40516 D I R E C T I O N S I N D E V E LO P M E N T Trade Quality Systems and Standards for a Competitive Edge J. Luis Guasch, Jean-Louis Racine, Isabel Sánchez, and Makhtar Diop Quality Systems and Standards for a Competitive Edge Quality Systems and Standards for a Competitive Edge J. Luis Guasch Jean-Louis Racine Isabel Sánchez Makhtar Diop THE WORLD BANK Washington, DC © 2007 The International Bank for Reconstruction and Development/The World Bank 1818 H Street, NW Washington, DC 20433 Telephone 202-473-1000 Internet www.worldbank.org E-mail feedback@worldbank.org All rights reserved. 1 2 3 4 :: 10 09 08 07 This volume is a product of the staff of the International Bank for Reconstruction and Development / The World Bank. The findings, interpretations, and conclusions expressed in this volume do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions The material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. The International Bank for Reconstruction and Development/The World Bank encourages dissemination of its work and will normally grant permission to reproduce portions of the work promptly. For permission to photocopy or reprint any part of this work, please send a request with complete information to the Copyright Clearance Center Inc., 222 Rosewood Drive, Danvers, MA 01923, USA; telephone: 978-750-8400; fax: 978-750-4470; Internet: www.copyright.com. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2422; e-mail: pubrights @worldbank.org. ISBN-10: 0-8213-6894-X ISBN-13: 978-0-8213-6894-7 eISBN-10: 0-8213-6895-8 DOI: 10.1596/978-0-8213-6894-7 eISBN-13: 978-0-8213-6895-4 Library of Congress Cataloging-in-Publication Data Quality systems and standards for a competitive edge / José Luis Guasch . . . [et al.]. p. cm. -- (Directions in development) Includes bibliographical references and index. ISBN-13: 978-0-8213-6894-7 ISBN-10: 0-8213-6894-X ISBN-10: 0-8213-6895-8 (electronic) 1. Quality control. 2. Quality control--Standards. 3. Competition, International. 4. Poverty. I. Guasch, José Luis. TS156.Q366 2007 658.5'62--dc22 2006100909 Contents Foreword xvii Preface and Acknowledgments xix About the Authors xxi Abbreviations xxiii Chapter 1 Introduction 1 Globalization, Quality, and Standards 2 Defining Standards and Quality 9 The National Quality System: An Overview 9 Organization of the Book 11 Chapter 2 Economic Impact and Effect of Quality and Standards 15 Functions of Standards 15 Positive Economic Effects of Standards 17 Negative Economic Effects of Standards 24 The Impact of Standards on Economic Growth: Empirical Evidence 28 The Impact of Standards on International Trade: Empirical Evidence 32 Standards and Quality as an Entry Point for SMEs: The Path to Industrial Upgrading 41 v vi Contents Chapter 3 The Jurisdiction Issue: Roles of the Private and Public Sectors 51 Coordination Mechanisms for Technology Adoption 52 Government Involvement in Standardization 54 National Standards Bodies 56 Chapter 4 Evaluating Compliance with Standards: The Conformity Assessment Framework 63 Certification 65 Testing, Calibration, and Inspection 68 Accreditation 69 Metrology 74 Chapter 5 Opening Doors: Standards, Conformity Assessment, and International Integration 81 The WTO Agreement on Technical Barriers to Trade 83 International Coordination in Standardization 84 The Need for International Recognition of National Accreditation Systems 92 The Need for International Recognition of National Metrology Institutes 95 Chapter 6 The ISO 9000 Quality Management Standards 97 The Function of ISO 9000 97 ISO 9000 and Competitive Advantage 99 Internal Firm-Level Barriers to ISO 9000 Certification 105 National Factors Influencing the Diffusion of ISO 9000 114 Chapter 7 Standardization in Latin America 127 The Standards-Setting Institutions 128 The Performance of Standards Bodies 136 International Integration 139 Chapter 8 Certification in Latin America 149 Certification Activities 149 Contents vii The Supply of Certification Services 159 Possible Factors Influencing ISO 9000 Diffusion 162 Chapter 9 Accreditation in Latin America 171 Accreditation Institutions 171 Accreditation Activities 177 International Integration 180 Chapter 10 Metrology in Latin America 185 The National Metrology Institutions 185 Metrological Activities 189 International Cooperation and Recognition of Metrology Institutions 190 Chapter 11 Existing Policies and Support Programs 193 Brazil: Center for Quality, Safety and Productivity 193 Colombia: Quality and Environmental Management Program 194 Peru: CERTIpyme, PYMEvaluación, and PRECISO 196 Mexico: National Committee on Productivity and Technological Innovation 197 Chile: Chilean Economic Development Agency 198 Argentina: Standards and Quality Accreditation Program 199 Chapter 12 Policy Recommendations and Conclusion 201 General Principles for Implementation and Reform of National Quality Systems 201 Consolidating the National Quality System 203 Accreditation 204 Metrology 207 Standardization 209 Supporting the Demand for Certification, Testing, and Calibration 212 Conclusion 216 viii Contents Appendix A An Evaluation of Mexico's National Quality System 219 Standardization 220 Metrology and Calibration 221 Accreditation 224 Certification, Testing, and Inspection 227 Appendix B An Evaluation of Turkey's National Quality System 233 Certification, Testing, and Inspection 236 Accreditation 243 Metrology and Calibration 249 Standardization 252 Conclusion 258 Notes 261 References 269 Index 281 Boxes 1.1 Surviving Crises in Low-Quality Markets: Brazilian Footwear and Chilean Wines 5 1.2 Leather PDA Covers in Bolivia: Using Quality to Enter New Export Markets 7 1.3 The Components of a National Quality System 12 2.1 Quality Standards in the Ugandan Coffee Industry 20 2.2 Transatlantic Competition for First- and Second-Generation Mobile Phones 22 2.3 The Economic Value and Impact of Measurement Standards 26 2.4 Differing Standards and Trade Costs: The Case of Pallets 34 2.5 Food Safety Standards in the European Union 35 2.6 Standards and Quality as an Entry Point to Upgrading in a Horticultural Value Chain 44 2.7 Grapes in India: Changing Market Requirements for Standards and Quality 46 4.1 The Consorcio del Jamón Serrano Español's Quality Control 67 4.2 Regional Cooperation on Accreditation: The Southern African Example 72 Contents ix 4.3 Evolving Needs of Metrology in Health Care 77 5.1 The Struggle to Create a Single Global Standard for 3G Telephony 84 5.2 South Africa's Contribution to International Timber Standards 88 6.1 China and the Drive for Quality 106 6.2 Quality Adoption in the Indian Software Sector: Moving Up the Value Chain 121 B1 Conformity Assessment and Access to the EU Market 254 Figures 1.1 Ratio of World Exports to Global GDP 3 1.2 Share of Manufactures in Global Exports of Goods and Services 3 1.3 Price of Developing-Country Manufactured Exports Relative to Price of Developed-Country Exports of Machinery, Transport Equipment, and Services, 1975­95 4 1.4 Schematic Representation of a National Quality System 11 2.1 Contributions of Different Factors to Output Growth in the United Kingdom, 1948­2002 30 2.2 Long-Run Growth of Labor Productivity in the United Kingdom, 1948­2002 30 2.3 Average Productivity Gains Due to Adoption of Quality Standards 32 2.4 Effects of British Standards on UK Exports, Imports, and Trade Balance, 1985­91 38 3.1 The Centralized Approach to Standards Development 57 3.2 The Decentralized Approach to Standards Development 58 4.1 Traceability of Measurements 64 5.1 Number of Standards in Selected Industrial Countries and International Standards 86 5.2 The Internationalization of British Standards, 1948­2003 89 6.1 Global Adoption of ISO 9000 Standards 98 Top Ten Countries for ISO 9000 and ISO 9001 Certificates, with Mexico Comparator 107 6.2 Share of Brazilian Firms Reporting Important Certification Investments in Different Categories 112 6.3 Regional Shares of Worldwide ISO 9000 Certificates, 2004 114 x Contents 6.4 Top 10 Sectors Registering the Most Worldwide ISO 9000 Certificates, 2004 116 6.5 Top 10 Sectors Registering the Most Worldwide ISO 9000 Certificates, Standardized by Global Export Volume, 2004 117 7.1 Funding Sources of the National Standards Bodies 133 7.2 Budgets of the National Standards Bodies 134 7.3 Number of National Standards, 1990 and 2002 137 7.4 Number of New Standards, 2004, and Active Technical Committees, 2005 138 7.5 Price of National Standards Based on ISO 9001:2000 139 7.6 Number of Participant and Observer Memberships in ISO Technical Committees and Subcommittees, 2005 141 7.7 Participation in ISO Standards Development, 2004 142 7.8 Participation in COPANT Standards Development Activities 145 7.9 Number of TBT Notifications on Technical Regulations and Conformity Assessment Procedures, 2004 146 8.1 ISO 9000 Certificates in Latin America and the Caribbean, 1993­2004 150 8.2 Number of ISO 14000 Certificates in Latin America and the Caribbean, 1995­2004 150 8.3 Number of ISO 9001:2000 and ISO 14001 Certificates, 2004 151 8.4 Growth Rates of ISO 9000 Certificates in 2004 vs. Number of ISO Certificates; World Average Annual Growth Rates, 2001­04 152 8.5 Growth Rates of ISO 14000 Certificates in 2004 vs. Number of ISO Certificates; World Average Annual Growth Rates, 1998­2004 153 8.6 ISO 9000 Adoption Standardized by Value Added in Manufacturing and Services, 2004 154 8.7 Number of ISO 9000 Certificates and Value Added in Manufacturing and Services, 2002 155 8.8 Aggregate Number of ISO 9001 Certificates Standardized by Number of Establishments in 10 Major Manufacturing Sectors 156 8.9 Number of ISO 14000 Certificates and Value Added in Manufacturing and Services, 2002 156 Contents xi 8.10 Aggregate Number of ISO 14000 Certificates Standardized by Number of Establishments in 10 Major Manufacturing Sectors 157 8.11 Number of QS 9000 and ISO/TS 16949:2002 Certificates vs. Number of Enterprises in the Automobile Sector 158 8.12 Accredited Domestic Certification Bodies and Foreign Affiliates Offering Quality Management Systems Certification, 2005 159 8.13 Market Shares of Top Five Certification Bodies in Brazil, Argentina, and Mexico 160 8.14 Number of ISO 9001:2000 Certifications in 2004 Standardized by Number of Accredited Certification Bodies Operating Domestically in 2005 161 8.15 Number of Lead Auditors Certified for ISO 9001:2000 with IRCA in 2005, Standardized by Total Value Added in Manufacturing and Services in 2004 161 8.16 Exports to the EU and Japan as a Share of GDP and Certification Rates 162 8.17 Inward FDI Stock as a Share of GDP and Certification Rates 163 8.18 Exported Goods in Global Quality­Focused Sectors as a Share of GDP 164 8.19 Regional SME Sector Share of Manufacturing Employment, 1990­99 Averages 165 8.20 SME Sector Share of GDP and Certification Rates 166 8.21 ISO 9001 Adoption and Firm Size in Mexico 166 8.22 Average Business Environment Indicator, by Region 167 8.23 Business Environment Indicator and Certification Rates, by Country 168 8.24 Average Institutional Development Indicator, by Region 168 8.25 Institutional Development Indicator and Certification Rates, by Country 169 8.26 Secondary School Enrollment and Certification Rates 169 9.1 Accreditation Body Personnel, 2005 174 9.2 Number of Qualified Lead Assessors, 2005 175 9.3 Number of Qualified Technical Experts, 2005 176 9.4 Minimum Time Required for the Laboratory Accreditation Process 177 xii Contents 9.5 Number of Accreditations of Management System Certification Bodies 178 9.6 Number of Accreditations of Inspection Bodies 179 9.7 Number of Accreditations of Testing and Calibration Laboratories 180 9.8 Proportion of Foreign Affiliates of Quality System Management Certification Bodies Accredited by the National Accreditation Body 181 10.1 Number of Measurement Areas in the National Metrology Laboratories 189 10.2 Number of Calibration Services Offered by NMIs, Standardized by Manufacturing Value Added, 2005 190 10.3 International Laboratory Comparison Activities, Absolute Number and Standardized by Number of Measurement Fields Offered by the NMI, 1995­2005 191 A1 Metrology Services Chain in Mexico 222 A2 Concentration of Calibration Labs in Mexico City, Jalisco, and Nuevo León 223 A3 Concentration of Testing Labs in Mexico City, Jalisco, and Nuevo León 229 A4 ISO 9000 Certificates, 2004 231 A5 Mexican ISO 9000 Certificates and Share of Latin American Total 231 A6 Mexican ISO 9000 Certificates and Share of World Total 232 B1 Contribution of Investment Climate Variables to Average Employment by Turkish Firms 234 B2 Contribution of Investment Climate Variables to a Turkish Firm's Probability of Becoming an Exporter 235 B3 Number of ISO 9001 Certificates as a Percentage of the Number of Firms in Manufacturing, Wholesale and Retail Trade, Business Services, and Real Estate in Selected OECD Countries 237 B4 Number of ISO 9001 Certificates vs. Number of Enterprises in Manufacturing, Wholesale and Retail Trade, Business Services, and Real Estate in OECD Countries 237 B5 Number of ISO 9001 Certificates as a Percentage of the Number of Manufacturing Firms in OECD Countries 238 Contents xiii B6 Number of ISO 9001 Certificates vs. Number of Manufacturing Firms in OECD Countries 238 B7 Growth of the Number of ISO 9000 Certificates in Turkey 239 B8 Number of ISO 9001 Certificates per Number of Establishments in Turkey, Spain, and Romania in Different Manufacturing Sectors 240 B9 Number of National Notified Bodies Recognized by the EU in Selected Countries, February 2006 241 B10 Number of Organizations Accredited by the National Accreditation Body in Selected Countries, March 2006 243 B11 Number of Organizations Accredited by the National Accreditation Body in Selected Countries, Standardized by Value Added in Manufacturing and Services, March 2006 243 B12 Number of Accreditations Granted Each Year by TÜRKAK 244 B13 Number of Accreditations per Permanent Staff Person in the Accreditation Body, Selected Countries 246 B14 Budget of the National Accreditation Body, Selected Countries 246 B15 Average Cost of an Accreditation, Selected Countries 247 B16 Number of Calibration Activities Relative to Manufacturing Value Added, and Number of Calibration Customers Relative to Number of Manufacturing Firms, in NMIs of Selected Countries, 2005 248 B17 Number of Key and Supplementary Measurement Comparisons by Signatories of the CIPM MRA 250 B18 Number of National Standards, Selected Countries 251 B19 Standards Stock vs. Value of Trade in Turkey 252 B20 Participant and Observer Memberships in ISO Technical Committees and Subcommittees 257 Tables 1.1 Share of Manufactures in Total Merchandise Exports, by Region percent 4 2.1 The Economic Effects of Standards According to Their Function 26 xiv Contents 2.2 Partial Production Elasticities in Selected European Countries 31 2.3 Empirical Results on the Effect of Idiosyncratic and Harmonized National Standards on Trade 37 5.1 Duplication of Testing Procedures to Meet Foreign Requirements 82 5.2 Members of IAF and ILAC Mutual Recognition Agreements, 2006 94 5.3 Firms with Products Subject to an MRA 94 5.4 Perceived Export Advantage of Participating in an MRA with an Export Destination Country 94 6.1 Summary of Studies Evaluating the Effect of Certification on Firm Performance 108 6.2 Firms Using External Consulting Services for Certification in Two Brazilian Surveys 109 6.3 Total ISO 9000 Implementation and Certification Costs 110 6.4 Typical Consulting Costs for Firms Seeking Certification, Argentina, 2005 112 6.5 Typical ISO 9000 Registration Costs in Latin America, 2005 113 7.1 Characteristics of the National Voluntary Standards Bodies 128 7.2 Activities of Standards Bodies 130 7.3 Proportion of National Standards Body Staff Involved in Standardization Activities (percent) 130 7.4 Private Sector Participation in the Governance of Standards Bodies 131 7.5 Staff of the National Standards Bodies 135 7.6 Share of Mandatory Standards in Total National Standards, 1990 and 2002 (percent) 136 7.7 Growth of the Standards Stock between 1990 and 2002 (percent) 137 7.8 Membership in International Standards Bodies 140 7.9 Evolution of the Share of National Standards Based on International Standards (percent) 140 7.10 Membership in Organizations and Preferential Trade Agreements with Standards-Related Measures 143 7.11 Principal Regional Agreements and Organizations with Standards-Related Measures 143 Contents xv 9.1 Legal Status of National Accreditation Bodies 172 9.2 Activities of National Accreditation Bodies 173 9.3 Private Sector Participation in the Governance of Accreditation Bodies 173 9.4 Implementation of ISO Guides and Standards for Accreditation Bodies 182 9.5 Principal Regional and International Organizations Operating Mutual Recognition Agreements for Accreditation Within, or in Cooperation with, Latin American Countries 183 9.6 Membership in International Organizations and Mutual Recognition Agreements 184 10.1 Organization of Scientific and Legal Metrology Institutions 186 10.2 Activities of the National Scientific Metrology Institutions 187 10.3 Accreditation of the National Metrology Laboratories 188 10.4 Membership in International Organizations 192 11.1 CYGA: Participation and Subsidies 195 11.2 Distribution of Enterprise Size in Two COMPITE Programs, 2004 197 11.3 COMPITE Programs: Participation and Subsidies 198 A1 Official Mexican Standards by Agency, June 2005 220 B1 Organization of Accreditation in Selected Countries 245 B2 Governance of the National Voluntary Standards Bodies, Selected Countries 254 B3 Turkey's National Quality System: Deficiencies, Ongoing Actions, and Recommendations 259 Foreword During the last decade, economic growth in the fastest-growing develop- ing countries, particularly China and India, has been accompanied by their rapid integration into world markets. This has created new oppor- tunities for all countries but also new competitive pressures, and it has placed new demands on policies to support trade development. As increased competition among developing countries in labor- intensive manufactures erodes economic returns, higher-quality markets and high-value goods are increasingly important to maintaining dynamic competitive advantage. Globally integrated production networks, typi- cally governed by buyers from developed nations, have raised competi- tiveness to the top of developing countries' policy agendas. Countries need to offer the high-quality products demanded by consumers and global supply chains and deliver them to markets to meet just-in-time production schedules. Against all major competitiveness rankings, Latin America has lost ground. Growth has been mediocre, especially if compared with the growth of most Asian economies. This has led to lively policy debates about the path back to high growth, from which has emerged a growing consensus that better and more effective coordination between the public and private sectors is required--and that this needs to be complemented by institutional and microeconomic reforms. For Latin America, a new trade and competitiveness agenda has three key elements: (a) upgrad- xvii xviii Foreword ing the value of traditional exports and diversifying exports away from primary products; (b) removing constraints to "speed-to-market" goods to fully exploit proximity to the United States and other markets; and (c) enhancing productivity growth to offset rising wages while boosting the development of technological capabilities and skills. But diversifying and upgrading exports--whether manufactured products within large supply chains or high-value food products--means developing quality and standards. It also means addressing weaknesses in logistic, financial, and administrative support services. These are not easy tasks, and they present a major challenge both for policy makers in the region and for development partners. This book responds to this challenge by providing a comprehensive account of quality systems for private sector development: what works on the ground and what doesn't, and why. It explains why quality and standards matter for export growth, productivity, industrial upgrading, and diffusion of innovation, all central ingredients in improving economic growth and generating real gains in poverty reduction. The book exam- ines the diversity of institutions, linkages, and arrangements involved in quality systems, identifying success factors and obstacles in the quality strategies of particular countries. A portion of the volume is devoted to experiences in Latin America and the Caribbean, a region with a great deal at stake in the drive to improve quality. (One of the authors, J. Luis Guasch, is one of the World Bank's leading experts on private sector development in Latin America.) Policy makers in Latin America and throughout the developing world will find Quality Systems and Standards for a Competitive Edge to be a valuable tool for meeting the challenges of building trade competitiveness in the new global economy. Danny Leipziger Vice-President Poverty Reduction and Economic Management Network The World Bank Preface and Acknowledgments Chi fa fala, e chi non fa sfarfalla. "Those who act make mistakes, and those who do nothing really blunder." Lackluster growth in many middle-income countries has led to the real- ization that while macroeconomic stabilization and liberalization are necessary, they are not sufficient to boost growth in developing countries. As the emphasis has shifted to microeconomic reforms, most developing countries have launched broad trade and competitive initiatives in search of that elusive goal, sustained high growth. Yet most of those initiatives have fallen short of providing the necessary infrastructure for the devel- opment and adoption of quality and standards that are critical for secur- ing access to external markets and supporting increased competitiveness. There is limited awareness of the benefits to be gained from systemati- cally producing high-quality products and adopting appropriate, interna- tionally expected norms and standards. These observed shortcomings provided the motivation for undertaking this book. As we began to help countries establish and improve national quality systems to support their trade and export development initia- tives, we saw the need for an integrated guide that would bring together in a coherent way all the different components that countries need to consider in building a national quality system. We hope that this book will raise awareness of the impact of quality and standards, place quality xix xx Preface and Acknowledgments issues high on the strategic agendas of firms and policy makers, and prove useful to reform-minded governments as they seek to improve, modern- ize, and implement their national quality systems. We are grateful to many people for encouragement in writing this book, particularly policy makers, development practitioners, and firms and private sector organizations in developing countries. They had a clear vision of the need for an integrated reference work to create broad aware- ness of the quality and standards issue and help promote the systematic adoption of good quality practices. Our dialogue with these stakehold- ers greatly enriched our analysis. Their perceptions and suggestions were lucid, and a number of them provided vivid examples of the challenges as well as illuminating examples of success. At the World Bank, we are most grateful to Danny Leipziger for his strong support and encouragement from the start; to Enric Banda, Carl Dahlman, Marianne Fay, Pablo Fajnzylber, Esperanza Lasagabaster, José Guilherme Reis, and Joseph Stiglitz, who provided comments and sug- gestions on the evolving manuscript; and to Amelia Camino-Forsyth, Thomas Haven, and Joy Troncoso, who very generously assisted us in the production of this book. The authors alone take responsibility for the content of the book and the views expressed here, which do not necessarily reflect the views of our colleagues in the World Bank Group. About the Authors J. Luis Guasch, a Spanish national, is currently Senior Regional Advisor for the Latin America and Caribbean Region of the World Bank, Wash- ington, DC, responsible for the areas of competitiveness, regulation, infra- structure, innovation, and technological development. In that role he has assisted a number of countries with the design and implementation of their national quality systems. A professor of economics at the Univer- sity of California, San Diego, since 1980, he has published extensively in leading economic journals and has authored or coauthored several books, including Managing the Regulatory Process: Design, Concepts, Issues, and the Latin America and Caribbean Story; Closing the Gap in Education and Technology, and Granting and Renegotiating Infrastructure Concessions: Doing it Right. He holds a Ph.D. in Economics from Stanford University, California, USA, and an engineering degree from the Polytechnic Univer- sity of Barcelona, Spain. Jean-Louis Racine is a policy consultant at ANGLE Technology Group where he focuses on devising technology-based economic develop- ment strategies for regional and national governments. Prior to joining ANGLE, he worked with the United Nations Industrial Development Organization (UNIDO), the Organization of American States (OAS), and the World Bank in several countries to promote innovation, technol- ogy diffusion, and private sector competitiveness. Mr. Racine's engineer- ing and policy experience spans a wide range of sectors in technology xxi xxii About the Authors development, from traditional (electric power generation and hydrau- lic systems) to cutting-edge (advanced medical devices, aerospace, and robotics). He holds a PhD in Mechanical Engineering from the Univer- sity of California, Berkeley (USA), and a Master of International Affairs in Technology Policy for Economic Development from Columbia Uni- versity, New York City (USA). Isabel Sánchez García is a consultant at the World Bank, specializing in the implementation of innovation and quality programs in Latin America and in East and Central Europe. She holds a PhD in Economics from the University of California, San Diego, and was Professor of Economics at the University of Rochester (Rochester, New York, USA) and at the Univer- sidad Carlos III (Madrid, Spain). As Chief of Staff for Spain's Vice-Min- ister for Science of Technology, she participated actively in the creation of that ministry and helped develop new policies supporting research and development and technology adoption. She has also been a member of Spain's Antitrust Commission and Energy Regulatory Agency. Makhtar Diop, a Senegalese national, is Sector Director for Finance, Private Sector and Infrastructure, in the Latin America and Caribbean Region at the World Bank, leading the Bank's assistance to Latin Ameri- can countries in strengthening and deepening their financial sectors, improving their business environments, and increasing the financing and performance of their infrastructure sectors. He has also served as country director for Kenya, Eritrea, and Somalia. Prior to the World Bank, he worked at the International Monetary Fund and held various positions in banking and finance. He has also served as Senegal's Minister of Economy and Finance. Mr. Diop undertook his PhD studies in economics at the University of Nottingham (UK) and holds a Master of Science degree in Economics from the University of Warwick (UK) and a Master's degree in Finance from Ecole Supérieure Libre des Sciences Commerciales Appliquées, Paris (France). Abbreviations ABNT Associação Brasileira de Normas Técnicas Brazilian Technical Standards Association AENOR Asociación Española de Normalización y Certificación Spanish Association for Standardization and Certification AMPS Advanced Mobile Phone System ANCE Asociación de Normalización y Certificación Association for Standards and Certification (Mexico) ANSI American National Standards Institute APEC Asia-Pacific Economic Cooperation APLAC Asia-Pacific Laboratory Accreditation Cooperation ASEAN Association of Southeast Asian Nations BIPM Bureau International des Poids et Mesures International Bureau of Weights and Measures CAN Comunidad Andina Andean Community CDMA Code Division Multiplexing Access CEM Centro Español de Metrología Spanish Metrology Center CEN Comité Européen de Normalisation European Committee for Standardization CENAM Centro Nacional de Metrología National Metrology Center (Mexico) xxiii xxiv Abbreviations CENELEC Comité Européen de Normalisation Electrotechnique European Committee for Electrotechnical Standardization CFE Comisión Federal de Electricidad Federal Electricity Commission (Mexico) CGPM Conférence Générale des Poids et Mesures General Conference on Weights and Measures CIPM Comité International des Poids et Mesures International Committee of Weights and Measures CIS Commonwealth of Independent States CITAC Cooperation on International Traceability in Analytical Chemistry CMM Capability Maturity Model CODEX Codex Alimentarius Commission COMPITE Comité Nacional de Productividad e Innovación Tecnológica National Committee on Productivity and Technological Innovation (Mexico) CONACYT Consejo Nacional de Ciencia y Tecnología National Council on Science and Technology (Mexico) COPANT Comisión Panamericana de Normas Técnicas Pan American Standards Commission CORFO Corporación de Fomento de la Producción Chilean Economic Development Agency CSRM Committee on Standards-Related Measures (of NAFTA) CYGA Calidad y Gestión Ambiental Quality and Environmental Management Program (Colombia) DGN Dirección General de Normas General Bureau of Standards (Mexico) DTI Department of Trade and Industry (UK) EA European Co-operation for Accreditation EFTA European Free Trade Association EMA Entidad Mexicana de Acreditación Mexican Accreditation Body EMS environmental management systems EN European standard ENAC Entidad Nacional de Acreditación National Accreditation Body (Spain) ETSI European Telecommunications Standardization Institute Abbreviations xxv EU European Union EUROMET European Collaboration in Measurement Standards FCC Federal Communications Commission FDI foreign direct investment FOMIPYME Fondo Colombiano para la Modernización y el Desarrollo Tecnológico de las Micro, Medianas y Pequeñas Empresas Colombian Fund for the Modernization and Technological Development of Micro, Small and Medium Enterprises GDP gross domestic product GNP gross national product GSM Global System for Mobile Communications HACCP Hazard Analysis and Critical Control Point IAAC InterAmerican Accreditation Cooperation IAF International Accreditation Forum ICONTEC Instituto Colombiano de Normas Técnicas y Certificación Colombian Institute for Technical Standards and Certification IEC International Electrotechnical Commission ILAC International Laboratory Accreditation Cooperation INDECOPI Instituto Nacional de Defensa de la Competencia y de la Protección de la Propiedad Intelectual National Institute for the Defense of Competition and the Protection of Intellectual Property (Peru) INEN Instituto Ecuatoriano de Normalización Ecuadorian Standardization Institute ININ Instituto Nacional de Investigaciones Nucleares National Nuclear Research Institute (Mexico) INMETRO Instituto Nacional de Metrologia, Normalização e Qualidade Industrial National Institute of Metrology, Standardization and Industrial Quality (Brazil) INN Instituto Nacional de Normalización National Standardization Institute (Chile) INTI Instituto Nacional de Tecnología Industrial National Institute of Industrial Technology (Argentina) IRAM Instituto Argentino de Normalización y Certificación Argentine Institute of Standardization and Certification IRCA International Register of Certificated Auditors ISO International Organization for Standardization xxvi Abbreviations IT information technology ITU International Telecommunication Union KAB Korea Accreditation Board KAS Korea Accreditation System KATS Korean Agency for Technology and Standards KOLAS Korea Laboratory Accreditation Scheme KOSGEB Small and Medium Industry Development Organization (Turkey) KRISS Korea Research Institute of Standards and Science LAC Latin America and the Caribbean Mercosur Mercado Común del Sur Southern Common Market MLA multilateral recognition arrangement MRA mutual recognition arrangement or agreement NAFTA North American Free Trade Agreement NMI national metrology institute NMT Nordic Mobile Telephone NWML National Weights and Measures Laboratory (UK) OAA Organismo Argentino de Acreditación Argentine Accreditation Bureau OAE Organismo de Acreditación Ecuatoriano Ecuadorian Accreditation Body OECD