3 food and beverages sector 23 growth rate 26
import growth 23 industrial composition of production 25–6 industrial policy: explicit approach 16–18; implicit approach 18–19; integrative strategy need 26–8 manufacturing sector 23–4 market liberalisation 129 production system fracture 21–4, 42–3 sales concentration 41 textiles and clothing sector 23 China entry into WTO 27, 36 Guang Dong clusters 103 cluster(s) concept 57 embryonic 5, 124, 136 hub-and-spoke 57, 83, 102 industrial districts vs 140 satellite 57, 83, 102, 105, 140 social strengths 66–9 state-anchored 57 trust in see trust see also survival clusters cluster development eclectic approach 6, 10, 103–4, 106, 131–2, 135; summary 85–9 empirical evidence: conclusions for policy-making 124–5; context 107–9; see also Forlì; Masaya; Sarchí new trends 106 policy-inducement approach see policy inducement social embeddedness approach 63–9, 78, 87–8 spontaneous approach 59–61, 62, 64, 78, 86–7 stage approach 84, 101–3, 134–5
155
156
Index
cluster governance 62 hierarchical 62 market 62 network 62 quasi-hierarchical 62 cluster policies 31–2, 33 clustering models 83 collective efficiency 6, 8, 60, 64, 86, 88, 94, 111, 117 conclusions 132–3, 134 collective goods 89 comparative advantage model 16–17, 39 competition, perfect 15 CONAPI 117 consumption cooperatives 138 cooperative system 138 CORFO 18–21, 26, 42, 129, 137 development programmes 19–20, 21 Costa Rica 46–8, 106, 117, 130–1 employment concentration 47 exports: promotion 46, 47; sectors involved 48 foreign direct investment 47 furniture cluster see Sarchí laws promoting SMEs 122, 131 software cluster 47 credit consortia 51, 60 credit cooperatives 138 dairy industry 74, 130 debt crisis 30, 35–6 diamond of competitiveness 73 dualism 14–15, 49, 128, 136–7 Chile 42–3 ex-post 128 Nicaragua 46 dynamism 123, 135 ECLAC, critique of neoliberal policies 25–6 eclectic approach 6, 10, 103–4, 131–2, 135 summary 85–9 economic democracy, promotion 3 economy of sacking 38 El Salvador 46 embryonic clusters 5, 124, 136
Emilia-Romagna regional development agencies 115 see also Forlì entrepreneurship 67 Forlì 111, 112–13 Italian industrial districts 93, 95 Nicaragua 76 Europe industrial restructuring 32–5, 50 SMEs: EU actions implemented 33, 34 firm creation 35 turnover growth 35 European Recovery Program 93 exclusion ex-ante 3 ex-post 3 export consortia 104 external economies 6, 60, 89–101, 107–24 family, as social cohesion factor 95–6 fishery industry 17–18 Nicaragua 73 flexible specialisation 6 food and beverages sector 23 Fordist model, breakdown 84–5, 93, 94, 95 foreign direct investment 44–5, 47, 69–70 forestry development 140–1 Forlì 8, 9, 108, 109–16, 135 collective efficiency 111 enterprises created by workers 113 entrepreneurship 111, 112–13 external economies 111–12, 113 growth rates 110–11 history 109–10 industrial areas around 115 information flow 112 innovation 112, 114 mode of basic features 108 policy inducement 114–15 retail shops investment 123 self-realisation 112–13 social cohesion 114 work relationships duration 114, 115
Index forums, local 79 free market policies 60 furniture clusters 8–9, 107–23 see also Forlì; Masaya; Matera; Sarchí global value chains 86, 98, 99 globalisation 128 governance cluster see cluster governance strategic role 6 within global value chains 98 Grabher, G. 139 Granada, leather industry 73–4 GTZ 72, 77, 122 Guatemala 46 Harris, J.R. 15 Heckscher-Ohlin model see comparative advantage model heterogeneity 137 high-technology policies 52, 131 see also innovation policies high-technology spin-offs 98–9, 100 hub-and-spoke clusters 57, 83, 102 Human Development Index (HDI) 36 import substitution industrialisation (ISI) 13, 21, 37 criticism 60 see also ISI INDE 72, 77 industrial development policy, systemic approach 5, 27, 50–3 industrial districts 57 clusters vs 140 competitive 83 development stages 84–5, 89–101 traditional 83 trust in see trust upgrading to 60 see also Italian industrial districts industrialisation, achieving 103 innovation policies 33, 52 INPYME 72, 73, 79, 123 institutions, local 66, 96 INTEL 47
157
ISI 13, 21, 37 criticism 60 see also import substitution industrialisation Italian industrial districts 7, 89–101 development stages 89–101; craft production to industrial concentration 92–4; industrial concentration to industrial districts 94–6; rural life to artisanal clusters 89–92; traditional IDs to competitive IDs 96–101 entrepreneurship 93, 95 external economies 94 groups of firms 99, 100 high-tech spin-offs 98–9, 100 immigration 100, 112 innovation 97 joint actions 94, 100 local institutions 96 policy inducement 91–2, 93, 99 self-realisation 91, 93, 95, 100 sharecropping 69, 88, 91 social cohesion 91, 93, 95–6, 100–1 specialised production 90–1 trust 95 see also Forlì; Matera; Third Italy Italy furniture clusters see Forlì; Matera industrial districts see Italian industrial districts Industrial Triangle 90, 93 laws promoting SMEs 33, 51, 91, 96, 99–100, 115; see also Sabatini Law merchant guilds 63, 64 southern 87 subcontracting relations 52 see also Third Italy Japan goodwill 63 subcontracting relations 52 joint actions 6, 60 joint-ownership firms 114 knowledge