Organisation for Economic Co-operation and Development OHSAS Occupational Health and Safety Assessment Series OIML Organisation Internationale de Métrologie Légale International Organization of Legal Metrology PABCO Planteles de Animales Bajo Certificación Oficial Animal Premises Under Official Control PAC Pacific Accreditation Cooperation PDA personal digital assistant PEMEX Petróleos Mexicanos Mexican Petroleum Company PNAC Programa Nacional de Aseguramiento de la Calidad National Program for Quality Assurance (Colombia) ProCal Programa de Normas y Acreditación de la Calidad Standards and Quality Accreditation Program (Argentina) QMS quality management systems Abbreviations xxvii QSP Centro da Qualidade, Segurança e Produtividade Center for Quality, Safety and Productivity (Brazil) R&D research and development RTA regional trade agreement SADC Southern African Development Community SADCA Southern African Development Community Accreditation SADCAS Southern African Development Community Accreditation Service SEI Software Engineering Institute SENA Servicio Nacional de Aprendizaje National Training Service (Colombia) SERNAC Servicio Nacional del Consumador National Consumer Service (Chile) SIC Superintendencia de Industria y Comercio Superintendency of Industry and Commerce (Colombia) SIM Sistema Interamericano de Metrologia InterAmerican Metrology System SME small and medium enterprise SPG standards preparatory group (Turkey) SPI software-process improvement STQC Standardization, Testing, and Quality Certification TBT Technical Barriers to Trade (of the WTO) TC Technical Committee (of the ISO) TNC transnational corporation TQM total quality management TSE Turkish Standards Institution TSG Traditional Specialty Guaranteed TÜBITAK Scientific and Technical Research Council of Turkey TÜRKAK Turkish Accreditation Agency UK United Kingdom UKAS United Kingdom Accreditation Service UME National Metrology Institute (Turkey) UMTS Universal Mobile Telephone System UNE-EN ISO Spanish national standard WTO World Trade Organization Note: All dollar amounts are U.S. dollars unless otherwise indicated. C H A P T E R 1 Introduction In the past few decades, changes in global trade flows have enhanced the role of quality and standards in economic development. Global trade flows have experienced changes with respect to their magnitude, nature, composition, and patterns. While the relative magnitude of trade flows has continuously increased, the nature of trade has been radically altered by the emergence of closely integrated global production networks. Shifts in trade composition have given a greater importance to manufacturing. Man- ufacturing has replaced other industries to become the most important export sector in many developing countries, if not the most promising export sector. Patterns of trade have significantly changed with the entry of developing countries in export markets for labor-intensive manufactures. However, an expanded presence of developing countries in global markets has not always coincided with rising economic returns, as intense com- petition has eroded the profitability of low-cost manufactures. All these changes have amplified the importance of standards and quality in the world economy. Higher-quality markets have not been subject to falling profitability and present a sustainable alternative to price competition. In this book we argue the urgent need for countries to move forward aggressively on adopting and upgrading quality and standards. The focus is on developing countries, many of which are lagging in the race for standards adoption, and within that category, on middle-income countries. The book analyzes the economic impact of quality and standards on economic growth, on international trade, and as an entry point for the industrial upgrading and mainstreaming of small enterprises. It offers detailed guidelines for the cre- ation of national quality systems that can effectively support the use and 1 2 Quality Systems and Standards for a Competitive Edge adoption of standards. It describes the optimal structure for a national qual- ity system, evaluates the precise roles of the public and private sectors, and proposes best-practice guidelines and norms for these roles. It also addresses the financing issue, including the extent of and rationale for targeted subsi- dies, and questions of jurisdiction. Special emphasis is given to international integration through mutual recognition agreements that enhance access to external markets--a key objective for developing countries. The first part of the book is generic and normative, providing empiri- cal evidence and guidelines for reform. The second part describes and evaluates the performance of a number of Latin American countries with respect to various components of the national quality system. Latin America offers several good examples of countries that are moving ahead on quality and standards. Moreover, most countries in the region are choosing exports as the engine for growth, and for that to be successful, quality and standards are essential. Finally, the appendix presents case studies of Mexico and Turkey to illustrate the methodology of analysis for making a diagnosis and appropriate recommendations. Globalization, Quality, and Standards Globalization, characterized by significant increases in global flows of information, ideas, production factors, technology, and goods, has brought about a deep and pervasive integration of the world economy. Traditional barriers to trade no longer play as important a role as before, in part because of advances in information, communication, and transportation technologies, but also because of a new international environment of trade policy. Successive rounds of international agreements have systemi- cally reduced trade barriers in rich countries, and developing countries have followed suit, inspired by the success of several East Asian econo- mies. Latin American and Caribbean countries have not been excluded from these trends, and policies aimed at outward-oriented growth have gradually overturned decades of protection and subsidies in this region. These changes have affected the magnitude of global trade, whose importance in the global output has grown rapidly in the past three decades. World exports as a proportion of global gross domestic product (GDP) have risen from a little more than 10 percent in 1970 to close to 30 percent in 2005 (figure 1.1). Manufacturing now accounts for a much greater share of trade than two decades ago, around 60 percent in 2002 versus 48 percent in 1980 (figure 1.2).1 Many developing countries have made major gains in their Introduction 3 Figure 1.1 Ratio of World Exports to Global GDP 30 25 GDP % as 20 ts expor 15 10 1970 1975 1980 1985 1990 1995 2000 2005 Source: IMF 2005. share of world trade in manufactures. These countries have increased their collective share of global manufactures exports from 10 percent in 1980 to almost 30 percent in 2002 (UNCTAD 2004). In Latin America, for instance, manufactures rose to become the largest export category in the late 1990s (table 1.1). Intense competition in global markets for cheap low-quality manufac- tured goods has eroded economic returns in developing countries.As a result, the prices of manufactures exported by developing countries continue to fall in relation to prices of the skill-intensive manufactured goods exported by developed countries. This has provoked a deterioration of terms of trade in manufacturing in developing countries. Figure 1.3 shows the evolution of a Figure 1.2 Share of Manufactures in Global Exports of Goods and Services 65 ts 63 61 expor 59 of % 57 as 55 es 53 tur 51 49 manufac47 45 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 Source: UNCTAD 2002. 4 Quality Systems and Standards for a Competitive Edge Table 1.1 Share of Manufactures in Total Merchandise Exports, by Region percent Region 1980­1983 1989­1992 1999­2003 All developing countries 31.4 55.7 68.1 Latin America and the Caribbean 32.6 35.9 56.6 Africa 12.7 15.7 23.0 West Asia 16.8 17.7 21.0 East and South Asia 54.9 76.5 84.8 Source: UNCTAD 2005. terms-of-trade index from 1975 to 1995. The index considers the price of manufactured exports from developing countries relative to prices of manu- facturing inputs and equipment exported by developed countries. These deteriorating terms of trade can have a sizeable impact on economies where manufactured exports represent a large share, such as in Latin America, where the manufacturing sector accounted for almost 60 percent of exports in 2003. In 2004 manufactures contributed to approximately a 2 percent decrease in terms of trade for both Peru and Mexico (UNCTAD 2005). Competition on quality can lead to a more sustainable competitive advantage than competition on price alone. High-quality market seg- ments are not as vulnerable to declining terms of trade. There are many examples in which producers of high-quality goods have been able to withstand an industry crisis much more easily than their low-quality counterparts (box 1.1). Figure 1.3 Price of Developing-Country Manufactured Exports Relative to Price of Developed-Country Exports of Machinery, Transport Equipment, and Services, 1975­95 120 110 100 ratio e 90 pric 80 70 60 1975 1980 1985 1990 1995 Source: Wood 1997. Note: 1980 = 100. Introduction 5 Box 1.1 Surviving Crises in Low-Quality Markets: Brazilian Footwear and Chilean Wines Brazil's footwear industry went into deep distress in the 1990s as a result of the entry of Chinese producers with lower labor costs. Brazil reduced its share of the global footwear market from 7.6 percent in 1985 to 4.1 percent in 1990 and 3.8 percent in 1998. Meanwhile, China's share increased from 1.4 percent in 1985 to 7.2 percent in 1990 and 23.3 percent in 1998. The Brazilian footwear industry experienced dramatic losses of pro tability and was squeezed out of its main segment of cheap, standardized leather shoes. To survive, Brazilian enterprises were forced to nd new strategies to produce higher-quality shoes and open up new markets and marketing channels. Some rms sought technical support from their suppliers to improve their productive processes. Others increased their quality by seeking the services of local testing institutions and investing in personnel training. Many of the rms decided to cooperate on the creation of a cluster brand with local design and quality requirements. Chile's wine industry su ered a similar shock. In the late 1990s, after a decade of soaring exports, prices of "popular premium" and "premium" wines began a decline triggered by excess supply in world markets. These were Chile's main wine export segments, and the average price of wine exports decreased from a high of $2.15 per liter in 1998 to $1.70 per liter in 2003. Just as Brazil's initial labor cost advantage did not guarantee the permanent pro tability of its foot- wear industry, natural conditions favorable to wine production in Chile were not su cient to ensure the sustainable growth of that country's wine industry. In response to this crisis Chile's wine industry has been trying to shift its focus to quality and product di erentiation. Some of the struggling rms have increased their pro ts by participating in a government program called PROFO (Proyectos de Fomento), which creates small networks of rms that are eligible for subsi- dies on collaborative projects. PROFOs have proven useful in upgrading soft technologies such as those in the organization and management of production processes and in quality control. Sources: Bazan and Navas-Alemán 2001; Schmitz 1999; Giulani and Bell 2004. 6 Quality Systems and Standards for a Competitive Edge In light of the increased competition associated with globalization, developing countries seeking sustained growth need to free themselves from dependence on primary products and diversify into manufacturing exports, whose value added translates into wealth. A poor investment climate and small market hampers that development, but an even big- ger obstacle is lack of the often sophisticated standards required to enter global trade markets. It is the systematic use and adoption of quality standards and technology that allows developing-country producers to close the gap with the leading countries. Increasingly, developing countries, particularly small ones, are adopting export-led growth strategies. Their relatively small internal markets and limited purchasing capabilities force them to look for markets abroad. Countries are liberalizing trade and aggressively signing bilateral free trade treaties to secure favorable access to their products. But while a free trade treaty is essential, it is not sufficient, as countries also have to offer the right products for sale. This is where quality and standards come to bear. To access global markets, independent manufacturers from develop- ing countries must join global production networks typically governed by transnational corporations (TNC) and global buyers from developed countries.The multiplicity of global sourcing arrangements, the numerous horizontal and vertical networks, and the dynamic nature of these net- works give a critical importance to standards in global production systems. Trade increasingly involves subcomponents and services, making it much more complex than the arms-length relationships of the past. TNCs and global buyers impose standards on their suppliers to ensure compatibility between products and processes throughout their global chains. They also use standards to ensure that they can satisfy the high customer require- ments of developed-country markets with respect to quality as well as environmental and social impact. Firms that are unable to meet these standards find themselves excluded from global markets, while those that meet them may be able to profit from new opportunities (box 1.2). Adoption of the right quality and standards can also have a significant impact on poverty and inequality. The entry of small firms into supply chains, supplier networks, and export consortia entails the use, adoption, and certification of quality standards. Increasingly, partners, networks, customers (in the case of final goods), and major firms require potential small suppliers to guarantee appropriate standards and quality in goods and services. The impact on earnings and growth and on employment for those small firms or farmers can be quite significant. There are many examples (in Peru, Brazil, Ecuador, Guatemala, Bolivia, and elsewhere) Introduction 7 Box 1.2 Leather PDA Covers in Bolivia: Using Quality to Enter New Export Markets In the mid-1990s personal digital assistants (PDA) made their successful entry into the mass consumer market thanks to novel designs such as the Palm Pilot. PDAs have since evolved from simple electronic organizers and address books to powerful handheld computers and communication devices. They are used throughout the world, with a global unit shipment of 15 million in 2005. When PDAs rst became popular, a leather tannery owner in Cochabamba, Bolivia, saw a unique opportunity for a new export market. With his industry experience, insight from trade shows, and export knowledge, he formed a new company called Macaws SRL.The company was to concentrate on leather cases for PDAs, in addition to exporting a broad range of leather products. Today the rm exports 150,000 PDA cases per month to markets in Europe, North America, and South America. These include large lots for PDA manufacturers as well as individual or customized items ordered over the Internet. Macaws has grown from 25 to 300 employees, with enough orders to hire another 150. Macaws attributes its success to the design and functionality of its prod- ucts, but also to the high quality of the workmanship. In fact, the rm's quality management system is based on the ISO 9001 standard. Macaws places heavy emphasis on training its new employees to make high-quality products. Initial employee training takes approximately two months. Production of leather cases is labor-intensive, and labor accounts for 60 per- cent of costs. Macaws's orientation toward the higher-quality market segments is important because it would be di cult to compete on price alone against countries where low-cost labor is abundant, such as China. In fact, some of the Macaws PDA cases are priced at around $40, the upper end of the price range, while competitors' products sell for $20 or $30. The rm's ability to expand its production, enter new markets with high prices, and pay its employees an aver- age of $650 per month is a testimony to the many advantages of competing on quality in export markets. Sources: World Bank 2005a; Macaws SRL Web site. 8 Quality Systems and Standards for a Competitive Edge of small firms, microenterprises, artisans, and farmers stepping up from poverty and subsistence-level production by adopting standards, or by making minor adjustments in their products, that have opened new mar- kets or facilitated linkages with major firms. In some cases they have tripled their earnings and output (World Bank 2004, 2006a). The use of quality standards is critical for the mainstreaming of small and medium enterprises (SMEs) and can also serve as a powerful incentive for the formalization of informal firms. However, upgrading quality and standards is not a simple matter for many firms, particularly SMEs in developing countries. They often lack the knowledge, expertise, and resources to adopt and implement high quality and standards. They may lack the information to make educated choices and often fail to understand the impact of their choices. Gov- ernments in developing countries can help by enabling the private sec- tor to learn about and adopt the quality and standards demanded by international markets. The main argument for government support in this regard rests on the "public good" aspect of standards--the associated externalities, critical mass issues, and network goods characteristics of standards--and the fact that the market is unlikely to produce the opti- mum standards and levels of adoption. Governments can compensate for market failures by creating programs and policies that promote the widespread use and diffusion of standards. It is quite common for govern- ments to develop and offer programs and training, with direct or indirect subsidies, to help firms adopt quality standards. But it is often the case that only formal firms can qualify for this assistance. In summary, in the past three decades, the rapid growth of global trade and the increasing share of manufactures in global trade has made quality a critical factor in export success. Low entry barriers in labor-intensive industries and new trade policies present export opportunities for low- cost manufacturers, but competition on price alone, at the expense of quality, is not sustainable. The manufacturing sector depends on stan- dards and quality to a much greater extent than do primary sectors such as agriculture and mining. This is not only because manufacturing by its nature offers more possibilities for specifying technical requirements, but also because the manufacturing sector is more concerned with the greater integration of global production networks. New technologies, trade policies, and corporate strategies have changed the nature of global trade and have integrated producers ever more tightly in global produc- tion networks, where standards and quality play an important role. Introduction 9 Defining Standards and Quality Standards are ubiquitous in today's world, defining much of the way people, products, and processes interact with each other and with their environment. In the most general sense, a standard can be considered as a model or an example that has been established by some form of authority, custom, or general consent. Standards define characteristics or performance, convey information, or provide a means of communication. They are used everywhere in both the public and private sectors. They allow governments to collect taxes fairly and efficiently, based on mea- surements of economic output or amounts of goods traded. People rely on standards when they purchase electrical appliances and expect them to fit a given electrical outlet. Standards allow railroads to connect so that shipments and passengers can cross multiple regions unimpeded. They instill confidence in the safety of drinking water by specifying minimum acceptable levels of harmful contaminants. For the purpose of this book, we will define quality as the degree to which a set of inherent characteristics fulfills stated and unstated cus- tomer requirements and expectations or complies with stated norms, regulations, and laws, or both. Standards and quality are intrinsically con- nected, in that standards are often used to codify the technical character- istics expected by customers. In fact, the entire industrial infrastructure that is used to create quality goods and services relies on standards. The National Quality System: An Overview Countries rely on their national quality systems to remain competitive in the global economy. Producers that face growing pressure to meet quality requirements cannot do so on their own, but must rely on a comprehen- sive system of interrelated actors that facilitate the diffusion of standards and quality in the economy. In the private sector, producers themselves diffuse quality through backward linkages, by exerting pressure on their suppliers, and through forward linkages, by supplying high-quality inputs to buyers. Producers also enhance quality awareness through horizontal linkages, through either informal communication channels, demonstra- tion effects, or formal associative programs. In addition, most countries have developed an entire industry of service providers that specialize in the diffusion of quality and standards through the provision of qual- ity assessments, technical assistance, information, and training services. 10 Quality Systems and Standards for a Competitive Edge Finally, public and nonprofit institutions also play a central role in ensur- ing that the private sector has the necessary infrastructure to coordinate the development and adoption of standards throughout the economy. These institutions provide vital links to other countries' national quality systems and allow for greater harmonization and recognition of different national quality systems. The activities used to evaluate whether a product, process, or service fulfills specified technical requirements are part of what is known as con- formity assessment procedures. These requirements can be described in terms of either voluntary or mandatory standards. The actual conformity assessment procedures differ according to the product or process and can include any combination of testing, inspection, calibration, and certifica- tion, performed by one of several conformity assessment bodies. Testing involves performing measurements with certain instruments to evalu- ate a product or process according to a specified procedure; inspection is similar but involves less sophisticated instruments or no instruments at all. Calibration is used to determine the relationship between an instru- ment's input and its output, and certification provides assurance that a product or process complies with a certain standard. In sum, conformity assessment activities provide the vital link between standards and the products, processes, and services themselves. Conformity assessment activities are supported by a multidimensional infrastructure of calibration, metrology, accreditation, and standards orga- nizations. Although conformity assessment activities can be performed by the supplier or by the purchaser of the good whose quality is to be ascertained, independent organizations often perform this function. Cali- bration laboratories ensure that the measurements performed by testing laboratories and inspection bodies are reliable. Third-party calibration of instruments is also sometimes required as a condition for certification. To demonstrate the accuracy and precision of their measurements, cali- bration laboratories can establish traceability to reference measurement standards by calibrating their own equipment at a national metrology institute. All conformity assessment bodies can demonstrate that they have the competence to offer such services by seeking accreditation from a national accreditation body. All aspects of these activities rely on stan- dards, which provide the basis for evaluation of conformity assessment bodies and define the requirements against which conformity assessment is performed. Box 1.3 describes the different components of a national quality sys- tem, and figure 1.4 shows the relationships between these components. Introduction 11 Figure 1.4 Schematic Representation of a National Quality System National metrology institute Definition of units National standards (may be required in standards) body or bodies Calibration Accreditation body standards certificate required for National accreditation body laboratory accreditation Accreditations Certification Inspection Testing Calibration body body laboratory laboratory standards standards standards standards Certification Inspection Testing Calibration bodies bodies laboratories laboratories Calibration certificate Standards required for Certification Inspection Testing Calibration certification certificate report certificate Enterprises Enhanced product quality and compatibility Enhanced safety and health Benefits Decreased environmental impact Consumers and the general public Standards and definitions Conformity assessment processes Source: Authors'elaboration. Organization of the Book Thefirsthalfofthisbookpresentsaconceptualframework,highlightingthe importance of a national quality system and explaining its role in interna- tional trade and economic development. Chapter 2 examines the systemic impact of quality and standards on economic growth and international trade. In particular, it looks at the usefulness of standards as an entry point for technological upgrading and for the mainstreaming of SMEs.Chapter 3 addresses the relative roles and jurisdictions of the public and private sectors in the implementation of an effective national quality system. 12 Quality Systems and Standards for a Competitive Edge Box 1.3 The Components of a National Quality System National standards bodies National standards bodies are organizations that bring together public and private stakeholders to develop o cial national standards. They are usually either public sector entities or nonpro t private entities. Standards bodies usu- ally adopt the standards through consensus and publish them so that they are available to industry, public sector institutions, and consumers. The standards may be voluntary, in which case there is no obligation to use them, or manda- tory, in which case they are enforced by governmental authorities. Certification bodies Certi cation consists of the provision of assurance that a product, service, sys- tem, process, or material conforms to one or more standards or speci cations. Although there exist some self-certi cation schemes, certi cation is usually conducted by a third party that is independent of the supplier or purchaser. Certi cation bodies are usually commercial for-pro t or nonpro t entities, although in undeveloped markets they are sometimes public sector organiza- tions. Product certi cation can involve various degrees of con dence, depend- ing on the referenced standard. In the simplest case, a prototype or a product from a preproduction run is tested and inspected against a speci c standard. More vigilant tests involve surveillance of the manufacturing process, random testing of samples, batch testing, and 100 percent testing, where every prod- uct is examined individually. Process certi cation involves an assessment of the variables that have an impact on a rm's output. For example, certi cation of a rm's manufacturing process quality to a certain standard may be based on an audit verifying the quality of the components or materials, equipment, equip- ment calibration and maintenance, the training and experience of the person- nel, and the environmental conditions. Testing laboratories Testing involves determining the characteristics or performance of a product or process according to a speci ed procedure. Testing is often a requirement for certi cation, but it is also used for a variety of purposes such as product design and research, quality control, satisfying contractual agreements, satisfying regu- Introduction 13 latory requirements, buyer protection and information, medical health and ser- vices, or product repair and maintenance. Testing laboratories come in all forms and sizes. They are usually private sector entities when they concern voluntary standards, but they can be found in both the public and private sectors when they test against mandatory requirements. Inspection bodies Inspection can be conducted alone or in combination with testing to determine whether a product or process meets certain requirements. The inspection of products is usually conducted by visual means or by using simple instruments. In contrast to test results, the outcome of the inspection process depends highly on the subjective judgment and experience of the inspector. Like testing laboratories, inspection bodies can be either private or public sector entities, depending on their role in enforcing mandatory requirements. Calibration laboratories Calibrationinvolvesdeterminingtherelationshipbetweenaninstrument'sinput and the magnitude or response of its output. It also serves to establish the accu- racy and precision of a measuring instrument. Calibration must be performed using equipment of known uncertainty. Commercial calibration laboratories compete for nal industrial users, or what is called the secondary calibration market. Those users do not themselves o er commercial calibration services. The national metrology institute The role of a national metrology institute (NMI) is to establish a country's national measurement system; to maintain, develop, and di use measurement standards for basic units; and to di use metrological expertise to the economy. Countries often have a single NMI, but when there are several NMIs each is responsible for distinct areas of measurement. It is common for NMIs to be pub- lic sector organizations, but they can also be private sector organizations. NMIs operate in the primary calibration market: they disseminate measurement stan- dards by providing calibration services to independent calibration laboratories and to other organizations responsible for regulations and standards. When their industrial measurements are traceable to the NMI through an unbroken chain of comparisons, rms are able to guarantee the accuracy and precision of their calibration instruments, process control instruments, and quality control instruments. continued on the next page 14 Quality Systems and Standards for a Competitive Edge Box 1.3 The Components of a National Quality System--Continued The national accreditation body Accreditation is de ned as the procedure by which an authoritative body gives formal recognition that an organization or person is competent to carry out spe- ci c tasks. Accreditation is sought on a voluntary basis as proof of competence in a given area. Most countries have a single national accreditation body responsible for all areas of accreditation. It can be either a public or a private not-for-pro t organization. Accreditation provides certi cation and inspection bodies, as well as testing and calibration laboratories, with a means to signal that they are con- ducting their work to appropriate standards and that they are able to provide reliable services. The accreditation body evaluates the personnel and supporting management system of the accreditation candidates and, when relevant, can request practical tests for laboratories. These tests take the form of pro ciency testing schemes through which the measurement results of di erent laboratories are compared. Accreditation is usually valid for a few years after initial assessment, during which time the accredited organization is subject to regular surveillance. Chapter 4 presents the structure, organization, and functioning of the certification, testing, calibration, inspection, accreditation, and metrology bodies of the national quality system. Chapters 5 and 6 discuss interna- tional aspects of the system, especially the ISO 9000 quality manage- ment standards. Chapters 7 through 10 focus on the experience in Latin America and the Caribbean.These chapters examine the standards-setting institutions, certification activities, accreditation processes, and metrology institutions in this region. The last two chapters of the text turn to the policy framework. Chap- ter 11 analyzes existing policies and support programs implemented in several Latin American countries to facilitate the diffusion of standards and quality in the productive sector. Finally, chapter 12 presents detailed policy recommendations for developing countries for implementing effective national quality systems and ends with a brief conclusion. C H A P T E R 2 Economic Impact and E ect of Quality and Standards Standards fulfill a variety of functions. They convey information, allow interoperability between products and processes, guarantee minimum levels of quality and safety, and result in economies of scale by reduc- ing variety. Positive economic effects of standards include their ability to exploit network externalities, to enhance productive and innovative efficiency, to reduce imperfect information, to diffuse information on innovation, and to promote competition. Depending on the context and the content of standards, they may also have negative economic effects by imposing constraints on innovation and decreasing market competi- tion. In light of these mixed positive and negative effects, a number of empirical studies have attempted to isolate the net economic effect of standards. The results of these studies all point to the fact that, under the right conditions, standards have a net beneficial effect on growth. Functions of Standards Standards are commonly classified according to their specific function. This classification is useful in understanding the economic effects of standards through analytical models, but it should be viewed as a sim- plification of reality. Most standards serve several purposes and cannot be neatly classified into a single category. Furthermore, standards with different intended or immediate functions may ultimately have very similar economic effects, and this limits the utility of a strict functional 15 16 Quality Systems and Standards for a Competitive Edge separation. Keeping this in mind, we can distinguish four basic categories of standards functions (DTI 2005; Swann 2000; Blind 2004). Information and reference standards, also called measurement standards, establish a common technical language in which to compare physical attributes and convey descriptive technical information. They include unit standards, such as the number system, which were probably the first technical standards. Weights and measures were first codified as early as 3500 BCE, largely to facilitate the fair and reliable collection of taxes. Centuries later, in 1799, different unit and reference standards in vari- ous kingdoms coalesced into the metric system (Krechmer 2000). These standards also include information standards, which present rules on how to communicate product characteristics. For instance, bolt standards explain how to designate bolt dimensions.A bolt manufacturer need only specify "M10 x 1.5-6g-S" to be perfectly understood by its customers; it need not describe a specific bolt as a "metric fastener thread profile M, fastener nominal size (nominal major diameter) 10 mm, thread pitch 1.5 mm, external thread tolerance class 6g, and thread engagement length group S (short)." Variety-reducing (or interchangeability) standards define the com- mon characteristics of two or more entities. In this way they provide interchangeability and generate economies of scale and economies of learning in production. The majority of standards fall into this category. Variety-reducing standards grew out of the Industrial Revolution, when the efficiency of mechanized manufacturing depended on codifying the characteristics of products and processes. A well-known standard of this type is the international paper standard, ISO 216, which defines the A4 format used in most of the world except for North America. The wide- spread use of A4 paper has many advantages other than economies of scale in paper production itself. It avoids the need to rework documents to fit different formats, and it allows consumers to choose between competing paper brands, calculate shipping weights from the number of pages (most A4 sheets have the same weight), and fit papers from dif- ferent sources into the same envelopes and binders, among many other advantages.1 Compatibility and interface standards define physical or virtual relation- ships between independent entities for the purpose of interoperability or communication. Most of a country's infrastructure uses compatibility standards to connect a number of disparate private and public entities. Consider the case of electricity distribution standards in the United King- dom, an early adopter of compatibility standards. All electrical outlets Economic Impact and Effect of Quality and Standards 17 accommodate a single type of plug, type G, and supply 230V at 50Hz. In contrast, in Cuba, there is a single 60Hz frequency standard, but the elec- tric voltage supplied is either 110V or 220V, and outlets can come in any of five types--A, B, C, F, or L. Hence, Cuban consumers might need to fit electrical appliances with burdensome adapters and converters before being able to use them at home. Minimum quality and safety standards allow consumers to assess the quality or safety of a product before purchasing it. The best-known qual- ity standards, the ISO 9000 standards of the International Organization for Standardization (ISO), are perhaps the most misunderstood. ISO 9000 standards do not specify the quality of a product, but of a manage- ment system. They are process standards that describe the organizational procedures that an organization must follow to ensure the consistent quality of its products and services. Suppliers adhering to these standards signal to buyers that they are able to produce goods and services of con- sistent quality. These standards play a particularly important role in the world economy today and they will be revisited many times through- out this book. Like quality standards, safety standards are widely present among consumer goods such as toys, food, drugs, and electrical appli- ances. They may specify requirements related to product design, product performance, or manufacturing process. Positive Economic Effects of Standards Productive and innovative e ciency Some of the most common economic benefits of standards are to increase productive and innovative efficiency. Variety-reducing standards lead to economies of scale and economies of learning. They allow suppliers to achieve lower per-unit costs by producing large homogeneous batches. In addition, producers gain skills and experience by focusing on fewer product variations and can increase their efficiency through economies of learning. By allowing producers to concentrate on a manageable number of product options instead of fragmenting their research and develop- ment (R&D) efforts, variety-reducing standards also increase innovative efficiency. The national standards body of the United Kingdom, now known as BSI British Standards, describes how the first public standards resulted in significant savings: On 26 April 1901 the first meeting of the Engineering Standards Committee took place. As a result, the variety of sizes of structural 18 Quality Systems and Standards for a Competitive Edge steel sections was reduced from 175 to 113 and the number of gauges of tramway rails was reduced from 75 to 5. This brought es- timated savings in steel production costs of £1 million a year. Steel merchants' costs were reduced due to fewer varieties. This made steel cheaper for the users so everyone benefited.2 Compatibility standards can facilitate both productive and innovative efficiency by providing interfaces for components in a system. They give producers the flexibility to quickly modify products or processes or cre- ate new designs and experiments by simply introducing new compo- nents, without having to generate a completely different system. Di usion of innovation An often unintended benefit of standards is to spur innovation by provid- ing information on the state of the art of a particular technology. When information on innovations is codified in standards and this information is nonproprietary, it is accessible to everybody, at least in principle. Firms, universities, and research organizations can use the knowledge embodied in standards to adopt innovations or generate new ideas. Standards play a particularly useful role in disseminating knowledge in industries where products and processes supplied by various providers must interact with one another. They ensure that information on innovations in one part of the sector will be diffused to other parts of the sector. A survey of British companies found that over 60 percent of product and process innovators used technical standards as a source of information for innovation (DTI 2005). This was twice the number of companies that cited universities or research laboratories as sources of innovation. Because standards can be acquired across borders, they can also constitute important sources of technical knowledge for firms in developing countries. The nature of the standardization process itself also facilitates the dif- fusion of technological information. Often, standards are the outcome of a formal development process in which interested parties come together to study and discuss technical specifications. Usually participants try to advance standards that are as close as possible to their own products or processes. As a result, engineers and other experts working in the same technological areas must agree to share their own technical and com- mercial information during the standard-setting process. In fact, a survey of 4,000 companies in Austria, Germany, and Switzerland found that businesses reduced both the economic risk and the cost of their R&D activities by participating in standardization (DIN 2000). Economic Impact and Effect of Quality and Standards 19 Reduction of imperfect information Problems arise when consumers, producers, or governments do not have all the information they need to make sound purchasing, investment, or policy decisions. Certain standards draw their utility from conveying information that corrects for problems of imperfect information. Minimum safety standards are the most straightforward example of standards used to solve imperfect information problems. By conveying information about the harmful effects of products, they allow custom- ers to avoid products that might negatively affect them, the public, or the environment, and they allow regulators to exclude unsafe products from the market. These standards thus reduce the societal costs of deal- ing with potentially harmful effects of certain products. These savings in turn allow a more efficient allocation of resources and can contribute to economic growth. Standards can also reduce the transaction and search costs that are caused by imperfect information. This is another role of minimum safety standards, but it can also be accomplished by quality standards and prod- uct description standards. These standards allow buyers to confirm that products and processes have the characteristics they want without the additional transaction costs of independent testing. By codifying market preferences, standards also save sellers the additional costs of defining consumer preferences. By reducing information asymmetries, standards can help mitigate adverse selection problems. In many situations of imperfect information, the seller has better information than the buyer. Sellers cannot charge a price premium for high-quality products if buyers are unable to distin- guish them from low-quality products. If the costs of high-quality pro- ducers are greater than the costs of low-quality producers, high-quality producers will undersupply or could even be driven out of the market. Minimum quality or safety standards help consumers confidently dif- ferentiate product quality and safety, and hence help mitigate adverse selection problems (box 2.1). Exploitation of network e ects Compatibility and interface standards add economic value to goods with network externalities and facilitate the development of networks. Com- patibility standards can increase direct network externalities by allow- ing products to work as part of a system or network. They allow each individual participant in the network to derive benefits from interacting with other participants in the network. This is the case of the telephone 20 Quality Systems and Standards for a Competitive Edge Box 2.1 Quality Standards in the Ugandan Coffee Industry Co ee plays an important role in the Ugandan economy and represented 17 percent of exports in 2004. However, in the early 1990s a combination of factors inside and outside Uganda started to erode the incomes and export rev- enues of co ee growers and threatened the long-term viability of the industry. These factors included aging co ee trees, pests and diseases, poor postharvest handling methods that resulted in lower quality levels, and, especially, an over- supply on the international market that drove down prices. Low world prices in turn worsened quality levels by acting as a disincentive for producers to invest in better crop husbandry. At the same time, liberalization of the co ee industry exacerbated the quality problems and threatened the image of Ugandan co ee overseas. The co ee industry has a long history of regulation in Uganda. Until the early 1990s, the Co ee Marketing Board, the state-controlled monopoly, had the exclusive rights to purchase and export all dry processed co ee. In 1991, the co ee industry was liberalized, and by 2001, about 200 licenses had been issued to private exporters. The liberalization improved the producers'revenues but also had undesirable outcomes. Increased competition and the entry of inexperienced exporters led to a decrease in co ee quality.This e ect was exac- erbated when a number of co ee buyers (middlemen) and producers realized that they were able to sell low-quality co ee and started to ignore quality at the expense of export quantity. These quality issues had negative repercussions for network, for example: subscribing to the network is only useful insofar as there are other subscribers with compatible telephones (box 2.2). Compatibility standards can also generate indirect network externali- ties when they allow users to derive benefits from a system involving two or more complementary components. In the case of the personal computer, the value of owning the hardware increases with the number of compatible software options. In turn, the market will provide more software options if there are more owners with compatible hardware. The indirect externalities are caused by the incremental effects of each purchasing decision on the future variety and price of complementary products available within the system. More hardware purchases will stimulate the supply of software and vice versa. This positive feedback Economic Impact and Effect of Quality and Standards 21 all Ugandan co ee exporters. Exports of low-quality co ee tarnished the image of Ugandan co ee abroad, which in turn reduced incentives for production of high-quality co ee. Given that the global production surplus had led interna- tional buyers to be more selective on quality, Uganda's inability to strengthen producer practices posed a serious threat to the viability of its co ee industry. Uganda reacted by establishing a mix of voluntary and mandatory co ee quality standards. In 1994, the Co ee Regulations were adopted to stipulate co ee standard and grade requirements for exporters. The quality standard requirements were enforced by the Uganda Co ee Development Authority through periodic inspections. The Co ee Regulations were complemented by the Uganda Co ee Trade Federation's Code of Practice, which directed mem- bers to conform to minimum quality standards. Federation members who con- sistently did not comply with quality requirements were issued nes. Moreover, the federation could make recommendations on the licensing of co ee export- ers to the Uganda Co ee Development Authority, and it would not support members that breached the Code of Practice. As an additional incentive for cof- fee producers, the federation made the list of its members who had subscribed to the Code of Practice available to co ee associations in importing countries. Although enforcement has sometimes been di cult, the main e ect of the Co ee Regulations and the Code of Practice has been to promote exports by projecting a positive image of Ugandan co ee quality in world markets and helping foreign buyers distinguish low-quality from high-quality suppliers. Source: Rudaheranwa, Matovu, and Musinguzi 2003. effect will increase the value of both the hardware and the software and expand the network of users. Increased competition By codifying market preferences and technical information and estab- lishing interchangeability, standards promote competition. By specifying the characteristics of products and services, standards give all produc- ers equal access to market information and level the playing field. In particular, compatibility standards can increase competition by reduc- ing entry barriers in markets with network effects. By specifying inter- face designs, standards make it possible for a company to enter a market without investing in an entire fully integrated system; it can compete 22 Quality Systems and Standards for a Competitive Edge Box 2.2 Transatlantic Competition for First- and Second-Generation Mobile Phones Standards have played a critical role in the development and di usion of mobile phone networks. While in 1990 there was only one cellular subscriber per 1,000 people, as of September 2005 there were more than 2 billion mobile phone subscribers in the world, or approximately 300 subscribers per 1,000 people. The success of the mobile phone industry has not been limited to high-income countries; for instance, in Sub-Saharan Africa there were six times more mobile phone subscribers than telephone land lines in 2003. First-generation cellular phone systems, based on analog technology, emerged in the 1970s, thanks in large part to the invention of the micropro- cessor. By 1978 the Bell telephone company was experimenting with the rst commercial cellular network using Advanced Mobile Phone System (AMPS) standards. Based on Bell's proposal, the U.S. Federal Communications Commis- sion (FCC) mandated AMPS as the cellular standard in the United States and imposed antitrust legislation to ensure competition in the mobile phone mar- ket. The establishment of a single standard created a critical mass of subscribers able to communicate with compatible networks throughout the country and leveraged network externalities. By 1993 more than half of the world's wireless cellular systems were based on the AMPS standard. In Europe, two dominant standards emerged for rst-generation phone systems. Ericsson and Nokia developed the Nordic Mobile Telephone (NMT) standard in the 1970s, which was subsequently adopted by a number of Euro- pean countries in the 1980s. The Total Access Communications System (TACS) standard originated in the United Kingdom and Italy and was the main Euro- pean rival to NMT. In addition, a number of other proprietary phone systems were established across Europe but for the most part each one operated in a single country. The incompatibility of rst-generation cellular networks across countries proved to be a serious handicap for the European cellular telecom- munications industry, as it failed to achieve economies of scale. As a result, there were far fewer mobile phone subscribers in Europe than in the United States in the 1980s. An important problem with rst-generation analog phones was their inef- cient use of a nite amount of available bandwidth. As more subscribers Economic Impact and Effect of Quality and Standards 23 adopted rst-generation phones, cellular networks approached saturation lev- els. In contrast, second-generation phone systems, based on digital technology, utilized spectrum much more e ciently and also o ered better voice quality. Commercial operation of second-generation phones began in Europe and the United States in 1991. This time, the United States and Europe reversed their original approach to cellular standardization. In the United States, the FCC decided not to adopt an o cial cellular stan- dard but to allow market competition to select the optimal technology. This approach led to segmentation of the market into a number of competing industry standards in the United States, including CDMA, TDMA (or "D-AMPS"), and the rst-generation AMPS system. However, the FCC's expected results did not materialize and no dominant second-generation technology emerged. Multiple standards coexisted well into the end of the 1990s. Limited network externalities signi cantly slowed digital phone adoption rates and by 1998 only 30 percent of subscribers were using digital phones in the United States. In Europe, the development of the Global System for Mobile Communica- tions (GSM) had already begun in 1981 as an e ort of the European Commis- sion, telecommunications operators, and equipment manufacturers to create a single open pan-European digital standard. In 1989 GSM was o cially accepted by the European Telecommunications Standards Institute and was adopted by each of its member countries. The GSM standard would allow one cell phone to work in all European countries. As a result of standards harmonization, GSM phone networks di used rapidly across Europe and displaced the old analog technologies. By 1998 over 90 percent of cellular subscribers in Europe were using GSM technology. The vast European market produced economies of scale and economies of learning that gave European phone manufacturers and network operators a critical advantage over their American and Japanese competitors. This enabled them to capture most of the world's cellular market. By September 2005 there were 1.6 billion GSM users worldwide, representing 77 percent of cellular subscribers. The GSM standard even captured an impor- tant share of the North American mobile telecommunications market and was used by 20 percent of its cellular subscribers in 2003. There are currently GSM operator members in 210 countries and territories. Sources: Ritchie et al. 1999; ITU 2005, 2006; GSM Association 2005. 24 Quality Systems and Standards for a Competitive Edge in the market for a single component of that system instead. Thanks to standards, a more open market results in more consumer choices, lower prices, and higher quality.The publication of interface standards decreases the ability of suppliers to extract high rents on replacement parts or new complementary products. If interface information is widely accessible, no single supplier can internalize the benefits of the entire network or has incentives to create switching costs for customers. Increased competition in turn leads to an optimal allocation of resources and greater efficiency in the economy. Cost reductions Standards are being developed by leading nations to reduce operating costs and increase competitiveness. For example, a group of 12 countries led by Singapore has developed an international standard for electronic seals on freight containers that is expected to save the global shipping community $64 million annually. The global standard, to be called ISO 18185, dictates a protocol for electronic recognition of the seal num- ber and for checking seal conditions electronically using radio frequency identification technology. The global standard is expected to speed up and improve these services, leading to quicker customs clearance for cargo in ports. It is also expected to help small and medium enterprises (SMEs) bring their supply chain and security planning needs up to multinational companies' standards. This will help avoid unnecessary delays during customs inspections. Negative Economic Effects of Standards Some of the very elements associated with the positive economic value of standards, such as compatibility and interchangeability, can also be detrimental to public welfare. Depending on the specific content of a standard and on market characteristics, standards can have a negative impact on two important contributors to economic growth: innovation and market competition. Constraints on innovation The positive effects of standards on technical change are counterbal- anced by their negative effects on product variety. By specifying product characteristics, such as form, performance, or interface, standards limit innovation. New designs need either to satisfy the given standards or to embody radical technological improvements that place them outside the Economic Impact and Effect of Quality and Standards 25 scope of existing standards. This can significantly increase development time and cost. By decreasing variety, standards also reduce the diversity in the pool of products that can be used as a basis for further innovation. Obsolete, inappropriate standards may hamper technical change by preventing the adoption of superior technology through the lock-in effect. Standards may create situations in which users are locked into a network or a product that uses less efficient technologies. If standards are formu- lated too late, the costs of switching to a new technology may become too high and prevent its diffusion. The QWERTY computer keyboard is a well-known case of technological lock-in. Although many alternative keyboard layouts have been designed to increase typing speed and com- fort, the nineteenth-century QWERTY layout is still used because the costs of retraining workers to use a different keyboard would be prohibi- tively high. Furthermore, new users will still prefer to learn to type on a QWERTY keyboard as long as this layout dominates the market. The effects of standards on innovation largely depend on timing.While standards encourage adoption of new technologies, newer technologies are not necessarily superior to older ones. An early switch of technology can prevent beneficial improvements to more mature technologies or can preclude the development of potentially more beneficial future tech- nologies. For example, had early typewriter manufacturers waited for fur- ther advances in keyboard design before adopting the QWERTY layout, we would perhaps have been using a more efficient keyboard today. Constraints on competition Standards can have anticompetitive effects if only one or a few compa- nies are able to internalize their benefits or control their content. In this case, standards can be used as strategic instruments to expand market power. This occurs when the content of the standards covers technologi- cal areas in which a limited number of firms have property rights, exclu- sive knowledge, or the exclusive resources needed to use a technology. In the case of compatibility standards, owners of proprietary information on interface technologies are able to control their use and extract high rents from their rivals. In 1998, Microsoft was taken to court for abusing its monopoly power. Some of the charges were directed at its exclusive control of application programming interfaces (APIs), interface standards that define how one piece of computer software communicates with another. Microsoft was accused of not disclosing the APIs that would enable other companies to produce Internet browser software that could work in the Microsoft Windows operating system and compete with 26 Quality Systems and Standards for a Competitive Edge Table 2.1 The Economic Effects of Standards According to Their Function Positive effects Negative effects Innovative Exploitation and Reduction Constraints Constraints of network productive of imperfect Innovation on on Function effects efficiency information diffusion innovation competition Compatibility X X X X X and interface Minimum X X X X quality and safety Variety X X X X reduction Information X X X X and reference Source: Authors'elaboration. Microsoft's own browser. It was found that Microsoft would only release this information to a company that entered in a "special relationship" with Microsoft.3 Table 2.1 provides a summary of positive and negative economic effects of standards according to their function. Again, it must be empha- sized that few standards fit clearly into a single functional category and that a combination of functions will result in some combination of eco- nomic effects. Box 2.3 provides an example of the quantifiable economic effects of measurement standards. Box 2.3 The Economic Value and Impact of Measurement Standards The economic role of measurement standards is of particular interest because they are the most widely used type of standard. Measurement standards involve the de nition of units of reference and their method of measurement.The accu- racy and consistency of measurements in an economy re ect the speci cation of measurement standards and their di usion throughout the economy. Measurement plays a signi cant role in assessing the quantity of goods being traded. A report for the International Committee of Legal Metrology (Birch 2003) Economic Impact and Effect of Quality and Standards 27 cited studies in several countries that point to the importance of measurement standards. A study of the Canadian economy reported that the total value of goods traded over all classes of trade-related weighing and measuring instru- ments totaled 32 percent of the gross national product (GNP) in 1989­90, exclud- ing prepackaged goods and utility metering. Similarly, the total value of trade transactions involving measurement in Australia in 1998 was estimated to be 60 percent of GNP, including prepackaged goods and utility metering. Another study estimated that weights and measures regulations impacted transactions involving 54.5 percent of United States GDP in 1996, including prepackaged goods and utility metering. Figures from these three countries are consistent with one another and con rm the importance of measurement in trade. Several economy-wide analyses have estimated the total cost of measure- ment activities in the United States and Europe. Early studies by the U.S. National Bureau of Standards between 1967 and 1984 found that industry, govern- ment, and community expenditures on measurement-related activities totaled between 3 and 6 percent of GNP (Birch 2003). A more recent study found that The European Union spends 83 billion euros per year, or nearly 1% of European Union GDP, on measurement activity (NMI turnover, accredited calibration laboratories turnover, certi cation costs to industry, instru- ment costs, and industrial spending on measurements). Adding social spending on health, environmental regulation, safety testing, antifraud projects and normal day-to-day measurement activities raises this gure considerably. (BIPM 2003: 116) Estimating the impact of measurement on the economy as a whole is a much more complex task. Studies commissioned by the European Commission's Directorate-General for Research (DG Research) illustrate the role measurement can play in speci c economic sectors and economy-wide. One study nds that a metering error of 1 percent in the gas industry equals about 4 billion cubic meters per year in Western Europe, with a commercial value of 800 million euros. The authors note that in reality metering errors easily attain 4 to 6 percent. Another DG Research study points to the cost of measurement and testing in European health services, which totals 13 billion euros per year. Econometric estimates also showed that expenditures on measurement activities generated 230 million euros of bene- ts, equivalent to 2.67 percent of European Union GDP.The model excluded social bene ts related to health, safety, and the environment. The resulting bene t-to- cost ratio of measurement was found to be 2.73 (Birch 2003; BIPM 2003). 