97
158
Index
Lancashire 58 Latin America debt crisis 35–6 dualism 49 liberalisation 36, 48–9 production system fracture 4 see also individual countries Law of Failure 91 leather industry 73–4 see also tanning industry Léon, tanning industry 80 Lewis, W.A. 15, 128 linkage policies 4, 27, 33, 52–3, 132 manufacturing sector Argentina 38 Chile 23–4 Nicaragua 45 market failures 4–5, 136 Marshall, A. 58 Marshall Plan 93 Masaya 9, 108, 116–23, 135 collective efficiency 117–18 dynamism 123 enterprises created by workers 120 external economies 119 hammock cluster 73, 74 history 117 information flow 119 innovation 119, 120–1 investment fund access 121 mode of basic features 108 policy inducement 122–3 self-realisation 119–20 social cohesion 121–2 work relationships duration 121–2 Matera 110 merchant guilds 63, 64 Mexico 30 micro-credit 72 Microfer 72 MIFIC 72, 141 Mise-à-Niveau 51, 132 modern sector 15 multinational networks 98 National Institute for Supporting the Small and Medium Enterprises (INPYME) 72, 73, 79, 123
neoliberalism 16–17 networking 33, 50, 98 networking programmes 79 Nicamueble project 72 Nicaragua 43–6, 69–78, 130, 133 agro industry 73, 130 business incubator programmes 79–80 civil conflict 46, 75 clusters; conceptualisation 73; identified 71 dairy industry 74, 130 development policies 6–7 dualism 46 enterprises, numbers 43–4 entrepreneurship 76 exports 44–5 fishery industry 73 foreign direct investment 44–5, 69–70 furniture sector 44, 45, 72, 74; see also Masaya governance systems 76–8 informal sector 70 infrastructure investments 69 labour division/specialisation limits 74–6 law to promote SMEs 122 macroeconomic measures 70 manufacturing sector 45 market liberalisation 75–6, 77 military conflict 75, 118 networking programme 79 policy inducement 77–8 prizes 139 production chain weaknesses 45 production system fracture 43 regressive coalitions 77 sales concentration 44 self-realisation 76 shoe industry 74 SME sectors, strongest 75 SME support programmes 71–4 social cohesion 75–6 tanning industry 80 textile sector 73 tourism sector 73 trade deficits 70 see also Granada; Masaya
Index Nicaraguan Institute for Entrepreneurial Development (INDE) 72, 77 North Africa 51 obligational contracting Ossola Law 96, 115
63
PAMIC 71–2, 73, 117 see also Programme for Supporting Micro Enterprises (PAMIC) policy inducement 8, 61–3, 66, 78, 86–7 conclusions 132–3, 134 Forlì 114–15 Italy 86, 91–2, 93, 99 Masaya 122–3 Nicaragua 77–8 Sarchí 122 prizes 80, 139 production cooperatives 60, 138 production systems fracture within 3–4 Argentina 37 Chile 21–4, 42–3 Nicaragua 43 Programme of Development Services (PROSEDE) 72, 123 see also PROSEDE Programme for Supporting Micro Enterprises (PAMIC) 71–2, 73, 117 see also PAMIC promotion consortia 99, 111 PROSEDE 72, 123 see also Programme of Development Services regressive coalitions Nicaragua 77
61, 63
Sabatini Law 27, 51, 96, 115 Sarchí 8–9, 108, 116–23, 135 collective efficiency 117 dynamism 123 enterprises created by workers external economies 118–19 history 117 informal exchanges 117
120
159
information flow 118–19 innovation 118–19, 120–1, 123 investment fund access 121 joint actions 117 mode of basic features 108 policy inducement 122 retail shops investment 123 self-realisation 119 social cohesion 121–2 work relationships duration 121–2 Sassuolo, tile industry 99 satellite clusters 57, 83, 102, 105, 140 see also survival clusters Schmitz, H. 67, 138 Schumacher, E.F. 107 self-realisation 6, 67, 68–9, 88, 138 conclusions 133 Forlì 112–13 Italian industrial districts 91, 93, 95, 100 Masaya 119–20 Nicaragua 76 Sarchí 119 sharecropping 69, 88, 91 Sheffield 58 Sinos Valley 138 SME clusters see cluster(s) social capital 66 social cohesion 6, 8–9, 67–9, 88, 138 conclusions 133 Forlì 114 Italian industrial districts 91, 93, 95–6, 100–1 Masaya 121–2 Nicaragua 75–6 Sarchí 121–2 social embeddedness 6, 8, 63–4 conclusions 133, 134 too much vs too little dilemma 139 social mobility 128 software clusters 47 spin-offs 51–2, 98–9, 100 Stabilisation Pact 137 state-anchored clusters 57 structural adjustment programmes (SAPs) 13 structural policies 33, 51, 131
160
Index
subcontracting relations 52 survival clusters 1, 5, 9, 82, 105–6, 116–23, 124–5 conclusions for policy-making 124–5 in developing countries 106–9 policy keys for 101–4, 134 typologies 140 see also embryonic clusters; Masaya; Sarchí
see also Italian industrial districts tile industry 99 Todaro, M. 15 tourism sector 73 traditional SMEs 4, 15, 136 transnational powers 61, 63 trust 64–6, 67, 88, 133 ascribed 64, 65, 67 earned/calculative 65, 67, 68 systemic 65, 68
tanning industry 80 ‘tariff perversion’ 77 textiles and clothing sector Chile 23 Nicaragua 73 Third Italy 7–8, 58, 60, 69 arts and corporations 88 institutions 96 policy inducement 86 political leader responsiveness
UF (Unidad de Fomento) 137 United Nations Development Programme 36 United Nations Industrial Development Organization (UNIDO) 73–4, 79, 80, 118, 122–3, 132
92
wage gap, urban-rural 90–1 welfare corporatism 63