28 Quality Systems and Standards for a Competitive Edge The Impact of Standards on Economic Growth: Empirical Evidence If standards affect technology diffusion, innovation, and technical change, endogenous growth theory implies an associated change in growth rates. The economic effects listed in table 2.1 give theoretical reasons to believe that standards impact economic growth, but do not establish in which direction. Two series of studies provide insight into the role of standards in innovation and economic growth. One set of studies was sponsored by the United Kingdom's Department of Trade and Industry (DTI 2005) and the other by the German Institute for Standardization (DIN 2000). The empirical relationship between innovation and standardization shows an inverted U nonlinear pattern, under several specifications. As sources of information on innovation, standards promote the adoption of new technologies by firms and consumers, but they also introduce economic constraints that impede the innovation process. Econometric models, based on surveys of firms and the corresponding standards stock in different economic sectors in the United Kingdom, suggest how the constraining and informative roles of standards vary with the number of standards and the median age of the standards stock (DTI 2005). A first model, concerned with information provision, finds that an increase in the number of national standards provides more information to producers. However, the model finds a nonlinear relationship between information provision and the median age of the standards stock in an industrial sector. As the median age of standards increases, producers find standards more informative, but after a certain point the standards start to lose their information content and median age begins to have a nega- tive effect on information provision. The explanation is simple: newly published standards are not widely diffused or properly understood and thus are of limited impact, while older standards become outdated and lose their information value as the technologies and products they refer to reach the end of their life cycle. A second model, concerned with the constraining role of standards, finds significant quadratic relationships with both the number and median age of standards. As the number of standards in a sector increases, producers at first find them less constraining on innovation, but after a certain point they start to find them more constraining. The model also finds that both rather new and rather old standards constrain innovation. Here, the authors contend that new standards constrain advances in the latest technologies by locking innovators into legacy systems, while older Economic Impact and Effect of Quality and Standards 29 standards constrain innovators by forcing them to conform to outdated technologies. The complex nature of the effect of standards on innova- tion may explain why, in another study based on time-series analyses using patents and standards, Blind (2004) only reports a weak positive impact of standards on innovation. Time-series investigations suggest a significant positive long-run con- tribution of standards to economic growth in the United Kingdom. A first study is based on an econometric analysis of time-series data over the period 1948­2002 in the United Kingdom (DTI 2005). The macro- economic model specifies total factor productivity (TFP) at time t as a function of the patent stock (PAT), expenditures for foreign patents and licenses (LEX), and the standards stock (STD). TFP(t) = f [ et PAT(t) LEX(t) STD(t)] This TFP is an index of the "technological level" of the economy. The construction of TFP in this model is based on the premise that standards alone cannot contribute to economic growth. Instead they contribute to technology diffusion and must work in conjunction with foreign and domestic innovation to increase productivity. In this model, the output Y is determined according to the following production function: Y(t) = TFP(t) F(K(t),L(t)) where K stands for capital, L for labor, and TFP is calculated as the Solow residual. The results of the analysis find that standards contributed to about 13 percent of the growth in labor productivity in the United King- dom over the given period, or to about 0.28 percent per year. The United Kingdom's GDP grew 2.5 percent per year between 1948 and 2002, of which 1.5 percentage points could be explained by labor and capital accumulation and 1 percentage point could be explained by technologi- cal change. More than 25 percent of the growth induced by technologi- cal change could be attributed to standards (figure 2.1). Central to the contribution of standards to productivity was an annual growth of the standards stock at an average rate of 5.1 percent per year (figure 2.2). The computed long-run elasticity of labor productivity on the stock of standards was about 0.05. Hence, a 1 percent increase in the number of standards could be associated with a 0.05 percent increase in labor productivity. Further analysis of the model shows that standards did not have a significant short-run effect. Rather, standards take time to diffuse among a user population and their impact is long run in nature. 30 Quality Systems and Standards for a Competitive Edge Figure 2.1 Contributions of Different Factors to Output Growth in the United Kingdom, 1948­2002 3.0 2.5 (%) 2.0 wthorg 1.5 1.0 annual 0.5 0 output capital employment technological unobservable standards change factors contributions to technological change Source: DTI 2005. Figure 2.2 Long-Run Growth of Labor Productivity in the United Kingdom, 1948­2002 10,000 stock of BSI standards capital-employment ratio labor productivity 1,000 100 1948195119541957196019631966196919721975197819811984198719901993199619992002 Source: DTI 2005. Note: 1948 = 100 log scale. Economic Impact and Effect of Quality and Standards 31 In Germany, a similar study found that, after capital accumulation, standards were the second factor contributing to growth. A slightly older macroeconomic analysis of standards was performed by Jungmittag, Blind, and Grupp (1999) for the German economy over the 1960­96 period. This study was the precursor of the DTI study and used the same methodology, incorporating standards in the production function of the model, along with technology imports and patents. The results showed that standards were at least as important to technical innovation as pat- ents and second only to capital accumulation in their contribution to growth. In the period 1960­90, the analysis found standards contributed to about 0.9 percentage points out of an average overall growth rate of 3.3 percent. However, from 1990 to 1996, after German reunifica- tion, the contribution of the standards stock decreased to 0.3 percentage points of an overall growth of 1.5 percent per year. An empirical study of different manufacturing sectors in Europe con- firms the major impact of the standards stock on productivity growth but draws attention to the differential impact of standards across sectors and countries (DTI 2005).The four countries included in this project, France, Germany, Italy, and the United Kingdom, all experienced a rapid growth of their standards stocks during the 1990­2003 period.The methodology was basically the same as in the two previously mentioned macroeco- nomic studies by DTI, but it was applied to 12 individual manufacturing sectors and did not consider technology imports. The estimated impact of standards on productivity across all 12 industries and all four coun- tries was found to be of the same order of magnitude as that reported in the economy-wide analysis. Country-level models suggest a differential impact across countries. Table 2.2 shows the partial production elastici- ties of patents and standards in each country and at the aggregate level. The contribution of standards was positive in each country. Standards were as important as patents for productivity in the United Kingdom, far more important in France, and less important in Germany and Italy. When the models are run individually on different industry sectors, a rough pattern in the results suggests that standards have a more signifi- Table 2.2 Partial Production Elasticities in Selected European Countries All four countries United Kingdom Germany France Italy Patent stock 0.105 0.047 0.094 0.072 0.059 Standards 0.079 0.052 0.027 0.147 0.017 Source: DTI 2005. 32 Quality Systems and Standards for a Competitive Edge Figure 2.3 Average Productivity Gains Due to Adoption of Quality Standards 15 (%) 12 gain 10 y tivit 4.5 5 oduc 1 pr 0 Central East China America Asia Sources: Escribano and Guasch 2005a, 2005b. Note: Central American sample includes El Salvador, Guatemala, Honduras, and Nicaragua. East Asian sample includes Indonesia, Malaysia, the Philippines, and Thailand. cant impact on productivity in mature industries than in R&D-intensive industries, where patents are more important. This observation makes sense, because low- and medium-technology sectors tend to rely on established technologies, which are documented in standards. Studies using firm-level surveys of investment climate factors in devel- oping economies also found that the adoption of standards had a significant impact on several measures of productivity. This analysis used a total of 10 productivity measures based on total factor productivity, rates of growth, and input-output elasticities based on the levels of the relevant variables. The adoption of standards was proxied by certification to ISO standards. Average productivity gains were estimated to be between 2.4 percent and 17.6 percent for four Central American economies, less than 1 percent for four Southeast Asian economies, and 4.5 percent for China, as shown in figure 2.3 (Escribano and Guasch 2005a, 2005b). The Impact of Standards on International Trade: Empirical Evidence The impact of standards on trade is a priori ambiguous. The role of stan- dards in trade can essentially be derived from their micro-level economic effects on public welfare, discussed earlier in this chapter, which may be either positive or negative. Many of the positive economic effects translate directly into positive effects on trade. By diffusing market and technological information across borders, standards allow countries to compete in new international mar- kets. They enhance transparency and allow foreign producers to appro- priate national preferences and technical specifications and adapt their Economic Impact and Effect of Quality and Standards 33 products and services accordingly. By establishing product and process characteristics or performance, standards reduce transaction costs between business partners in distant countries and reduce information asymmetries. When the same standards are used in different countries, they promote trade by enabling specialization and the generation of economies of scale. Also, compatibility and interface standards allow countries to specialize in system components for which they have comparative advantage. However, there are many other circumstances in which standards will have no effect on trade or even act as a deterrent to trade. Many of the theoretical arguments that support a positive role for standards in trade are based on the assumption that standards are public goods that can be accessed and adopted on an equal basis by producers in any country. In some cases this may be far from the reality. For example, standards may not be well documented, they may be difficult to find, they may be poorly understood, or they may be protected by intellectual prop- erty rights. If each country uses its own idiosyncratic or country-specific standards, global economies of scale will not be realized and each coun- try may end up focusing exclusively on its domestic market. If products must be adapted to conform to different national standards, additional costs can hinder trade (box 2.4). Some countries may even find them- selves excluded from international markets when compliance with for- eign standards requires technological capabilities that are beyond their reach. But shared standards that reduce variety can also hamper trade. Standards that are harmonized between countries can distort trade if their content strongly favors one trade partner over the other. A country may be placed at a competitive disadvantage if it does not have the tech- nological capabilities, the industrial infrastructure, or the natural resource endowments needed to comply with a shared standard. Some studies involving business surveys highlight the potential of standards as barriers to trade. This is the approach used in a study spon- sored by the German Institute for Standardization (DIN 2000). Business surveys in Austria, Germany, and Switzerland showed that 27 percent of firms had to adapt their products and services to foreign standards. Adaptation costs averaged DM 350,000 per year at the firm level, rang- ing from DM 2,500 to DM 6 million.4 However, the data presented in the DIN report make it difficult to evaluate the relative importance of these costs to exporters. A survey by Wilson and Otsuki (2004) offers more insight on the trade barriers posed by mandatory standards.5 This survey, which covers 17 devel- oping countries, shows that a high share of firms face technical regulations 34 Quality Systems and Standards for a Competitive Edge Box 2.4 Differing Standards and Trade Costs: The Case of Pallets Raballand and Aldaz-Carroll (2005) provide a detailed account of how the absence of harmonized pallet standards increases trade costs. The pallet is a portable platform for storing, stacking, handling, and transporting goods. Pal- lets were initially developed to protect products, facilitate storage, and improve distribution e ciency. They are used extensively in most countries and carry 80 percent of traded goods in the United States. Part of the e ciency of pallets is compromised by the incompatible pallet standards used in di erent countries. There are hundreds of di erent pallet sizes around the world, so goods shipped from one country to another must often be manually unloaded from the pallet of the country of origin and reloaded onto the pallet of the destination country. This operation increases delivery costs for a number of reasons. Transferring goods between pallets requires additional time and labor and a warehouse might need to own di erent equipment to handle each type of pallet. Moreover, manual handling of goods between pallets can lead to product losses and deterioration of the pallets. A study of banana exports from Ecuador to Europe found that di erent pallet sizes introduce ine ciencies that increase total transport costs by 21 percent, or $27 million annually. Developing countries are particularly a ected by the costs of using multiple pallet sizes because these costs are high relative to the low value added of their exports. In Uganda, for example, packaging costs account for 50 percent of the product value for the export of a pallet of bananas. Additional packaging costs can create trade costs that reduce market access to countries with di erent pallet sizes. Source: Raballand and Aldaz-Carroll 2005. in export markets, including 76 percent of firms in Latin America and the Caribbean. In all countries, firms consider foreign technical regulations to hinder the expansion of trade. Many of the firms facing technical regula- tions have additional compliance costs for each export market, which in total make up between 1 and 10 percent of investment costs. These compliance costs most often concern investments in additional plant or equipment (38 percent of firms), product redesign, and contracting of additional workers. On average, investments to comply with technical regulations amount to 4.4 percent of total sales. Another study of technical barriers to trade by Chen, Otsuki, and Wil- son (2006) attempts to quantify the impact of standards and technical Economic Impact and Effect of Quality and Standards 35 Box 2.5 Food Safety Standards in the European Union New EU food safety regulations that came into force in 2005 include stringent reporting requirements for developing-country farmers. As reported by Wal- lace (2004), the new regulations "make it mandatory for all fruit and vegetable products arriving in the EU to be traceable at all stages of production, process- ing and distribution." EU assistance to help farmers meet these stringent new standards, which involve tracing production back to the seed, is reported to be inadequate. Consequently, many developing-country farmers risk being closed out of this important market. An additional problem is that the standards are applied in a discriminatory fashion, and meeting them requires specialized skills and equipment beyond the capability of most low-income countries. Such a tightening of standards can have economic rami cations. Otsuki, Wilson, and Sewadeh (2001) examined proposed EU standards for a atoxin (toxic compounds produced by molds) in food exports from Africa. The authors estimated that the standards, which would reduce EU health risks by less than 2 deaths per billion per year, would decrease African exports of cereals, dried fruits, and nuts by 64 percent ($670 million). Source: Goldin and Reinert 2006. regulations on the export performance of firms. The study is based on the above-noted survey of 17 developing countries. The survey collected information on the perceived effects of standards and technical regula- tions on the ability to export. Using these data and a model of a firm's decision to export, the study suggests that testing procedures reduce export share by 9 percent. Meeting standards also affects a firm's abil- ity to enter foreign markets, reducing a firm's likelihood of exporting in more than three markets by 7 percent. While the previous studies offer valuable insight into the perceived effects of standards on trade at the firm level, they do not establish a solid relationship between standards and trade at the macro level. Nonethe- less, a number of empirical studies have attempted to do just that, and they have delivered interesting results. Box 2.5 on food standards in the European Union (EU) presents an example of how setting standards can have a significant impact on trade. The balance between the costs and benefits needs to be carefully evalu- 36 Quality Systems and Standards for a Competitive Edge ated so the right balance is achieved, and remedial measures to respond to new requirements are available and do not become an insurmountable burden. A review of empirical studies finds that there is mixed evidence on the role of idiosyncratic standards but that shared or harmonized standards are trade-promoting (table 2.3). The empirical evidence does not pro- duce a strong consensus on the role of country-specific standards, and it reveals their ambiguous effect on trade. Idiosyncratic standards promote trade by reducing information costs but hamper trade by introducing adaptation costs, so it is the relative weight of these two effects that will determine the net trade effect. However, the empirical evidence shows that the adoption of harmonized standards by trade partners increases imports, exports, or both. When countries use the same standards, many of the trade-hampering effects of standards are eliminated. Exporters no longer need to adapt their products for other markets, information on standards is more easily available to exporters, and economies of scale promote the exploitation of national competitive advantages. Blind and Jungmittag (2005) perform a panel study of bilateral trade flows between Germany and the United Kingdom between 1980 and 1995 in 31 sectors to ascertain the causality of the relationship between different types of standards, exports, and imports. The model, which con- siders both German and British standards, is based on trade equations that incorporate factors of technological competitiveness and macroeconomic indicators. Blind and Jungmittag find evidence that it is specifically har- monized standards that enhance trade. International standards adopted by Germany have a significant positive effect on German exports, whereas German idiosyncratic standards have a negative effect on German exports. These results support the hypotheses that national idiosyncratic standards create a competitive disadvantage for exporters, and that, conversely, international standards represent a competitive advantage for German producers. As for imports, German international standards have a slightly positive effect on German imports from the United Kingdom, whereas national German standards have no impact on imports. In another study, Swann, Temple, and Shurmer (1996) perform an econometric analysis of the effects of standards on UK trade performance. Their study considers total intra-industry trade in the United Kingdom in 83 manufacturing sectors from 1985 to 1991. British standards in each sector are introduced in the trade equation, and although this study is not restricted to trade with Germany, German standards are also intro- duced as a measure of international best practice. These authors find that Table 2.3 Empirical Results on the Effect of Idiosyncratic and Harmonized National Standards on Trade Effect of idiosyncratic standards Effect of harmonized standards Trade Trade Authors Model Imports Exports volume Imports Exports volume Blind and Jungmittag (2005) Time series, 1980­95, bilateral trade, 0 ­ ­ + + + 2 countries Swann, Temple, and Shurmer (1996) Time series, 1985­91, bilateral trade, + + + + + + 2 countries Moenius (2004) Time series, 1985­95, bilateral trade, + n.a. + + n.a. + 12 countries Chen and Mattoo (2004) Time series, 1986­2001, bilateral trade, n.a. n.a. n.a. + + + 42 countries Hummels and Klenow (2005) 1995 data, bilateral trade, 126 countries n.a. n.a. n.a. + + + Source: Authors'compilation. Note: + = positive e ect; ­ = negative e ect; 0 = no e ect; n.a. = not available. 37 38 Quality Systems and Standards for a Competitive Edge an increase in the number of either idiosyncratic or international stan- dards has a comparable positive impact on the trade ratio, but that the stock of idiosyncratic standards encourages imports and exports more than the stock of international standards (figure 2.4). Specifically, 100 additional idiosyncratic British standards increase imports by 34 percent and exports by 48 percent, while 100 additional international standards increase imports by 3 percent and exports by 24 percent. The study sug- gests that the weaker effect of international standards on trade may be due to the specific nature of these standards. If international standards are mainly variety-reducing standards, the negative effects of reducing variety on intra-industry trade may offset the positive effects of sharing standards. Moenius (2004) uses a slightly different approach to investigate the role of shared versus idiosyncratic standards. Standards documents adopted in a given country usually include links to the international stan- dards to which they are related, if any. The author's measure of shared standards is based on a count of the number of links between similar standards in two countries. This contrasts with the traditional approach used by Blind and Jungmittag (2005) and Swann, Temple, and Shurmer (1996), which counts the number of international standards in each country and assumes that these standards are shared on a bilateral basis. Figure 2.4 Effects of British Standards on UK Exports, Imports, and Trade Balance, 1985­91 6 UK exports 5 UK imports UK trade balance 4 ease 3 incr % 2 1 0 Original Internationally BSI standards equivalent BSI standards Source: Swann, Temple, and Shurmer 1996. Note: Values shown represent percentage increases in trade caused by the adoption of 10 additional standards in the United Kingdom. At the time of the study, British standards were administered by what was then called the British Standards Institution (BSI), the precursor of BSI British Standards. Economic Impact and Effect of Quality and Standards 39 Moenius uses a gravity-based model, which predicts that the volume of trade between two countries will be proportional to their economic mass, usually measured by GDP, and inversely proportional to the physical dis- tance between them and other obstacles to trade. The model uses sectoral bilateral trade data for 12 countries of the Organisation for Economic Co- operation and Development (OECD) during the time period 1985­95. The results show that trade volumes are higher for countries that share more standards. On average, a 1 percent increase in the number of shared standards increases the bilateral trade volume by 0.3 percent. This implies a $6 billion increase in U.S. trade volume for a 1 percent increase in the number of shared standards between the U.S. and its trading partners. When the analysis is repeated for imports only, both shared standards and country-specific standards are also found to promote trade. An examination of the effects of standards at the industry level pro- vides additional insight into the effect of each type of standards. The results show that for simple product sectors, such as food and bever- ages or crude mineral fuels, the stock of either country-specific or shared standards in the importing country hampers imports, while for complex manufacturing sectors, both types of standards promote imports. In sum, the Moenius study reveals that country-specific standards are not neces- sarily a barrier to imports. In technologically complex sectors, informa- tional requirements to export in foreign markets are high, so standards reduce information costs by documenting technical specifications and market preferences. In simple industrial activities, products are relatively homogeneous and informational requirements are low, so the reduction of information costs through standards will not outweigh product adap- tation costs. However, while this explains why country-specific standards have a positive effect on trade in manufactures and a negative effect on trade in nonmanufactures, it does not fully explain why shared standards should negatively affect trade in nonmanufactures. In another study, Chen and Mattoo (2004) examine bilateral trade among 42 countries to determine the effect of shared standards on trade. This study takes a unique approach in that it does not actually count shared standards but quantifies the extent of standards harmonization by identifying harmonization directives in different industrial sectors. The model also includes regional trade agreements and mutual recog- nition agreements.6 The time-series analysis yields a statistically signifi- cant relationship between shared standards and trade. The results imply that a harmonization directive implemented by two countries raises bilateral imports in the concerned industry by 32 percent and imports 40 Quality Systems and Standards for a Competitive Edge from countries outside the harmonizing region by 10 percent. This study supports the argument that shared standards reduce the transaction and adaptation costs of trade. However, because this study relies on harmoni- zation directives, it does not cover all shared standards and is principally concerned with mandatory standards. Hummels and Klenow (2005) provide evidence of the impact of qual- ity on export performance. Using 1995 data on exports from 126 coun- tries to each of 59 importers in over 5,000 six-digit product categories, they show that large exporters are able to sell large quantities at higher prices, consistent with producing higher-quality goods. Their wider set of goods accounts for 62 percent of the greater exports of larger econo- mies. By exporting higher-quality goods, richer economies can export larger quantities without lowering the prices of their varieties on world markets. Quality is a demand shifter, raising the quantity a country can export to a market at a given price. This study also finds that quality dif- ferences can account for 9 percent of country differences in real income per worker. To summarize, the empirical evidence in the literature supports the trade-enhancing role of harmonized standards but does not produce a strong consensus on the role of country-specific standards. Nonethe- less, there are important limitations to the methodologies used in these analyses. A first caveat is that apart from the Chen and Mattoo (2004) study, the studies mainly concern developed countries. As suggested by Moenius (2004), the different economic structures of developed and developing countries could lead to significant difference in the effects of idiosyncratic standards. For instance, the content of a standard depends on the process used to develop the standard. In theory, countries with more mature standards bodies should have a lesser tendency to create trade-distorting standards. A second caveat is that it is not completely appropriate to examine the role of different standards by measuring the number of standards because a standard's economic impact varies highly with its content. Harmonized standards and idiosyncratic standards could have different effects on trade simply because of the nature of their content. Also, different coun- tries have different methods of documenting technical specifications in standards. The same specifications may be included in a single standard in one country and distributed among 10 standards in another. These two cases would register different numbers of standards even though the amount of standardized information is effectively the same in both countries. Finally, many countries harmonize their standards for political Economic Impact and Effect of Quality and Standards 41 reasons, or as part of a regional trade agreement. While these standards may be completely irrelevant to a country's economy they would still be counted as shared standards in the models. Standards and Quality as an Entry Point for SMEs: The Path to Industrial Upgrading With the new organization of global trade, buyer-driven value chains are playing a growing role in exports of manufactured goods to devel- oped countries. Value chains are composed of enterprises that add value to a product or service through discrete, though interrelated, activities involved in the production and distribution of goods and services. For example, in the case of the apparel industry the value chain extends from "raw materials (e.g., cotton or petrochemicals), to the production of nat- ural or synthetic fibers and textiles, then to the design, cutting, assembly, laundering, and finishing of apparel, and, finally, to the distribution, mar- keting and retailing of garments" (Bair and Gereffi 2001). In buyer-driven chains, "large retailers, brand-named merchandisers, and trading compa- nies play the central role in shaping decentralized production networks in a variety of exporting countries, frequently located in the periphery. Buyer-driven value chains are of particular relevance to the integration of small developing countries' exporters to the global economy. This pat- tern of industrialization is typical in relatively labour-intensive consumer goods such as garments, footwear, toys and housewares" (Gereffi 1994 quoted in Schmitz 1999).7 Firms in developing countries that adopt standards and quality practices can find opportunities for upgrading in buyer-driven value chains (Kaplin- sky and Readman 2001).Increasingly,global trade in low-cost manufactures is characterized by an ever-expanding pool of producers that compete for an ever-more-concentrated group of buyers in developed countries. Global buyers often do not own any production facilities of their own but play pivotal roles in coordinating decentralized production processes. These global buyers are the gatekeepers to the markets in developed countries and control strategic segments of the value chain, such as research, design, sales, marketing, and branding. As in other forms of global production net- works, conformance to standards and quality requirements is a precondi- tion for participation in buyer-driven value chains. Participation in buyer-driven chains can also facilitate industrial upgrading and innovation. Upgrading can be defined as the transition to higher-value-added activities. Much of it involves innovation, because 42 Quality Systems and Standards for a Competitive Edge the firm must learn to develop new processes, products, and services. Humphrey and Schmitz (2000) distinguish three types of upgrading: 1. Process upgrading is transforming inputs into outputs more efficiently by reorganizing the production system or introducing superior tech- nology. 2. Product upgrading is moving into more sophisticated product lines in terms of increased unit values. 3. Functional upgrading is acquiring new, superior functions in the chain, such as design or marketing, or abandoning existing low-value-added functions to focus on higher-value-added activities (see also Giulani and Bell 2004). Upgrading is not only necessary to increase earnings, but offers an opportunity for a firm to react to competition from other low-wage countries and globally mobile capital as well as unpredictable mar- ket demand and stringent product requirements. To do so, firms need to adopt dynamic learning and innovation-based business strategies. In developing countries, remaining competitive without upgrading often involves squeezing wages and profit margins. Japan, the newly industrial economies, and China all achieved their initial export success through functional upgrading, in addition to process and product upgrading. These countries started as mere assemblers of imported inputs for labor- intensive consumer products and gradually moved to higher-value-added activities such as full-package supply and original equipment manufactur- ing production, where the supplier is involved in producing to the speci- fications of the buyer. Some of these countries moved beyond original equipment manufacturing into original brand-name manufacturing by combining their production expertise with design and marketing activi- ties. Functional upgrading, as has occurred in East Asia, is more difficult to achieve than product or process upgrading, but examples of successful functional upgrading can still be found in many buyer-driven chains (box 2.6). According to Gereffi (1999), "participation in global commodity chains is a necessary step for industrial upgrading because it puts firms and economies on potentially dynamic learning curves." Buyers in value chains play an active role in transmitting technical and organizational knowledge to their suppliers. Local producers have ben- efited a great deal from buyers' knowledge about improving production Economic Impact and Effect of Quality and Standards 43 processes, attaining consistent product quality, developing new prod- ucts, and increasing delivery times (Humphrey and Schmitz 2000). For example, in the 1970s, when Brazil's shoe industry was still at an incipi- ent stage, international buyers maintained a substantial technical staff in Brazil and played a key role in helping firms reach international quality and delivery standards (Schmitz 1999). Today, support has diminished but buying agents for foreign retailers still actively support Brazilian foot- wear producers in product and process upgrading. In a study of the Sinos Valley cluster, Brazil's main footwear cluster, Bazan and Navas-Alemán (2003) found that approximately half of firms that serviced the U.S. mar- ket had introduced new lines in this market, mostly at higher prices. The producers acquired new skills by developing these new lines. East Asian economies were able to transition to higher-value-added activities by using their tight linkages to buyers in value chains to receive technology and knowledge embedded in inputs, equipment, and technical advice. Producers of higher-quality goods benefit from much tighter relation- ships with global buyers or transnational corporations (TNCs). Buyers in quality-driven value chains need to engage in intense communication with their suppliers, instruct them on specifications, and assist them with technology transfer to ensure that they produce high-quality products. Having invested in their suppliers, these buyers are unlikely to change partners for the sake of lower short-term prices because they would face high switching costs (Schmitz and Knorringa 2000). Evidence from the Costa Rican electronics industry showed that when TNCs sourced high- quality inputs from local SMEs, this resulted in stable relationships and cooperation with the public sector to assist in SME upgrading (Monge- González, Rosales-Tijerino, and Arce-Alpízar 2005). Experience from Brazil's Sinos Valley footwear cluster showed that buyers that focused more on quality needed to work closely with suppliers to meet the desired standards, and these close relationships made local suppliers less vulnerable to competition from low-cost countries. Furthermore, firms that had benefited from buyers in higher-quality export markets were able to use their knowledge to penetrate new export markets with their own brands and designs (Vargas and Alievi 2003). Another interesting example of adaptation to market requirements and its effects is the case of grapes in India (box 2.7). 44 Quality Systems and Standards for a Competitive Edge Box 2.6 Standards and Quality as an Entry Point to Upgrading in a Horticultural Value Chain In the past two decades the fresh vegetable industry in the United Kingdom has increasingly taken the form of a buyer-driven value chain. This restructuring has created both new challenges and new opportunities for Sub-Saharan African exporters. On the one hand, they are faced with a growing number of standards and quality requirements that they are expected to ful ll. On the other hand, the restructuring of the vegetable trade has opened up new opportunities for upgrading and innovation. In the 1960s, 90 percent of fresh fruits and vegetables in the United Kingdom were traded in wholesale markets. There were no direct relationships between supermarkets and growers. As a result, barriers to entry were low. Supermarkets could not impose speci c requirements on their suppliers, and suppliers in turn had no incentives to invest in meeting the requirements of any particular super- market. Relationships were loose across all links in the value chain, and most exporters of African horticultural products purchased their vegetables through spot markets in rural areas. In the 1990s, the fresh fruit and vegetable trade experienced a dramatic restructuring, and supermarkets began to bypass wholesale markets to buy directly from producers. This allowed supermarkets to enhance their coordina- tion and control of the supply chain. These changes occurred in a context of demand for greater product variety, product innovation, and increased packag- ing and processing. Supermarkets began to view the quality of fresh produce as the most important factor in building a competitive advantage. At the same time, the UK supermarket industry was becoming increasingly concentrated, with the top four retailers accounting for nearly 75 percent of all food sales. The change in structure to a buyer-driven value chain and the growing focus on quality induced supermarkets to seek tighter and more exclusive relation- ships with exporters and growers. Sub-Saharan African countries were par- ticularly a ected by this restructuring because their vegetable exports to the European Union had grown signi cantly in the 1990s (by 151 percent between 1989 and 1997).The largest supermarkets, such as those in the United Kingdom, purchase 70­90 percent of their fresh produce imports from Africa, from a large number of suppliers. Economic Impact and Effect of Quality and Standards 45 The change in chain governance meant that supermarkets increasingly speci ed both product and process parameters all along the chain. Quality, regulatory, environmental, and labor requirements guided all the activities, from how products were grown to how they were transported, stored, pro- cessed, and packed. All export rms in Africa were required to have sophisti- cated quality assurance systems that documented all the steps from plant- ing to shipping. Standards included both company standards and external standards developed by trade associations, consortia of trade unions, non- governmental organizations, and enterprise associations. The supermarkets demanded regular audits of production and processing facilities to control and monitor compliance with the standards. This was done either through the supermarket's own sta or through third-party auditors from recognized certi cation bodies. The restructuring of the horticultural value chain thus brought many challenges to African growers and exporters as they were obliged to invest in new skills and equipment to meet quality and process requirements. However, these buyer-driven value chains have also presented African coun- tries with new opportunities for industrial upgrading. Not only have growers and exporters acquired new capabilities in areas of quality control, logistics, storage, distribution, and transport, but processing activities have also been transferred from UK importers to African exporters. These include simple tasks like washing and trimming and more technically complex activities such as packaging, bar- coding, and labeling. Processing functions in Africa have recently extended to the production of ready-to-eat foods, which requires high hygiene levels. It is cheaper for supermarket chains to push back these processes to Africa because they are labor-intensive, repacking in the United Kingdom is avoided, and pro- cessed products have a higher value-to-weight ratio, which reduces transport costs. Participation in the horticultural value chain has also opened up new opportunities for innovation. Many importers work closely with their African suppliers to develop new products, packaging, and presentations. Innovation projects have led to the development of long-term relationships between sup- pliers and buyers through which African exporters have acquired signi cant technical and market information from European buyers and have minimized their risks of substitution. Sources: Dolan, Humphrey, and Harris-Pascal 1999; Dolan and Humphrey 2004. 46 Quality Systems and Standards for a Competitive Edge Box 2.7 Grapes in India: Changing Market Requirements for Standards and Quality Exports of grapes from India have been largely stagnant over the past decade. While a mix of internal and external factors is responsible for this trend, non- tari barriers are an increasingly critical factor (Chaturvedi and Nagpal 2003). To export to the European Union, grape-producing farms have to meet the standards issued by EurepGAP, an initiative of retailers belonging to the Euro- Retailer Produce Working Group. EurepGAP's aim is to develop widely accepted standards and procedures for the global certi cation of good agricultural prac- tices (GAP). EurepGAP's July 2003 standards require that exporters meet a set of conditions pertaining to worker training, planning and production, pesticide record keeping and disposal, testing for pesticide residues at the farm, and post- harvest operations. One estimate contends that compliance with EurepGAP increases production costs by 40 percent (Chaturvedi and Nagpal 2003). Physical and chemical characteristics are used to determine grape quality. Standards vary by country. 1. United Kingdom. The United Kingdom has set rigid standards. The minimum acceptable diameter of an individual grape is 18 mm, and the berry must be a "light, milky green." Grape bunches must be a uniform 350 to 700 grams and packages are to hold 9 kilograms. British grocers sell grapes by the bunch, not by weight, the practice in India. 2. European Union. The minimum acceptable diameter of the berry is 16 mm. The speci ed color is a light, milky green.The packaging has to be a uniform 4.5 kg. 3. Middle East. India's main export market is an extension of the local market and is lax on standards. Given India's small-scale production, individual farmers are hard-pressed to meet demand for quality grapes. Moreover, the e ort increases costs by an additional Rs 50,000 to Rs 75,000 per hectare. Pruning is the most important practice in quality grape production. To achieve market-speci ed sizes for bunches and berries, yields must be sacri ced. Growers must thin shoots and clusters and apply gibberellic acid, a growth hor- mone that promotes elongated clusters. Berries themselves must be thinned and girdled. These practices reduce the number of bunches per vine, giving the fruit ample space to attain the required size. Thus yields are much lower in vine- Economic Impact and Effect of Quality and Standards 47 yards that cater to the export market. The average yield for a vineyard catering to the local market is about 35 tons per hectare; from vineyards that cater to the export market, the yield drops to about 25 tons per hectare. Besides lower yields, the quality of the export and the exportable volumes also matter. Even in for-export vineyards, only the very well maintained ones are able to ensure that the entire available yield is exportable. In other vineyards, on average, only about 80 percent are of export quality--i.e., about 20 tons/ hectare. It has been estimated that from most export-oriented vineyards, about 8 tons would meet the United Kingdom's quality standards; the remaining 12 tons could be exported to other EU countries. The color speci cations ("milky green") of the UK and EU markets require cer- tain changes in preharvest practices--namely, the use of shade or light-cutting nets. Exposure to sunlight turns the berries golden, an acceptable color for the domestic market but deemed undesirable abroad. The costs of these shade nets must be borne by the farmer. To achieve the desired size of berries and bunches, grape farmers must employ techniques that lead to crop suppression, namely the application of syn- thetic chemicals and arid plant-growth regulators, the residues from which may be toxic. All pesticides have minimum residue limits prescribed by the importing countries. The UK standards are more stringent than those of other countries and in fact are stricter than the international CODEX standards. To keep pesticide residues within the limits, there is a waiting period prior to harvesting. Because this waiting period is di erent for each pesticide, spraying must be done according to a special, sequenced schedule. Grape growers normally require technical assistance. The application of chemicals also adds to the costs by way of testing and certi cation, practices required for the European market. Grape exports require two kinds of certi cation: 4. The EurepGAP certi cation. This is a mandatory certi cate needed to market produce in Europe.The certi cate is based on three safety criteria: labor, envi- ronmental, and consumer. 5. Certi ed pesticide-residue tests. The cost of testing and certi cation for pesticide residue is currently Rs 7,000/sample/pesticide.The number of tests undertaken depends on the export destination. Countries with more stringent measures, like the United Kingdom, require more tests, incurring higher costs. continued on the next page 48 Quality Systems and Standards for a Competitive Edge Box 2.7 Grapes in India: Changing Market Requirements for Standards and Quality--continued In addition to certi cation, other postharvest processes for export grapes include grading, packing, precooling, cold storage, and transport. Grapes are immediately transported to the packhouse after harvesting and then sorted and graded. Bunches are put in polypacks, which are placed in properly cush- ioned boxes treated with grape guard, according to market requirements. At each stage, the proper temperature and humidity levels have to be maintained, which constitutes a major bottleneck to growth in exports because of infra- structure limitations. The investments are substantial and normally beyond the means of individual farmers. Facilities are currently provided either by coopera- tive societies or private enterprises. In consequence of the above-mentioned issues, new challenges are emerg- ing for fruit and vegetable exports. The now-mandatory EurepGAP certi cation is costly, with a registration charge of Rs 35,000 per farm, making exports pro- hibitively expensive for the small grape grower. Pesticide-residue limits (set by importing countries) are growing more stringent, and satisfying them creates additional costs and delays. In light of those requirements the Indian grape sec- tor has been adapting and responding to the challenge and the government has assisted in its corresponding jurisdiction to facilitate the response.The grow- ers'association and other organizations are seeking to meet those challenges in order to keep the grape sector competitive and pro table. APEDA [Agricultural Economic Impact and Effect of Quality and Standards 49 and Processed Food Products Export Development Authority], for example, has developed a pesticides-monitoring program, while the Maharashtra state gov- ernment is arranging to issue pesticide-residue certi cates. By 1995, the main producer of grapes, Mahagrapes, and the cooperatives had the infrastructure and knowledge they needed to produce and export quality grapes. Rejected consignments fell to less than 10 percent by 1995. But while Mahagrapes was able to build brand recognition, exports fell from 1,800 tons in 1991/92 to 357 tons in 1994/95. The drop grew out of a conscious decision to export only quality produce. Since 1998/99, exports of Mahagrapes have increased to about 800 tons, about 2 percent of the EU market during the period. The prices commanded for Mahagrapes have steadily improved over time. Mahagrapes, in short, succeeded in establishing a system for the export of high-value perishables from India. Mahagrapes is extending its product line to mango, pomegranate, strawberries, and other fruits and vegetables. Ten new cooperatives in the state of Maharashtra have recently emerged to export grapes, and it is estimated that cooperatives now account for 35 percent of total grape exports. Many producers are also exporting grapes individually. These and other exporters, a total of 150, have gained from association e orts to building a grape-cultivation and -export knowledge base. India now is able to send about 60 percent of its export produce to the high-value EU market and capture about 35 percent of that market during the export season. Source: Excerpted from Naik 2006. C H A P T E R 3 The Jurisdiction Issue: Roles of the Private and Public Sectors As shown in chapter 2, different types of standards can have differ- ent economic impacts, often acting in opposite directions. One factor that is likely to influence the economic effects of standards is the stan- dards development process itself. The actors, processes, and institutions involved in elaborating standards may ultimately shape their contribu- tion to technical change and economic growth. Standards can be classified into categories according to their origin and the type of cooperation mechanism. There are two main categories, market standards and formal standards. Market (de facto) standards are the result of industry self-regulation. There is no explicit mechanism for cooperation between interested parties; these standards evolve through a series of unilateral, irrevocable purchasing choices. One agent chooses first, and others follow with their choices based on previous choices and personal preferences, creating a bandwagon effect. There are three basic types of market standards: 1. Market standards can be unsponsored if they do not involve any pro- prietary rights or identifiable author and are accessible to all market participants. This is often the result of an iterative design process based on uncoordinated collective innovation. The four filing holes commonly found on A4 paper are an example of an unsponsored de facto standard. Although the location and dimension of the two center holes are specified in an international standard (ISO 838), the 51 52 Quality Systems and Standards for a Competitive Edge two other holes, located 80mm above and below the center holes, are not documented in any standard or patent. 2. Sponsored market standards involve the proprietary rights to a set of technical specifications, where a specific design wins a position of market dominance. The owners can determine their specifications and extract rents from their use. An example of a proprietary stan- dard is the set of technical specifications for Adobe Acrobat's Por- table Document Format (pdf). Files in this format can only be read using Adobe's Acrobat Reader software. 3. A growing number of market standards have been established through industry consortia. These are voluntary but exclusive alli- ances of organizations or individuals that coordinate specific indus- try standards. The archetypal consortium is the Internet Engineering Task Force, which manages the Internet. Formal (de jure) standards involve an explicit coordination and nego- tiation process prior to commitment to a particular standard. These stan- dards are of two types: 1. Voluntary standards can be developed by interested groups through voluntary standards development organizations. Here stakehold- ers, including both producers and consumers, agree on a standard through an open committee. This contrasts with the exclusive nature of standard development in industry consortia. Formal standards are used, for instance, to specify the material, size, and strength of bolts. Any interested party is in principle free to use these standards. 2. Mandatory standards or technical regulations are elaborated by govern- mental authorities and their application is compulsory. Ideally, they occur when externalities are so strong that the private incentives to standardize are suboptimal. This is the case of standards that regulate the environmental impact of products and processes. The individual costs of complying with a standard may outweigh individual benefits because the detrimental effects of environmental contaminants may be distributed over a large number of people and over time. Coordination Mechanisms for Technology Adoption Market standards can have divergent effects on technical change, depend- ing on market structure and corporate strategy. If a variety of standard- ization choices are available and consumers favor producers who are The Jurisdiction Issue: Roles of the Private and Public Sectors 53 offering the best technology, then market forces may drive convergence toward a single industry standard using that technology. But many factors can steer markets away from the best technology. Often, it is not techno- logical choice but corporate strategy that determines who wins the stan- dards race. In that case, market standardization can lead to the adoption of a suboptimal technology as an industry standard. In the early 1980s, although Sony's Betamax format for video recorders was technologically superior to JVC's VHS format, JVC's ability to license the technology to other manufacturers eventually overwhelmed Sony and led VHS to become the dominant standard. The market mechanism to generate standards incorporating new tech- nologies can be particularly weak when it is confronted with network effects. Individual decisions to participate in a network only consider personal ben- efits, not the benefits to the network community. Costs may discourage participation in a network when there are few members in the network and the individual benefits are small. Hence, a new technology that derives its value from being part of a network will only be accepted if a sufficient number of economic units have already accepted it. This may create a sub- optimal equilibrium where society is locked into a network using older, less efficient technology. The lock-in effects and high switching costs inherent to the video recorder market ensured that once VHS became the standard it would be difficult for new entrants to compete in this market. In fact, it took two decades before a new technology, the DVD, could challenge VHS as the de facto standard. In the absence of perfect competition, dominant firms with installed bases are likely to prevail in standards races because there are often increasing returns to cumulative adoptions. Firms can also use anticompetitive practices to gain a critical mass of customers and tip the market toward a preferred proprietary standard. Market forces may lead not only to a suboptimal standard but to understandardization, where various technologies coexist in the market. While in some cases, variety in the market reflects consumer preferences, in other cases, adoption externalities hinder the convergence toward a single standard because social costs and benefits are not taken into account by consumers or producers in their individual choices. When market mechanisms fail to yield an optimal solution, coopera- tive committees can be used to develop standards that make a better outcome possible. For example, cooperative committees can remedy lock-in effects by coordinating the diffusion of new technologies. Open participation in the standardization process enhances the likelihood that all relevant information will be included in the cost-benefit analysis 54 Quality Systems and Standards for a Competitive Edge of competing standards. To be effective, a committee should include a diverse set of actors with competencies in technological, social, and eco- nomic issues. Industry participants should represent both the supply and demand sides of the market. If committee decisions are based on con- sensus and voting rights do not depend on market power, the process becomes more egalitarian than a market-based approach. However, there are limitations to standardization through open volun- tary associations. In particular, open participation can slow standardization because it requires coordinating a large number of participants, document- ing the process, and organizing reviews. This can be long and cumbersome. Moreover, it is often the case that not all participants have a genuine inter- est in contributing to the process. The knowledge spillovers involved in the open standardization process may encourage participants to free-ride. When participants have vested interests in incompatible positions, the pro- cess can become long and costly. In contrast, industry consortia are much quicker and more efficient because they involve fewer participants and these participants tend to be like-minded. The use of consensus in vol- untary committees can also lead to strategic behavior, where participants falsely represent their preferences. Minority interests can be favored over the interest of the group as a whole.While latecomers in the market would accept a new technology, users of an old technology could block a change that would be associated with high switching costs. However, if majority rule is adopted instead of consensus, decisions will be reached quickly but solutions will tend to ignore individual interests. Finally, open participa- tion is not always a viable option because it requires the involvement of all the players in the market. Market leaders may not wish to participate in a formalized standardization process if they can instead invest in new proprietary standards without incurring too much risk. Government Involvement in Standardization The strong "public good" element of a standard can sometimes justify government involvement. Formal standards are nonrivalrous because anyone can benefit from them without diminishing their utility to oth- ers, and they are nonexcludable because once they are created anybody can use them. The economic rationale for government intervention in standardization is that a free market is unlikely to produce the optimum amount of a public good. The market may provide for too many stan- dards, as in the case of incompatible measurement units, or standardiza- tion of the wrong sort, as in the case of the QWERTY keyboard.1 The Jurisdiction Issue: Roles of the Private and Public Sectors 55 Information is a public good, and this also suggests a role for govern- ment. Knowing the characteristic of standards required by users or estab- lished in norms by governments in diverse potential markets is key for access to those markets. The acquisition of that information is costly: it is not efficient for each producer to invest in that endeavor, and coordina- tion and cost sharing are difficult. This points to a role for government in the process. Likewise, government can also play a role in raising aware- ness about impacts of the adoption of standards. Governments can support standardization by ensuring open partici- pation and a balanced representation in voluntary committees. Many of the advantages of formal standards over market standards stem from the fact that they incorporate the views of a much broader range of stakeholders. However, uneven representation in voluntary standards committees can occur for a variety of reasons and can lead to short- sighted and biased standards. Large firms, which tend to benefit more than small firms from the economies-of-scale benefits of standards, are more likely to have the resources and personnel necessary to partici- pate in the standards-setting process. As a result, large companies are often overrepresented in voluntary committees and can capture the standardization process. Standardization may also exclude potential stakeholders because it requires significant technical expertise. While consumers are directly affected by standards, they are unlikely to have the qualifications to participate in the standardization process. Often, consumers form a large and uncohesive group, which makes it difficult for them to organize and enhances their tendency to free-ride the stan- dardization process. Hence, producer interests are likely to be more rep- resented than user interests in standards committees. A possible role for the government is to help balance standards committees by promoting the involvement of marginal participants or by representing excluded interests in the committees. Governments can also play a role in correcting market failures in the diffusion of formal standards. Some standards are network goods, and the value of adopting the standard increases with the number of adopters. If the cost of adopting the standard is high, the market may never reach the critical mass of users necessary for the individual benefits of standards to exceed their costs. Many standards also involve significant adoption bar- riers caused by a lack of technical skills, time, or resources, which make them poorly understood and difficult to adopt. Governments can com- pensate for these market failures by creating programs and policies that promote the widespread diffusion of standards. 56 Quality Systems and Standards for a Competitive Edge Mandatory standards can provide an optimal solution when private agents would not be able to internalize the benefits of standardization. This is often the case when there are so many stakeholders that a com- mittee process would be unmanageable. Government bodies can be used to represent the interests of all members of an economy in the standardization process. Rather than relying on voluntary coordination committees, governments can enforce the adoption of standards through binding regulations, with punitive measures for nonobservance. Manda- tory standards may have a much stronger effect in terms of decreasing product variety because only products and processes conforming to these standards can circulate in the market. This may be desirable in domains where negative externalities are high, such as in safety, health, environ- mental protection, and consumer protection, as well as where positive externalities are high, such as in markets of network goods. Nonetheless, decreasing variety can have detrimental effects on innovation because it places constraints on the development of new technologies. Government standards can also result in suboptimal solutions if the state does not have the necessary technological foresight to make informed decisions. National Standards Bodies International experience shows that a wide variety of institutions can be used to organize the standards development process.2 There is no dominant system, and most standards bodies cannot be neatly classified into specific categories according to typology (Stephenson 1997). The standardization process can be more or less centralized, the standards organizations can be either public or private, and there may be more or less juxtaposition of vol- untary and mandatory standards development within a single organization. In the European Union (EU), most countries have adopted a very cen- tralized approach to standardization. A central national organization is involved in developing all of the national standards (figure 3.1). Typically, the organization implements a work program and delegates the techni- cal aspects of standards development to technical committees consist- ing of representatives from government agencies, the productive sector, consumer associations, academia, and research institutions. Members of technical committees are usually volunteers who do not receive remu- neration from the standards body for their work, but they may be spon- sored by their organizations. Each technical committee is responsible for developing standards in a specific area and may be formed and dissolved by the standards body as the need arises. The Jurisdiction Issue: Roles of the Private and Public Sectors 57 Figure 3.1 The Centralized Approach to Standards Development Member of the primary standards body or technical committee Proposes standard development Revises drafts Primary standards body based on public Assigns comments standard development to technical committee Technical Technical Technical Technical Technical Public committee committee committee committee committee Elaborates a standard Submits draft drafts for public review Primary standards body Votes on draft approval Approves standard Standards Publishes standard Source: Authors'elaboration. Members of the standards organization, often members of a technical committee, propose plans for new standards, but proposals can some- times come from outside the standards development organization. If there is sufficient support for a plan in the standards organization, the technical committee begins to study and elaborate a standard. Once the technical committee has reached a consensus, a draft of the standard is submitted to a vote by members of the national standards body. If approved, the standards body then subjects the draft to public enquiry. During the public review process, the draft is typically made available to the public for two months. Once the technical committee has revised the draft to incorporate public comments, the standards body finalizes, adopts, publishes, and distributes the standard. In the United States, by contrast, the standardization process is extremely decentralized and market-oriented. Although the American National Standards Institute (ANSI) serves as a coordinating body and 58 Quality Systems and Standards for a Competitive Edge accredits standards bodies, it does not itself develop standards, and no single organization has oversight over the entire voluntary standards pro- cess. Instead, more than 600 private sector organizations develop volun- tary standards (DeVaux 2001). These include professional and technical organizations, trade associations, research and testing institutions, and oth- ers. The individual standards bodies follow basically the same standards development process as is used in the EU: they work through technical committees, approve drafts by vote, and then submit the drafts to public enquiry through ANSI (figure 3.2). ANSI is not involved in determining Figure 3.2 The Decentralized Approach to Standards Development Primary standards body Accredits standards Member of Proposes development the primary standard organizations standards development Requests revisions body or based on technical Accredited public Accredited Accredited committee organization comments organization organization #1 #2 #3 Public Prepares Primary Technical Technical Technical review standard standards Public committee committee committee drafts body Votes on draft approval Accredited organization Submits revised Submits draft #1 standard for approval for public review Verifies the standard Primary development process standards and approves the standard body Accredited Publishes standard organization #1 Standards Source: Authors'elaboration. The Jurisdiction Issue: Roles of the Private and Public Sectors 59 the actual content of standards and does not even monitor whether stan- dards conflict. Rather, accredited standards bodies must document the development process of each new standard and ANSI performs periodic audits seeking evidence of compliance with the approved development procedures. The individual standards bodies then publish and sell the standards. Canada and Mexico have adopted similar decentralized sys- tems of standards development. The legal status of standards bodies varies widely across countries. These bodies can be part of the public sector, operating as government agencies or autonomous statutory bodies, or they can be not-for-profit organizations in the private sector. When the organization is private, a general assembly or consultative council often provides for representa- tion of both the private and public sectors. A 2003 survey of the 147 members of the International Organization for Standardization showed that half of these standards bodies were public organizations, while the other half were private organizations (ISO 2003a). Another distinction among standards bodies is membership policy. In Europe, membership in a national standards body requires that the mem- ber be a European organization or enterprise or have business interests in Europe. In contrast, membership in standards bodies in the United States is unrestricted (Delaney and van de Zande 2000). Standards bodies must adhere to some globally accepted principles to respond effectively to both social and economic demands. These prin- ciples are contained in publications of the International Organization for Standardization (ISO), the International Electrotechnical Commis- sion (IEC), and the World Trade Organization (WTO), especially the Code of Good Practice for Standardization (ISO/IEC Guide 59) and the WTO's Code of Good Practice for the Preparation, Adoption and Appli- cation of Standards (see chapter 5). Standardization procedures reflect- ing these international principles are usually described in the laws and charters that govern standards bodies. In decentralized systems, following these procedures is typically a requirement for becoming accredited as a national standards developer. They can be summarized as follows: · Written procedures based on the consensus principle should govern the methods used to develop standards.3 These procedures should be available to interested parties. · The standards development process should be open. Standards bodies should be proactive in opening participation to all interested parties and there should be broad-based public reviews of draft standards. 60 Quality Systems and Standards for a Competitive Edge Foreign parties should be given equal access to the public review pro- cess. · The standardization process should be based on a balanced represen- tation of interests, with no single interest dominating. · There must be an appeals mechanism for participants who believe that the procedures guiding the standards development process have not been respected. · Standards should be reviewed periodically and revised in a timely manner. Most governments provide some level of support to the standards body. In fact, 63 percent of ISO members receive at least 50 percent of their revenues from government grants, and 83 percent receive at least 20 percent (ISO 2003a). However, it is important to interpret these sta- tistics carefully because ISO membership is only open to one standards body per country. In decentralized models, the primary standards body may be delegating most of the work to other organizations that may receive different levels of subsidies. It is common for standards bodies that receive little government sup- port to offer a number of income-generating services. Membership fees and publications are often used to complement the standards body's budget, but the level of income from such sources rarely exceeds half of the standards body's total income. This shows that it may not be possible to achieve self-sustainability through standardization activities alone. In fact, a memorandum of understanding between BSI British Standards and the UK government acknowledges the role of other business activi- ties in cross-subsidizing standards development. Standards bodies often offer certification, testing, training, and quality consulting services. It is difficult to estimate the total national budget for standardization activities. Estimating the resources that a country dedicates to standard- ization is hard because the costs of standardization are usually distrib- uted among several entities and most of these entities are also involved in activities other than standardization. An ISO survey in 2002 showed that 58 percent of national standards bodies had annual budgets over $660,000 and 14 percent had budgets over $13.2 million (ISO 2003a). National standards bodies are sometimes involved in the develop- ment of mandatory standards, but these rarely make up the majority of a country's standards. In some cases, the voluntary standards developed by private standards bodies are referenced in law or technical regulations The Jurisdiction Issue: Roles of the Private and Public Sectors 61 and effectively become mandatory. In other instances, the national stan- dards bodies are directly responsible for developing both voluntary and mandatory standards. Sometimes the primary standards body restricts its activities to mandatory standards, while decentralized private sector bodies are responsible for voluntary standards. When both mandatory and voluntary standards are decentralized, a national standards body may be responsible for accrediting both private and government standards developers. The ISO survey shows that for 56 percent of ISO members, less than 10 percent of standards are mandatory (ISO 2003a). C H A P T E R 4 Evaluating Compliance with Standards: The Conformity Assessment Framework The mere existence of standards does not ensure their proper diffusion. Standards contain technical specifications that can enhance quality, safety, interchangeability, compatibility, and information diffusion, but these benefits can only be realized if producers understand and comply with the standards. For many standards, especially quality and safety standards, the incentives for self-enforcement are low and producers have much to gain by claiming that a noncompliant product or process adheres to a standard. Furthermore, the highly technical content of some standards may make it difficult for producers to know whether they have appro- priately complied with a standard. If there is no means of differentiating products that conform to a standard from those that do not, standards are of limited use. A country requires institutions that evaluate and verify compliance with standards to reap the economic benefits of standards. The confor- mity assessment framework consists of all the activities that are necessary to demonstrate that a product or process conforms to technical specifi- cations such as those included in technical standards. The full economic benefits of standards can only be realized if there exist organizations for certification, testing, calibration, inspection, accreditation, and metrol- ogy in the national economy. This multidimensional system includes, at a lower level, certification, testing, calibration, and inspection bodies 63 64 Quality Systems and Standards for a Competitive Edge evaluating the conformance of products, processes, services, and orga- nizations to technical specifications, and at a higher level, accreditation and metrology bodies ensuring that the lower-level bodies themselves conform to relevant standards (figure 4.1). Conformity assessment can be performed through three alternate channels. Assessment activities can be performed by the suppliers them- selves (first-party assessment), by the purchasers (second-party assess- ment), or by independent organizations (third-party assessment). In first-party assessment, a supplier performs the conformity assessment procedures internally, using its own staff and equipment. In some cases the supplier performs first-party assessment for internal quality control reasons, but in other cases it does so to declare to potential buyers that a product or process satisfies one or more requirements. Confidence in the Figure 4.1 Traceability of Measurements BIPM: definition of units Primary standards in foreign NMIs Secondary standards Primary standards in NMI in NMI National metrology infrastructure Calibration laboratories Legal metrology authorities Manufacturers, repairers, installers, Designers, manufacturers, researchers, universities repairers and installers of regulated measuring instruments Retailers, traders, students Users of regulated measuring instruments: traders, utility companies, gas stations, supermarkets, and so forth National and international consumers, public health and safety, natural environment Source: Authors'elaboration. Note: BIPM = International Bureau of Weights and Measures; NMI = National Metrology Institute. Evaluating Compliance with Standards: The Conformity Assessment Framework 65 supplier's declaration relies on the supplier's capabilities, integrity, and reputation. When a buyer cannot have confidence in a supplier's declara- tion of conformance, the buyer must conduct a second-party assessment on each individual supplier. This method can be expensive if a buyer has many potential suppliers, and the purchaser may not even have the capabilities to assess the conformance of each supplier. Instead, an inde- pendent organization can validate a producer's claim of conformance through third-party assessment. Certification Certification consists of the provision of assurance that a product, ser- vice, system, process, or material conforms to one or more standards or specifications. Certification is usually conducted by a third party that is independent of the supplier or purchaser. Certification schemes increasingly involve systems rather than specific products. Quality management system certification involves evaluating conformance to a quality standard through auditing of a firm's manage- ment practices, in addition to collecting data and generating test results, if relevant. Similarly, environmental management system certification involves an examination of a firm's environmental management proce- dures. Many of the activities involved in the certification process can be either conducted by the certifier or delegated to other parties, such as independent testing and calibration laboratories or external auditors. Cer- tification often results in the granting of a mark, certificate, label, or regis- tration, such as ISO 9000 or ISO 14000, with a quality system registrar.1 The certification market usually involves a number of competing pri- vate bodies. In the vast majority of countries, subsidiaries of the 20 or so largest global certification bodies have an important share of the market.2 While most certification bodies operate as private for-profit companies, others operate as individual not-for-profit organizations or as part of industry associations. There are internationally accepted guidelines outlining the proper operation of a certification body.3 To produce credible assessments of a product or process, it is generally accepted that a certification body should adhere to the following principles: 1. The certification body should be impartial and independent. This con- cerns the overall strategy and policy of the body, the evaluation pro- cess, and the decision to certify. The certification body should ensure 66 Quality Systems and Standards for a Competitive Edge that its ownership or shareholding and its involvement in other activi- ties do not affect its impartiality and independence. For example, a certification body should not offer consulting services that are aimed at improving a client's chances of obtaining certification, and auditing personnel should not be simultaneously employed in such activities. 2. The certification body should be objective. For this purpose, it should have the proper resources to undertake the relevant tests and inspec- tions, or it should contract these out appropriately. The credibility of the certification will depend on the quality of the auditor. Auditors should have proper technical expertise and competence, including an adequate educational level, specialized training, and significant experience in their area of auditing. To demonstrate their competen- cies, auditing personnel often seek auditor certification by special- ized national bodies or through international associations such as the International Register of Certificated Auditors (IRCA). 3. Finally, all information collected through the evaluation process should remain confidential. Certification bodies often obtain access to de- tailed information about the operations and technology of a candi- date organization. Employees and subcontractors of the certification body should not share this information with third parties, employees, or employers. Accreditation by a national accreditation body, as discussed later in this chapter, will ensure that the conformity assessment body operates according to the above guidelines and is competent and credible, pro- vided that the accreditation body itself is recognized. Membership in international organizations such as IQNet, an international certification network with 38 members, can also enhance the credibility and reputa- tion of a certification body. Certification provides benefits for producers of goods and services, consumers, and government regulators, as well as for international trade. Manufacturers and service providers can have their products or their management systems certified to particular standards to distinguish themselves from less reputable suppliers. Buyers benefit from certifica- tion because it allows them to compare and differentiate products and services in terms of quality, safety, or other desirable characteristics (box 4.1). It avoids the costs of having each buyer independently validate the characteristics of a supplier's products or services--which buyers may not have the resources or expertise to do--and reduces the risk of pur- chasing faulty products or services. Certification is also more transparent Evaluating Compliance with Standards: The Conformity Assessment Framework 67 Box 4.1 The Consorcio del Jamón Serrano Español's Quality Control In 1990, the producers and exporters of air-dried cured ham in Spain formed the Consorcio del Jamón Serrano Español to harmonize standards and create a quality brand. The Consorcio's seal, which is given only to hams that meet its standards, guarantees the high quality of the certi ed product. Under EU regulation, the "Serrano ham"denomination is protected as a Tra- ditional Specialty Guaranteed (TSG). The TSG standard for Serrano ham speci es the method of processing the meat, although it does not refer to a speci c processing area or to the origin of the raw material. Cured ham cannot be sold in the EU with the words"Serrano ham"on the label unless it is duly certi ed as meeting the TSG standard for the product. In addition to meeting the TSG requirements, the Consorcio also imposes its own standards, which in certain aspects are more demanding than the TSG standards. For a ham to earn the Consorcio seal of quality, it must 1. be"Serrano"ham (meetingTSG requirements), produced by a certi ed company; 2. use only Spanish raw material (Spanish pigs slaughtered in Spain); 3. be processed exclusively in Spain; 4. be cured a minimum of nine months; 5. have a fat covering of at least 1 centimeter (to ensure the ham's texture and aroma); 6. have shrunk 34 percent in relation to the weight of the original fresh ham; 7. pass an individual sensorial inspection (piece by piece); and 8. be produced by a company that has passed the quality inspections that the Consorcio constantly carries out. The inspections performed by the Consorcio are certi ed according to the UNE-EN ISO (Spanish national) standards. The Consorcio strives to ensure that hygienic, temperature, and humidity conditions established in the TSG standard, as well as the boning, slicing, and packaging procedures, are respected during the di erent stages of the process. In addition, each piece of ham is subjected to visual inspection. A ham that meets all the standards will have a re seal on the skin with the Consorcio del Jamón Serrano Español logo and will also have a num- bered control label. Consumers who purchase the certi ed products pay a pre- mium price in exchange for the quality assurance that the certi cation provides. Source: Adapted by authors from Consorciodel Jamón Serrano Español. 68 Quality Systems and Standards for a Competitive Edge than relying on a producer's reputation,4 and it conveys standardized information about the capabilities of a producer that may not be obtain- able elsewhere. It reduces search costs for global suppliers and hence reduces technical barriers to trade. In fact, verification by third-party organizations is increasingly included in trade contracts (Schuurman 1997). Finally, regulators benefit from certification because it provides them with a tool to enforce governmental health, safety, and environ- mental legislation. There are costs associated with certification. The range and magnitude of these costs vary, depending on the standard for which a firm is seeking certification. An organization is faced with both internal and external certification costs. Internal costs include salaries of employees dedicated to meeting the certification requirements, along with documentation costs, overhead costs, internal training expenses, and investment in pro- duction and quality control technologies. External costs involve the pro- curement of services outside the firm. The firm can prepare itself for the certification process by hiring internal auditors or seeking assistance from external consultants to provide diagnostic and training services. Where relevant, laboratories may be involved in technical support for calibra- tion and quality control equipment. External costs are also associated with the auditing process itself and include administrative fees paid to the certification body. These vary according to the complexity of the cer- tification task. Because certification involves continuous surveillance and periodic reassessment of the firm, some of the costs must be added to the firm's operational costs. Testing, Calibration, and Inspection Testing, calibration, and inspection involve performing consistent and reliable measurements. Testing seeks to determine the characteristics or performance of a product or process according to a specified procedure. Calibration determines the relationship between an instrument's input and the magnitude or response of its output. Inspection is another form of assessment that relies on less sophisticated instruments than testing. The inspection of products is usually conducted visually or by using sim- ple instruments. Testing and calibration laboratories and inspection bodies can differ widely from one another in size, range of services, legal status, purpose, and technical competence. They are plentiful in most economies: they can operate as government regulatory laboratories or bodies, government Evaluating Compliance with Standards: The Conformity Assessment Framework 69 research laboratories, university laboratories, independent private sector laboratories or bodies, manufacturers' in-house laboratories, or custom- ers' in-house laboratories. Testing, calibration, and inspection are integral components of the conformity assessment process. Independent testing laboratories and inspection bodies can be contracted by a firm to obtain a test or inspec- tion certificate as evidence that a product or process conforms to cer- tain characteristics. Independent calibration laboratories can be used to guarantee the performance of a company's measurement equipment. A third-party laboratory calibration certificate can provide evidence that a firm has the quality control equipment that meets the requirement for product or system certification. In certain cases, testing, calibration, and inspection are required for firms to implement a quality control system, such as ISO 9000. These may be essential elements in proving that a firm is complying with customer requirements. There are internationally accepted standards guiding the operation of testing and calibration laboratories and inspection bodies.5 To play a credible role in the conformity assessment system, testing and calibra- tion laboratories and inspection bodies must display many of the same characteristics as certification bodies, notably impartiality, objectivity, and confidentiality. Objectivity relies heavily on the procedures guiding the evaluation process, the equipment used, and the skills and qualifica- tions of staff. Equipment and measurement reference systems must be calibrated to other widely accepted metrological references. This is the concept of traceability of measurement discussed in more detail in the section on metrology in this chapter. Accreditation The function of accreditation Accreditation is defined as the procedure by which an authoritative body gives formal recognition that an organization or person is competent to carry out specific tasks. Accreditation is sought on a voluntary basis as proof of competence in a given area. Accreditation provides certification and inspection bodies, as well as testing and calibration laboratories, with a means to signal that they are conducting their work to appropriate standards and that they are able to provide reliable services to support quality in firms. The accreditation body evaluates the personnel and sup- porting management system of the candidates for accreditation and can request practical tests for laboratories when relevant. These tests take 70 Quality Systems and Standards for a Competitive Edge the form of proficiency testing schemes through which the measure- ment results of different laboratories are compared. Accreditation is usu- ally valid for a few years after initial assessment, during which time the accredited organization is subject to regular surveillance. There are well-recognized international criteria that guide the accred- itation process. Most accreditation bodies base their accreditation deci- sions on the performance of interested entities against widely recognized international guides and standards that establish operational require- ments, such as ISO/IEC Guides 62, 65, and 66 for certification bodies, ISO standard 17025 for laboratories, and ISO/IEC standard 17020 for inspection bodies. These requirements include the impartiality, objectiv- ity, and confidentiality practices described in the sections above on certi- fication and testing, in addition to sound management practices. Accreditation contributes to economic development by reducing risk, ensuring quality, and increasing productivity. Accreditation protects customers from the risk of receiving poor service by guaranteeing that accredited organizations are both independent and technically competent. Accreditation enhances quality in the economy by strengthening the cred- ibility of the entire conformity assessment system and decreasing informa- tion asymmetries. Accreditation also enhances productivity by providing independent assessments of commercial and in-house laboratories. The laboratory assessment process always contains an implicit element of tech- nology transfer, with comments and suggestions that can lead to improve- ments in a laboratory's quality and performance (UNIDO 2003). The scope of accreditation can be broad or specific, depending on the accreditation body and on the product and standard for which the certification body is seeking accreditation. In some cases a certification body will receive multiple accreditations to allow it to cover many areas. Likewise, laboratories can be accredited for a broad or narrow range of tests and calibrations. The International Laboratory Accreditation Coop- eration (ILAC) recommends specific practices for defining the scope of laboratory accreditation through standardized descriptors (ILAC 1994), but many accreditation bodies use their own methodology to establish the scope of their testing and calibration accreditations. As a result, there is often no one-to-one correspondence between the scopes of laboratory accreditations offered by two different accreditation bodies. Accreditation institutions Establishing an accreditation body requires significant investments and government support. Accreditation requires well-trained personnel and Evaluating Compliance with Standards: The Conformity Assessment Framework 71 may have important technological requirements, even for simple sys- tems. The accreditation process is costly and time-consuming, because it must be accompanied by surveillance activities. It also involves a lengthy learning process that can only be realized by having a large customer base. With the exception of countries where there are large markets, such as the United States, accreditation is not usually financially self-suf- ficient and depends on government subsidies (UNIDO 2003). Investing in accreditation needs to be justified by an adequate demand for accredi- tation from certification bodies and testing and calibration laboratories, or at least a demand for conformity assessment services from domestic firms and organizations. In most countries, accreditation bodies indirectly benefit from government support through policies that provide incen- tives for accreditation or that require product certification by accredited entities in public procurement provision. Some countries do not have a national accreditation body and have decided to provide accreditation services by engaging in agreements with foreign accreditation bodies or through regional cooperation. This is pos- sible because national accreditation bodies often allow organizations from other countries to seek accreditation. However, to be effective, a foreign accreditation must be recognized in the accredited organization's home country. One option is to create a nominal national accreditation body that contracts with a foreign body to undertake accreditation activ- ity on its behalf, with the national body retaining some authority over the process. Alternately, regional accreditation bodies can be established to spread costs among several countries. The Joint Accreditation System of Australia and New Zealand is an example of such a model. The South- ern African Development Community (SADC) is also building a novel regional model for accreditation (box 4.2). It is important to note that although regional bodies may provide superior solutions in some cases, they can be very difficult to implement for political reasons. Accreditation bodies have different structures and legal status in dif- ferent countries, with no single model demonstrably superior to any other. Accreditation bodies may be subject to various degrees of govern- ment involvement depending on the country. Nonetheless, countries are increasingly finding the government-managed model less desirable than a public autonomous agency or nonprofit organization. This is due to a recognized need for greater flexibility and independence. The struc- ture of the national accreditation system may also vary but the norm is to have a single national accreditation body. In a few countries, such as the Republic of Korea, laboratory accreditations and certification body 72 Quality Systems and Standards for a Competitive Edge Box 4.2 Regional Cooperation on Accreditation: The Southern African Example The Southern African Development Community (SADC) was formed in 1992 and now consists of 14 states. One of the community's objectives has been to lower technical barriers to trade. Toward this end, in 1997 SADC in Accredita- tion (SADCA) was tasked with de ning a suitable regional accreditation infra- structure or forming a regional accreditation body, or both. One of its principal objectives has been to create a pool of internationally acceptable accredited laboratories and certi cation bodies in a region where most countries do not have the resources to establish a national accreditation body. Currently only two member states, South Africa and Mauritius, have established national accreditation bodies. In its rst phase, SADCA's objective was to become a regional accreditation management committee to allow member states to cooperate at the interna- tional level with the International Accreditation Forum (IAF) and ILAC, along with other regional organizations. In a second phase, which started in 2002, SADCA initiated a novel approach to regional cooperation. This consisted of establishing a regional accreditation service, SADCAS, which will work alongside the current and future national accreditation bodies to provide a ordable services to member states. SADCAS will provide a cost-e ective and transparent mechanism for member states that do not want to establish their own national infrastructure, for lack of resources, but that want to provide some input in the decision-making process and tech- nical operations of accreditation through "national accreditation focal points." Any state that nds it cost-e ective to do so is allowed to create a national infrastructure for accreditation that will coexist with SADCAS. A project man- agement committee has thus far been able to secure some donor funding for initial training for SADCA members and national accreditation focal points. The project will train technical assessors in each country to become part of a pool of regional assessors for SADCAS. By pooling scarce resources to create a regional accreditation infrastructure, SADCAS has become a unique model for develop- ing countries. Sources: UNIDO 2003; SADCA Web site; SADC Web site on Standardisation, Qual- ity Assurance, Accreditation, and Metrology. Evaluating Compliance with Standards: The Conformity Assessment Framework 73 accreditations are the responsibilities of two distinct bodies. An extreme case is the United States, where a number of federal, state, and local government accreditation bodies operate in parallel with private sector accreditation bodies. What matters more than legal status and structure is the ability of an accreditation body to comply with a few key prin- ciples. Internationally accepted standards, guides, and codes of conduct6 emphasize the need for impartiality, objectivity, and nondiscriminatory policies and practices, and call for the avoidance of conflicts of interest. Involving stakeholders in the governance and technical functions of the accreditation body can help ensure impartiality and build confidence in the organization. A governing board is often established to represent the interests of both the public sector and potential clients, although this is less common when there is tight governmental control over the accredi- tation body. The establishment of technical consultative committees of external specialists also ensures that accreditations are granted on the basis of sound and objective technical decisions. The technical commit- tees should include a balanced representation of the concerned sectors. Conflicts of interest can arise when accreditation bodies are involved in other activities. Offering certification, testing, and calibration services leads to conflicts of interest because it puts the accreditation body in competition with potential candidates for accreditation. Not all related commercial activities present these problems, and an experienced accred- itation body may be in a unique position to offer some services to indus- try. This notably includes training services, not only for future assessors but also for technical laboratory staff who conduct internal audits and quality management staff in firms. The objectivity of the accreditation process largely depends on the quality of its technical staff.Assessor training programs are highly special- ized and are usually only found in mature accreditation bodies. The staff of new and developing-country accreditation bodies often receive train- ing through formal courses and field practice in more experienced bod- ies abroad. Assessment teams must have sufficient collective scientific and technical skills to understand the testing and calibration activities of laboratories and the management systems of certification bodies. In the area of laboratory accreditation, technical expertise requires assessors to have current knowledge of the latest technologies and scientific practices. In certain fields it may be difficult for full-time assessors to maintain an adequate level of current knowledge to perform their tasks reliably. For this reason, assessors are often selected from accredited organizations or from academic or technical institutions and are engaged on short-term 74 Quality Systems and Standards for a Competitive Edge contracts or as volunteers. For assistance with accreditation, lead asses- sors can also rely on external technical experts, although these experts may have no formal training in conformity assessment. Metrology Metrology is the study of a system of measures. Measurements are not only an integral part of the national quality system, but an integral part of our daily lives. Because the engineering behind most of the technology we use relies on measurements, accurate measurements allow equipment to be calibrated to produce consistent product quality, and they allow tests to be performed to verify that a product or process conforms to predetermined requirements. Correct measurements also play a key role in the economy by facilitating commercial transactions and their regulation. Metrology is usually supported by one or more institutions in a given country.They provide a support infrastructure for all measurement-related activities performed by enterprises, individuals, and governments. Scien- tific metrology is the branch of metrology concerned with developing measurement standards and promoting their acceptance and equivalence. Legal metrology consists of the legislative, administrative, and technical procedures established to regulate the credibility of measurements related to official controls, trade, health, safety, and the environment. Scienti c metrology The role of a national metrology institute (NMI) is to establish the national measurement system; to maintain, develop, and diffuse measurement standards for basic units; and to disseminate metrological expertise to the economy (figure 4.1).The NMI establishes the national measurement sys- tem by adopting a national system of units with a view to incorporating it in legislation. In establishing the measurement standards for these units, an NMI can make use of either primary or secondary standards. Using a primary standard is the most fundamental approach because it provides an absolute basis of reference for a unit and its value is accepted without reference to another standard of the same quality. An NMI can create the physical realization of the unit from its definition to estab- lish a primary standard. For example, a standard for temperature can be obtained by realizing the equilibrium state of a pure substance that occurs at a highly reproducible temperature. Realizing the triple point of water, which corresponds to the thermal equilibrium between solid, liquid, and vapor, will provide a temperature of exactly 273.16 kelvin. Evaluating Compliance with Standards: The Conformity Assessment Framework 75 Obtaining the physical realization of a standard can be a difficult and costly approach, but it is possible in the case of some physical quantities to establish a primary standard by creating a reproduction of the standard. For example, the meter is defined as the length of the path traveled by light in a vacuum during a time interval of 1/299,792,458 of a second. The practical reproduction of this standard is not a physical realization because it does not involve timing the passage of a light beam; rather, it consists of comparing length to be measured to the wave length of a frequency-stabilized laser. An NMI can also establish a secondary standard, its value being deter- mined through periodic comparisons with a primary standard in another country. For instance, the kilogram is defined as equal to the mass of the International Prototype of the Kilogram, a cylinder made of platinum and iridium fabricated in 1889 and maintained in Paris by the Interna- tional Bureau of Weights and Measures (BIPM). Because the definition and the construction of this unit are based on an artifact, the original kilogram is considered to be the only primary standard. To establish their secondary standards, NMIs must compare their own mass standards to the original kilogram or to other secondary standards that are traceable to the original kilogram. The NMI is the primary scientific metrology laboratory in a coun- try and its exact measurement standards are disseminated to industry and other users. In countries where the commercial metrological net- work is well developed, the NMI conducts very few calibrations directly for industrial clients. Instead, measurement standards are disseminated through a network of calibration laboratories and through organizations responsible for regulations and standards. When their industrial mea- surements are traceable to the NMI through an unbroken chain of com- parisons, firms are able to guarantee the accuracy and precision of their calibration instruments, process control instruments, and quality control instruments. International standards used to assess the competence of accredited laboratories and quality systems require the internationally recognized traceability of measurements and test results.7 Often, experts from the NMI participate in the technical evaluation of laboratories seek- ing accreditation. The NMI disseminates metrological expertise by offer- ing specialized technical consultancy and training services to industry, research institutions, and educational institutions. NMIs also conduct research and development activities in areas related to metrology to develop new systems and deepen their own tech- nological capabilities. Measurement standards evolve continually and 76 Quality Systems and Standards for a Competitive Edge reflect advances in science as well as changing industrial and other needs. The primary purpose of research in measurement science is to develop more advanced and accurate calibration methods to ensure that the long- term needs of society, industry, and government will be fulfilled. Research and development (R&D) also serves the important purpose of enhanc- ing an NMI's ability to identify, assimilate, and exploit existing knowledge in the field of metrology, or what Cohen and Levinthal (1989) refer to as "absorptive capacity" in their seminal work on the dual role of R&D. This absorptive capacity depends on tacit knowledge, which cannot eas- ily be transferred through traditional means such as manuals and courses. Absorptive capacity gives an NMI the opportunity to acquire outside information and technology that its scientists and engineers will find most relevant to national needs. R&D is also necessary to attract and retain the high-skilled personnel required for many of the NMI's activities. A survey of 26 NMIs indicated that on average, 32.5 percent of their annual budgets were spent on R&D and 34.6 percent of their employees could be clas- sified as research scientists (KPMG Consulting 2001). NMIs in develop- ing countries are generally less involved in R&D than those in developed countries, and when they do conduct R&D, they tend to focus more on the development of new standards than on basic measurement research. It is necessary to continuously invest in new metrology infrastructure to keep up with the pace of technological and market changes. These changes are driven by growing demands for quality and performance in global markets. A former director of the International Bureau of Weights and Measures, the primary international organization concerned with metrology, notes Engineering tolerances, i.e., the amount by which dimensions are permitted to depart from specification, have tightened in practi- cally all industrial production by a factor of three every 10 years since 1960. The result is that production engineers in the large- scale manufacture of automotive and electronic products are now required to work at tolerances previously attempted only in fine, small scale work. (Quinn and Kovalesky 2005) The recent growth of certain industrial sectors has made it necessary to redefine the scope of activities of the NMI. In the past decade, NMIs have invested in expanding metrology from the traditional areas of phys- ics and engineering to chemistry and biology, as the latter two fields have become more prominent factors in high-growth industries (box 4.3). Evaluating Compliance with Standards: The Conformity Assessment Framework 77 Box 4.3 Evolving Needs of Metrology in Health Care Metrology has a direct in uence on people's lives when it involves medical diag- nosis or therapy.The accuracy requirements for measurements in medicine are not as high as in other elds, but the reliability of these measurements must exceed all standards because they can have life-or-death consequences for human health. In economic terms, the impact of medical diagnosis measurements is very large and growing. In 2003, spending on medical care reached 8.62 percent of GDP on aver- age in Organization for Economic Co-operation and Development countries, up from 7.8 percent in 1993. In the United States, 15 percent of the GDP was spent on health in 2003. As much as 30 percent of health costs have been found to concern measurement and tests related to diagnosis. Progressinmeasurementcapabilitieshasimprovedthediagnosesofpatients in the past decade, increasing their chances of successful treatment, but there are still many problems in health care metrology. The EU has recently addressed this problem by issuing the In Vitro Diagnostics Directive, which came into force in 2003 and requires that all diagnostic kits be calibrated to higher-level standards. A key challenge for metrology authorities in the EU and in countries exporting products to the EU has been to make available reference materials, seroconversion panels, and critical samples for manufacturers. In medical therapy, there are still accuracy issues in ionizing radiation, and scienti c progress has been slow in this eld in the past 20 years.Without proper calibration, underdosing and overdosing would be common, because the radi- ations themselves are invisible and have no immediately observable e ects on the patient. Errors in radiotherapy greater than a few percentage points can easily lead to detectable physiological e ects. Ultimately, the accuracy of ioniz- ing radiation depends on the development and maintenance of high-accuracy standards by the NMIs. This requires considerable resources and e ort. Sources: OECD 2005; Quinn and Kovalesky 2005; BIPM 2003. Most NMIs in industrial and rapidly developing countries receive sig- nificant funding from their central governments. This is even the case when the NMI is managed by a private organization. Public funding is usually justified by the wide range of individuals and organizations that benefit from the existence of measurement standards. The private sec- tor alone would tend to provide fewer resources to the production and maintenance of measurement standards than is desirable from the point 78 Quality Systems and Standards for a Competitive Edge of view of the economy as a whole because of the "public good" nature of these standards. In industrial countries, governments spend between 0.002 percent and 0.007 percent of GDP on metrology infrastructure. This figure is roughly equivalent to $300 million for the United States. Governments of some rapid-growth East Asian economies have invested as much as 0.01 percent of their GDP in establishing a measurement infrastructure (BIPM 2003; Quinn and Kovalesky 2005). NMIs also fund part of their activities through the fees they charge for their calibration, consulting, and R&D services. In most developed countries these fees are determined by the NMIs themselves, while in other countries they are sometimes specified in national legislation. Laws require the establishment of a national metrology institute in practically every country (BIPM 2003), but these institutions can be organized in different ways. NMIs are usually public institutions but are often granted some operational autonomy. It is often the case that all metrological activities are incorporated within a single organization because this facilitates efficiency and transparency. However, some countries do have several national metrology institu- tions. This occurs when a central NMI is not able by itself to establish metrological activities in all the fields of interest. In that case, a central NMI or the government designates other public or private institutes to maintain national measurement standards and related calibration facilities for one or more quantities and measurement ranges. In Spain, a total of six institutions maintain national standards. These include laboratories affili- ated with the Ministry of Defense and the Ministry of Science and Educa- tion, but also an independent foundation affiliated with a university. In the United Kingdom, one of the four custodians of the national standards is a private for-profit engineering consultancy company called the National Engineering Laboratory. In Chile, the national metrology institute operates no laboratories of its own but coordinates a network of five custodians. A decentralized national metrology system should guarantee a level playing field, especially when commercial companies are involved. A drawback of operating multiple NMIs is that they are unlikely to enjoy the same cred- ibility and recognition as a single NMI responsible for a broad range of scientific research. A number of countries require their multiple NMIs to demonstrate their competence by having their laboratories accredited. Legal metrology Legal metrology involves the legislative, administrative, and technical procedures that cover regulated areas where there is a public interest Evaluating Compliance with Standards: The Conformity Assessment Framework 79 in the correctness of units, measurements, and measuring devices. These areas cover all official measurements, measurements involved in enforc- ing mandatory standards, and commercial transactions that involve the use of measuring instruments for the determination of prices, such as weighing devices, taxi meters, and electric meters. The role of a national legal metrology organization is to guarantee accuracy by ensuring that measuring instruments comply with legal regulations. This is achieved by approving the types and models of measuring instruments used for certain activities, and by coordinating or conducting verifications and inspections of measurements and measuring instruments. Although most governments find it generally undesirable to make standards mandatory, measurement standards present significant exter- nalities and are usually integrated into legislation. Some of the benefits of legal metrology include 1. Reduction of disputation and transaction costs. Enforcing the use of official measurement units provides transparency and decreases information asymmetries in trade transactions involving measure- ment. Market efficiency is increased, consumers are protected, and fraud involving faulty measurements or measuring instruments is reduced. 2. Helping the government fulfill its fiscal functions. Governments col- lect significant amounts of revenue through excise and taxes based on measurement. Legal metrology reduces fiscal arbitrariness and increases the predictability of business costs. 3. Contributing to public safety, health, and the environment. Regulations aimed at the public interest often involve specific measurements, such as determining permissible amounts of contaminants in water and monitoring automobile emissions.These regulations cannot be enforced without a properly functioning legal metrology infrastructure. Empirical evidence confirms that reducing measurement errors in legal metrology decreases trade costs. Canada's legal metrology agency, Measurement Canada, conducted a study to quantify the benefits of met- rological control on trade. The study combined information on the total value of goods traded over trade weighing and measuring instruments and estimates of the compliance rates of the different instruments to find the annual trade costs of measurement errors. Measurement Canada found that each dollar spent on periodic inspection of measuring instru- ments saved $11.40 in measurement errors. When targeting measuring 80 Quality Systems and Standards for a Competitive Edge instruments for which errors would be the most costly, the benefit-cost ratio increased to 28.7 (Birch 2003). Because legal metrology affects the public interest so extensively, legal metrology institutions are not often privatized or given substantial auton- omy. Regulations are established by public authorities and national legal metrology institutions have a role in their formulation and enforcement. Furthermore, legal metrology organizations cannot be self-supporting because they conduct supervision and consultancy tasks for which they cannot charge (Marban and Pellecer 2003). There are different approaches to organizing legal metrology. In some countries a single organization covers both legal and scientific metrology. This is usually not possible unless the national metrology institute is inte- grated in the public sector. In other countries there are separate institu- tions for legal and scientific metrology. In this case, there should be close cooperation between the two fields to ensure that the country has the technical capacity to enforce regulations. In most countries legal metrol- ogy is decentralized and a large number of responsibilities, including verification and inspection, are delegated to states or municipalities. In principle, all measurements performed outside the national legal metrol- ogy body should be traceable to the national standards (figure 4.1). Fees are usually charged for model approval, verification, test center accredi- tation, and maintenance service accreditation, and fines are collected for infractions of metrology law, such as use of nonauthorized measure units and use of nonverified instruments. C H A P T E R 5 Opening Doors: Standards, Conformity Assessment, and International Integration Differences in standards and technical regulations between countries, even when justified, may sometimes create technical barriers to trade. Differences in standards can reflect differences in national consumer preferences, income levels and distributions, the natural environment, technological capacity, or historical technological trajectories. Similarly, differences in technical regulations can reflect national preferences for promoting health, safety, or the environment. Hence, there is sometimes sound economic justification for not aligning national standards and tech- nical regulations with those of actual or potential trade partners. However, as discussed in chapter 2, there are cases in which standards and technical regulations are purposely used as nontariff trade barriers to protect markets from foreign competition. This occurs when differ- ences between countries are not justified by legitimate health or safety objectives or technological capabilities, or when standards and technical regulations are not properly publicized. In those cases, a lack of harmoni- zation can create a net welfare loss in both an importing and an exporting country. Even when national standards or technical regulations have been harmonized, incompatible conformity assessment procedures can deter trade. Complying with a standard or a technical regulation is only useful if compliance can be demonstrated to the buyer or the government at 81 82 Quality Systems and Standards for a Competitive Edge reasonable cost. Demonstrating compliance through conformity assess- ment is itself only useful if the testing and certification requirements are similar in the exporting country and the importing country. If testing laboratories are not recognized abroad, tests on products carried out in the exporting country have to be repeated by a recognized laboratory in each of the importing countries. An adverse test report in the import- ing country can result in the rejection of an entire shipment. Likewise, if certification in one country is not recognized abroad, domestic firms requiring quality system and environmental management certification for export purposes1 need to be certified by organizations in each of the importing countries. Conformity assessment procedures vary widely across countries and in many cases constitute a larger technical barrier to trade than standards. Nonrecognition or nonharmonization of con- formity assessment procedures do not persist due to inherent national differences, but because conformity assessment is particularly vulnerable to misuse if bureaucratic procedures are not transparent. In developing countries, duplication of testing and certification is fre- quent and very costly. A recent survey of firms in developing countries showed that 44 percent of firms had to conduct significant duplication of testing procedures to meet foreign requirements after domestic require- ments had been met, and 30 percent had to conduct complete duplica- tion of testing procedures, as shown in table 5.1 (Wilson and Otsuki 2004). In the same survey, 68 percent of firms cited testing and certifica- tion costs as an important reason for not exporting. Because many small and medium enterprises can barely afford a single certification, redun- dant conformity assessment procedures make exporting prohibitively expensive. In sum, a lack of confidence in another country's accreditation system, or the absence of infrastructure for accreditation, are significant obstacles to trade. Table 5.1 Duplication of Testing Procedures to Meet Foreign Requirements Duplication effort Share of firms (%) No duplication 7 Minor duplication 13 Signi cant duplication 44 Complete duplication 30 Not answered 6 Source: Wilson and Otsuki 2004. Opening Doors: Standards, Conformity Assessment, and International Integration 83 The WTO Agreement on Technical Barriers to Trade In light of these problems, the World Trade Organization's (WTO's) Agreement on Technical Barriers to Trade (TBT) forbids the use of stan- dards, conformity assessment procedures, and technical regulations as nontariff trade barriers. The TBT Agreement is an integral part of the WTO Agreement, which now extends to 148 countries. The principles of the TBT Agreement include the following: 1. Avoidance of unnecessary obstacles to trade. The agreement allows for legitimate divergences between countries with respect to tastes, income, geography, and other factors. It accords flexibility in the preparation of technical regulations on condition that they be con- sistent with a given policy objective and have legitimate objectives.2 Conformity assessment procedures are not to be stricter or more time-consuming than is necessary to evaluate a product. 2. Harmonization. When appropriate, countries should use interna- tional standards, such as those produced by the International Organi- zation for Standardization (ISO) and the International Electrotech- nical Commission (IEC), as a basis for technical regulations to fulfill given policy objectives. The agreement requires signatories to fol- low international guides or recommendations in national conformity assessment procedures unless they are inappropriate. Exceptions can be made for developing countries in cases where implementing and enforcing international standards is not possible due to specific tech- nological or socioeconomic conditions. The agreement also encour- ages members to participate in the work of international standards bodies to ensure that international standards reflect their production and trade interests. 3. Nondiscrimination. Technical regulations and conformity assessment procedures should treat all products equally, irrespective of national origin. 4. Equivalence. Members should accept technical regulations that are different from their own, provided they fulfill the same objectives. 5. Mutual recognition. Members are encouraged to enter into negotia- tions with other members for the mutual acceptance of conformity assessment results. 6. Transparency. Members should provide notification to the WTO when work or international agreements on technical regulations and conformity assessment procedures could have a significant impact on 84 Quality Systems and Standards for a Competitive Edge trade. Notifications should also be given for measures taken to ensure the implementation of the agreement and acceptance or withdrawal from the Code of Good Practice.3 Countries should establish a national enquiry point where WTO members can obtain information on technical regulations, standards and test procedures, and relevant international agreements. International Coordination in Standardization International and regional standards The three organizations responsible for the vast majority of international standards are the ISO, the IEC, and the International Telecommunication Union (ITU). Their role is to elaborate and diffuse international stan- dards that allow businesses to compete in markets around the world. ISO and the IEC are private, nongovernmental organizations. ISO has 153 members, consisting of the primary national standards bodies from each country. The IEC membership consists of 63 national committees representing the public sector, the private sector, or both, depending on the country. The ITU is an international organization within the United Nations whose membership includes 189 governments and 620 members of the private sector. Development of international standards requires consensus between all major global stakeholders and can sometimes be a very difficult process (box 5.1). Box 5.1 The Struggle to Create a Single Global Standard for 3G Telephony When countries join e orts to develop an international standard it is usually because all major public and private stakeholders agree on the bene ts of using a shared standard. However, even with full global support, coordination problems, market forces, and technological trajectories can hinder the adoption of common standards when stakeholders cannot agree on which standard to choose. This was the fate of the ITU's attempt to establish a single global stan- dard for third-generation (3G) mobile phones. Opening Doors: Standards, Conformity Assessment, and International Integration 85 Third-generation phone systems are based on superior technology that can transmit both voice data and nonvoice data such as e-mail, Internet, and real- time video images. In 1985, as a response to the emergence of incompatible cellular standards in Europe, Japan, and the United States, the ITU started to develop speci cations for a 3G standard. Its objective was to establish a single worldwide standard that would enhance the performance of rst-and second- generation phones while o ering seamless global roaming, enabling a user to move across borders while using the same number and handset. By 1998, two technologies were competing to become the global 3G standard and were pro- posed to the ITU. European and Japanese rms formed a coalition to back Uni- versal Mobile Telephone System (UMTS), while in the United States, Qualcomm supported Code Division Multiplexing Access (CDMA-2000). Today, it is clear that strong strategic and economic interests prevented the adoption of a single 3G standard. In 2000, the vision of the ITU could not be real- ized and it adopted the IMT-2000 standard as a family of standards (comprising UMTS, CDMA-2000, and three other standards) rather than as a single standard. In this context, regions have been adopting incompatible 3G standards and the role of the IMT-2000 project has become one of increasing the compatibility between cellular technologies. A number of Global System for Mobile Communications (GSM) countries, mainly in Europe, have freed up the larger frequency ranges necessary for UMTS networks. UMTS is not compatible with GSM, so UMTS handset manufacturers in Europe have decided to produce dual-mode phones that can operate using both technologies. In North America, the adoption of UMTS has been ham- pered by spectrum limitations. The UMTS spectrum established in the ITU stan- dard has not been made available for cellular telephony in the United States or Canada. As a result, UMTS handsets used in North America operate at di erent frequencies and cannot be used in Europe. CDMA-2000 does not require new spectrum allocation and is compatible with second-generation CDMA tech- nology, so it has been widely adopted outside the GSM zone in the Americas, Japan, and the Republic of Korea. GSM operators in North America have tempo- rarily adopted another IMT-2000 standard, EDGE, which does not make use of additional spectrum and is compatible with existing GSM handsets. Meanwhile, China has been developing its own IMT-2000 3G standard, TD-SCDMA, and is preparing to launch its rst operational network. Sources: Ritchie et al. 1999; ITU 2005, 2006; Henten and Saugstrup 2004. 86 Quality Systems and Standards for a Competitive Edge International standards are playing a growing role in the global econ- omy as they become the basis for an increasingly large share of national standards. Together, ISO and the IEC produce about 85 percent of all international standards (WTO 2005). ISO has produced a total of 14,941 international standards, including 1,247 new standards in 2004 alone. These standards include all fields except for electrotechnologies and tele- communications. The IEC has developed 4,840 international standards, including 343 in 2004. The ITU has more than 3,000 standards in force.4 Although these numbers are smaller than the total number of standards in some industrial economies (figure 5.1), international standards are quickly gaining momentum in terms of adoption rates. In some countries, such as the Netherlands and the United Kingdom, roughly 45 percent of national standards are based on international standards (ISO 2003b). International standards are developed through the voluntary participa- tion of members. ISO, the IEC, and the ITU develop standards using gener- ally similar systems. In the specific case of ISO, standards are first proposed by an industry or business sector that communicates the requirement to their national member. If there is sufficient support for the standard among ISO members, responsibility for the development process is given to tech- nical committees, subcommittees, or working groups consisting of national delegations of experts. These experts incorporate representatives of the public or private sectors. If a new technical committee needs to be cre- Figure 5.1 Number of Standards in Selected Industrial Countries and International Standards 30,000 26,544 27,179 25,000 22,053 22,589 ds 20,000 standar 14,941 15,000 of 10,000 9,009 number 6,664 4,840 5,000 0 Australia France Germany Japan Netherlands United ISO IEC Kingdom Sources: ISO 2003a; IEC and ISO Web sites. Note: Country data are December 2002; ISO and IEC data are December 2005. Opening Doors: Standards, Conformity Assessment, and International Integration 87 ated, its administration is given to the national member organization that proposed the committee. When committees, subcommittees, and working groups reach consensus on a draft standard, it is submitted to a vote of the member states before becoming a standard.The main costs of standardiza- tion are borne by the member bodies, which manage standards develop- ment projects, and business organizations, which provide the expertise for the technical work. ISO, IEC, and ITU members are not obliged to adopt the international standards developed by these organizations. There are important benefits to participating in the international standardization process. Active participation increases the chance that national concerns will be reflected in international standards (box 5.2). It also makes possible the compatibility of national standards with inter- national standards. Participation in international standards development can also provide a forum for exchange of technical information with rep- resentatives from international industrial and scientific organizations.The international standardization process tends to be dominated by Organi- sation for Economic Co-operation and Development (OECD) countries and their needs are usually more closely reflected in international stan- dards. In fact, four countries--the United States, the United Kingdom, France, and Germany--account for roughly 60 percent of all secretariats and governorships of ISO committees and working groups. Along with international standards, regional standards represent a growing share of the standards stock of many countries. The most developed regional system can be found in Europe, where the European Committee for Stan- dardization (CEN), the European Committee for Electrotechnical Standard- ization (CENELEC),and the EuropeanTelecommunications Standardization Institute (ETSI) develop regional voluntary standards. Member states of the European Community and European Free Trade Association (EFTA) are rep- resented in CEN and CENELEC through their primary standards organiza- tions. ETSI is open to all organizations, governmental and private. The European regional institutions use a standardization process simi- lar to that followed by their counterparts in other regions. A number of organizations in other regions have organized regional standards devel- opment schemes, but these are far less developed than those in the EU. In the Americas, the Pan American Standards Commission (COPANT) develops common standards for most countries; in the Asia Pacific region, the Asia-Pacific Economic Cooperation (APEC) carries out standards har- monization activities; and in Africa, harmonized standards are adopted in the Southern African Development Community (SADC) through the SADC Cooperation in Standardization (SADCSTAN). 88 Quality Systems and Standards for a Competitive Edge Box 5.2 South Africa's Contribution to International Timber Standards South Africa is active in approximately 270 ISO technical committees (TCs) and subcommittees and is an observer in 108 more. The main strategy of the South African Bureau of Standards is to in uence international standardization in areas where South Africa participates in global trade. For example, South Africa is very active in ISOTC 165, the technical committee concerned with structural timber.TC 165 develops international standards that establish design requirements for timber structures, structural properties, and performance, and standards for test methods and requirements to establish physical properties and performance of timber. South Africa has few indigenous natural forests and has gained signi cant expertise in obtaining construction timber from plantations. Plantation forestry trees in South Africa grow faster than the same species in the northern hemi- sphere. As a result, one of the main problems with South Africa's trees is that they produce a lower quality or grade of timber, based on the strength of the timber. Much of South Africa's wood is of grade 5, 6, or 7, whereas 7 is usually the low- est grade in the northern hemisphere. Although grades lower than 7 were not originally to be included in the standards, one of South Africa's important con- tributions in TC 165 was the inclusion of these grades, so that the standards take technical problems of the tropical and subtropical regions into consideration. South Africa was also successful in in uencing standards related to methods of testing treated timber. Many timber species are not suitable for speci c uses in many regions of Africa and are susceptible to local environmental threats such as the African termite. However, foreign donors would often specify the use of such unsuitable or untreated imported timber for their projects. Through its involve- ment inTC 165, South Africa was able to have the ISO standards include test meth- ods to ensure that laminated beams are adapted to an African environment. Sources: Jooste, Kruger, and Kotzé 2003; ISO Web site. In many European countries, a growing majority of standards are regional. The European standards bodies produce two kinds of standards: European standards and harmonization documents. Member organiza- tions are obliged to adopt European standards and withdraw conflicting national standards. Harmonization documents can be implemented some- what more flexibly, and members are obliged to incorporate their techni- Opening Doors: Standards, Conformity Assessment, and International Integration 89 Figure 5.2 The Internationalization of British Standards, 1948­2003 100 otalt 90 of 80 % ds as 70 ds standar 60 50 standar British 40 30 national 20 1948195119541957196019631966196919721975197819811984198719901993199619992002 Source: DTI 2005. cal content in one or several corresponding national standards. CEN has an agreement with ISO through which many of the standards are devel- oped in common. As a result, about 30 percent of European standards are identical to ISO standards (CEN 2004). The number of European standards grew from 774 in 1992 to 9,320 in 2005, and the number of harmonization documents grew from 85 to 1,554 during the same period (CEN 2005). Given the magnitude of the regional standards stock, it is not surprising that purely national standards now account for only a minority of standards in many European countries. In 2003, 26 percent of standards in the United Kingdom were purely national, a sharp decline from 64 percent in 1990 and 98 percent in 1948 (figure 5.2). Approaches to upgrading standards in developing countries In developing countries, upgrading standards toward international norms is necessary but costly. Standards that are idiosyncratic to developing countries are generally less demanding than international standards and refer to less modern technologies. For this reason, a standard can be con- sidered "upgraded" when its content has been changed to bring it closer to that of a corresponding international standard. Thus, not only does standards upgrading allow for greater economic integration, as discussed in chapter 2, but it also diffuses foreign technology and increases export quality. Moreover, the regional harmonization of standards between developed countries, such as in the EU, has increased intraregional trade while reducing exports from excluded developing countries. In an empir- ical study of standards, conformity assessment recognition agreements, 90 Quality Systems and Standards for a Competitive Edge and trade, Chen and Mattoo (2004) find that regional harmonization induces a 16 percent reduction in imports from developing countries that are excluded from the harmonization process. Developing countries can benefit from these integrated markets if they are able to align their own standards more closely with the harmonized standards. However, there are significant costs involved in standards upgrading. The technological content and the health, safety, and environmental objectives of international standards, which are usually developed by OECD countries, may not be consistent with a developing country's stage of technological and socioeconomic development. Upgrading uni- laterally to an intermediate standard is risky because trading partners may not consider the new standard up to par and market access is not guaranteed. Upgraded standards may also render an industry less com- petitive if neighboring trading partners have less demanding standards. Rather than unilaterally upgrading their standards, some develop- ing countries have adopted a coordinated approach in which a group of countries agrees to upgrade standards simultaneously. Products meeting the upgraded standards are guaranteed market access to all countries in the group. This is the approach used by APEC. The problem with such an agreement is that it is difficult to enforce. The agreement is not necessar- ily supported by formal regional institutions, and dispute panels are typi- cally given a mediation role instead of an arbitration role (Aldaz-Carroll 2006). Because a cooperation approach does not entail the elimination of other market barriers, members can also undermine the agreement by raising tariffs in a particular industry sector once standards have been harmonized. Regional trade agreements (RTAs) can be an effective forum for devel- oping countries to coordinate standards upgrading. RTAs are character- ized by more formal institutions, greater enforcement, and a greater level of trust caused by frequent interactions between RTA members. The comprehensive nature of RTAs allows for cross-issue retaliation. RTAs also imply the reduction of tariffs in some areas of trade, so it can be more difficult for countries to raise tariffs to protect a market once stan- dards have been harmonized. RTAs have adopted several strategies for standards harmonization. One approach, followed by the Association of Southeast Asian Nations (ASEAN), is based on harmonizing standards in key sectors around inter- national standards. Members are not forced to adopt the identified stan- dards as their national standards but are compelled to accept products from partners complying with these standards. This approach does not Opening Doors: Standards, Conformity Assessment, and International Integration 91 require standards development committees because regional standards are simply equated to international standards. Other RTAs have followed a gradual, coordinated approach to stan- dards upgrading. Here, members adopt regional standards and gradually upgrade them toward more demanding international standards.The main potential benefit of a gradual, coordinated approach is that it avoids the adoption of international standards that may not be aligned with a coun- try's level of development (Aldaz-Carroll 2006). Nonetheless, the costs of adopting a regional standard may outweigh the benefits if exports to the region of a particular product are small relative to its exports to other countries where international standards are required. In Latin America, the Andean Community and Mercosur (the Southern Cone Common Market) have adopted a gradual, coordinated approach to harmonize their standards and technical regulations. Some RTAs with weak regional institutions, such as free trade areas,5 have chosen to remove barriers to trade by making standards compatible with each other rather than harmonizing them. Unlike harmonization, compatibility does not imply the development of new regional stan- dards shared by more than one country. Instead, compatibility requires that the content of national standards in one country not conflict with the content of one or several standards in another country. This is the case of the North American Free Trade Agreement (NAFTA), which establishes committees and working groups to facilitate the develop- ment of compatible standards and technical regulations. The provi- sions of the NAFTA agreement are similar to the WTO TBT provisions and emphasize the need for members to adopt international standards, except where these would be inappropriate or would not effectively fulfill legitimate objectives. Besides lowering the cost and risk of upgrading, there are other ben- efits to regional coordination. These include the possibility of pooling resources in the area of international standards information diffusion, pooling resources to participate more effectively in shaping international standards, and dividing standardization tasks among countries with dif- ferent regulatory comparative advantages. Limiting standards harmoni- zation to small groups of countries can render the coordination process more effective by simplifying and expediting the harmonization process, although it decreases economies of scale. It is important to emphasize that in the short run, upgrading standards is not always beneficial because stricter regional standards may divert trade from low-cost extraregional partners with looser standards. 92 Quality Systems and Standards for a Competitive Edge The Need for International Recognition of National Accreditation Systems Mutual recognition of accreditation procedures promotes trade by decreasing transaction costs and eliminating technical barriers. When accreditation in one country is recognized by other countries, the work performed by that country's certification bodies, inspection bodies, and calibration and measurement laboratories will be accepted in other coun- tries. Countries that harmonize or upgrade their national standards will only gain greater market access if they can prove that their products and services truly conform to the standards. In the study by Chen and Mattoo (2004) described in chapter 2, the authors demonstrate that the regional mutual recognition of conformity assessment procedures increases intra- regional trade in affected industries and even promotes trade with third countries if there are no restrictive rules of origin. In the EU, for exam- ple, the European Commission estimates that the perfect operation of mutual recognition could yield trade benefits of as much as 1.8 percent of the EU's gross domestic product (GDP) (Commission of the Euro- pean Communities 2001). Membership in the two principal international organizations for accred- itation, the International Accreditation Forum (IAF) and the International Laboratory Accreditation Cooperation (ILAC), enhances an accreditation body's prospect of gaining international credibility. Accreditation bodies must demonstrate that they operate at high international standards to join the IAF or the ILAC. In this sense, IAF and ILAC membership is a form of assurance that accreditation bodies are competent to undertake their work and are not subject to conflicts of interest. Furthermore, these international organizations facilitate technology transfer in areas related to quality assessment and provide a forum for learning from other experi- enced accreditation systems.Although membership in these organizations enhances the reputation of an accreditation body, in no way does it guar- antee the recognition of its accreditation process by other countries. A first step to achieving mutual recognition of accreditation processes across countries is to adopt harmonized standards and guidelines for con- formity assessment procedures. Although trading partners can formulate and adopt their own requirements for accreditation, most countries have now decided to follow the international requirements published by ISO and the IEC. These requirements concern accreditation of certification and inspection bodies and measurement and calibration laboratories6 and establish the criteria for accreditation in areas concerning operational Opening Doors: Standards, Conformity Assessment, and International Integration 93 procedures, quality system, personnel, and equipment. However, recog- nizing that requirements for accreditation are equivalent across coun- tries is not very useful if decisions to grant accreditation are affected by idiosyncratic features of accreditation bodies. Hence, to ensure that accreditations in different countries are performed with similar levels of objectivity, impartiality, and transparency, the accreditation bodies them- selves should operate according to harmonized standards or guidelines. Such standards and guidelines have also been established by ISO and the IEC and are followed by accreditation bodies globally.7 For full recognition, accreditation bodies must establish agreements with other countries based on mutual evaluation and acceptance of each other's accreditation systems. Membership in a mutual recognition arrangement or agreement (MRA) at the bilateral level (with one other country) or at the regional or international level (with many other coun- tries) is critical to guaranteeing the credibility of the national confor- mity assessment system. MRAs are based on peer evaluation processes through which signatories evaluate each other's compliance with the agreed-upon requirements and evaluate the performance of assessment staff. They usually cover specific types of accreditation (for example, quality management system certifiers).8 MRAs are now widespread among industrial countries and affect an increasing portion of traded products. In an effort to integrate its internal market, the EU has long depended on mutual recognition of conformity assessment systems. In 2001, 21 percent of industrial production or 7 percent of GDP inside the EU was covered by mutual recognition (Com- mission of the European Communities 2001). The EU has exported the MRA model and has been at the forefront of the proliferation of bilateral MRAs with other countries. The MRA between the EU and the United States covers approximately $41 billion in bilateral trade (Maskus and Wilson 2000). The importance of the EU in world trade has created global incentives for countries to consider adopting the MRA model. Several regional accreditation bodies sponsor MRAs, including those in Asia and the Americas. At the international level, the IAF and the ILAC administer the most important mutual recognition agreements, referred to as multilateral recog- nition arrangements (MLA) in the case of the IAF (table 5.2). Often, sign- ing an MRA or MLA at the regional level is less demanding and represents a first step toward signing an international MRA. Some regional MRA organi- zations have signed the agreements of the IAF and the ILAC, thus extend- ing international MRA membership to all of their regional signatories. 94 Quality Systems and Standards for a Competitive Edge Table 5.2 Members of IAF and ILAC Mutual Recognition Agreements, 2006 MLA or MRA Number of member countries IAF Quality Management System MLA 42 IAF Environmental Management System MLA 38 IAF Product MLA 29 ILAC Testing MRA 50 ILAC Calibration MRA 50 Source: Wilson and Otsuki 2004. Note: Countries are only counted once, even if multiple national bodies are MRA or MLA signatories. National members of regional accreditation body members are counted as individual members. A survey of firms in developing countries reflects the potential benefits MRAs could have on trade with developed countries (Wilson and Otsuki 2004). Although the majority of firms (69 percent) do not subject their products to an MRA (table 5.3), most firms agree that it would be easier to export to developed countries if an MRA with those countries were in place (table 5.4). Table 5.3 Firms with Products Subject to an MRA The firm has products subject to an MRA Share of firms (%) Yes 23 No 69 Not answered 8 Source: Wilson and Otsuki 2004. Table 5.4 Perceived Export Advantage of Participating in an MRA with an Export Destination Country % of firms Much Somewhat No Somewhat Much Not Destination harder harder effect easier easier answered Australia 6 2 17 58 11 3 Canada 7 3 18 55 13 3 European Union 9 4 15 54 12 4 Japan 5 2 13 69 8 1 United States 7 3 15 62 10 2 Total 7 3 16 60 11 3 Source: Wilson and Otsuki 2004. Note: Rows may not total 100 because of rounding. Opening Doors: Standards, Conformity Assessment, and International Integration 95 The Need for International Recognition of National Metrology Institutes When the capabilities of a national metrology institute (NMI), its mea- surement units, or its physical embodiments of the national standards are not recognized internationally, the credibility of all measurements trace- able to that NMI is affected. Calibration and test certificates establishing product or process characteristics can be refused by overseas buyers or governments if they are not traceable to recognized measurements. Lack of recognition affects the entire accreditation system because interna- tional standards used to assess the competence of accredited laboratories and quality systems require the internationally recognized traceability of measurements and test results. In sum, the full recognition of conformity assessment systems requires the recognition of both the national accredi- tation body and the NMI. NMIs can increase the credibility of their metrology infrastructure by joining international metrology organizations and participating in interna- tional laboratory comparisons. Membership in regional organizations, such as the Inter-American Metrology System, and international organizations, such as the International Bureau of Weights and Measures, offers a host of advantages. These organizations allow NMIs to exchange information on the latest developments in metrology; they also coordinate the dissemi- nation and use of measurement units, facilitate technology transfer, and allow members to receive technical support from more experienced NMIs. International and regional metrology organizations also provide the insti- tutional framework to organize regional and international comparisons of national laboratories. These inter-laboratory comparisons play a key role in demonstrating the accuracy and equivalency of measurements. A mutual recognition arrangement in metrology provides for the for- mal recognition of national measurement standards and calibration capa- bilities. Through measurement comparisons, it establishes the degree of equivalence of national measurement standards maintained by NMIs and thereby guarantees the international acceptance of measurement results that are traceable to the NMI by the other MRA signatories. It also pro- vides governments and other parties a secure technical foundation for wider agreements related to trade and regulatory affairs. Although mutual recognition was originally conducted at the bilat- eral level, most countries now focus on joining regional or international MRAs. The most comprehensive multilateral MRA, that of the Interna- tional Committee of Weights and Measures (CIPM), is organized by the 96 Quality Systems and Standards for a Competitive Edge International Bureau of Weights and Measures and had 64 participating institutes as of January 1, 2006. As in most multilateral agreements, par- ticipation in the CIPM MRA is more cost-effective than participation in individual bilateral MRAs. An impact assessment conducted by KPMG Consulting (2001) concluded that if NMIs were to maintain separate individual MRAs with each partner NMI, the annual cost of establish- ing and maintaining mutual recognition with each partner NMI would be 75,000 greater than it currently is with the CIPM MRA. Exports between CIPM MRA participants accounted for 89 percent of world trade in merchandise in 2001 (Quinn 2003). As a result, the measure- ment comparisons required by the CIPM MRA have become the princi- pal reference for information on the technical capabilities of NMIs. The criteria for joining the CIPM MRA concern not only participation of NMIs in international measurement comparisons, but also verification through a peer-review process that the national calibration services make use of an internationally recognized quality system. International cooperation in legal metrology can also reduce technical barriers to trade. Bulk and prepackaged goods subject to metrology laws account for a significant portion of exports and national income in many countries. Without harmonized means and procedures for verifications and tests, it is difficult to establish whether metrological control is equiv- alent in different countries. Similarly, measuring instruments accepted in one country may not be accepted in another country because of differ- ing regulations or procedures for product approval. Membership in the International Organization of Legal Metrology (OIML) helps countries harmonize policies regarding the trade of products and services with a commercial value based on measurements, as well as trade in measur- ing instruments. When necessary, the OIML also provides governments with technical assistance for the development of sound metrology poli- cies.9 The OIML Certificate System for Measuring Instruments facilitates administrative procedures and lowers transaction costs in the trade of legal instruments subject to legal requirements.10 C H A P T E R 6 The ISO 9000 Quality Management Standards The ISO 9000 standards are the most widely known and fastest-growing international quality standards. They were first developed by the Interna- tional Organization for Standardization (ISO) in 1987 and represent the state of the art in quality management. The ISO 9000 family of standards is implemented in more than 670,000 organizations in 154 countries (figure 6.1). The Function of ISO 9000 ISO 9000 standards provide an organization with a model to follow for the design, implementation, and assessment of quality management sys- tems. Quality management refers to the steps an organization can take to fulfill the customer's requirements and regulatory requirements while aiming to enhance customer satisfaction and achieve continual improve- ment of its performance in pursuit of these objectives. Mechanisms are established to research and understand customer needs and expectations and act on the result. Special emphasis is placed on the consistent use of documented, standardized procedures to guide processes in the organi- zation. Processes that are subject to such constraints are more likely to enhance product uniformity and conformance to specifications. ISO 9000 standards are generic management system standards. They can be applied to any organization regardless of its size, products, sector, and activity, and regardless of whether it is a business enterprise, a public 97 98 Quality Systems and Standards for a Competitive Edge Figure 6.1 Global Adoption of ISO 9000 Standards 800,000 Worldwide total of ISO 9000 certificates 700,000 Worldwide total of ISO 9001:2000 certificates es 600,000 500,000 tificat erc 400,000 of 300,000 number200,000 100,000 0 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 Source: ISO Web site. agency, or a government department. In effect, ISO 9000 incorporates quality system requirements but does not dictate how they should be met. This provides great scope and flexibility for implementation in dif- ferent types of organizations. The ISO 9000 family of standards consists of three standards: · ISO 9000:2000 describes the fundamentals and vocabulary of quality management systems. · ISO 9001:2000 specifies requirements for quality management sys- tems.This standard was published in 2000 to improve and replace three previous 1994 versions of ISO 9001, ISO 9002, and ISO 9003.1 · ISO 9004:2000 provides guidelines beyond the requirements given in ISO 9001:2000 to consider both the effectiveness and the efficiency of a quality management system, and consequently the potential for improvement of the organization's performance. Certification bodies cannot register organizations against ISO 9004. In this book, the term "ISO 9000" will refer to the family of ISO 9000 standards recognized by the International Accreditation Forum for accredited certification. This includes ISO 9001:1994, ISO 9002:1994, and ISO 9003:1994 up to December 15, 2003, and only ISO 9001:2000 after that date. The ISO 9000 Quality Management Standards 99 ISO 9000 standards can bring benefits to an organization due to both internal and external factors. Benefits due to internal factors include · lower costs and shorter cycle time due to more effective use of resources; · higher-quality processes, leading to fewer costly inspections, warranty costs, and reworking; · greater customer focus, resulting in flexible and fast responses to mar- ket opportunities; · greater management involvement in improving quality performance and control over employee performance; and · better working conditions and motivation for employees. Benefits due to external factors include · greater consumer confidence that products will meet their require- ments or regulations, leading to an increase in customer base; · greater consumer satisfaction, leading to repeat purchases; and · better image of the organization. ISO 9000 and Competitive Advantage The overwhelming popularity of ISO 9000 certification has prompted much debate about the business value of this quality assurance system. While most of the debate has been fueled by anecdotal evidence from practitioners, 18 years of ISO 9000 have also provided plenty of opportu- nity for more rigorous empirical studies of the costs and benefits of certi- fication. The debate has been focused on two main questions. First, does ISO 9000 certification actually improve operational performance? And second, regardless of its effect on operational performance, does certifica- tion affect financial performance based on revenue growth and profitabil- ity? Unfortunately, much of the empirical work addressing these issues has produced conflicting results, and this has posed a serious challenge to unambiguous arguments in support of certification. Nonetheless, a closer examination of the literature helps explain some of the mixed evidence and does yield several reasons to support ISO 9000 certification. Why do rms get certi ed? A firm's motivation for seeking certification can affect the benefits it derives from the certification process. Firms seek ISO 9000 certification 100 Quality Systems and Standards for a Competitive Edge for two types of reasons: internal reasons, related to operational perfor- mance, and external reasons, related to market requirements. Opinions and empirical results have essentially converged regarding the effect of motivation on the ensuing benefits of certification. Many researchers posit that certification serves as a necessary but not sufficient condition for improving organizational performance and quality. Thus, firms with a genuine interest in establishing a quality management system will use certification as a foundation for further improvements and are likely to derive internal benefits. In contrast, firms pursuing certification as a reaction to market requirements tend to adopt a minimalist approach to quality management and derive few internal benefits. Firms seeking certification for internal reasons believe that ISO 9000 can help them improve their quality management systems (Casadesús and Giménez 2000; Singels, Ruël, and van de Water 2001). A quality management program can increase profitability by reducing defects, duplicate work, and waste, enhancing internal organization and commu- nication, and enabling a better response to customer needs. These firms see certification as a means of acquiring a better quality management system (Wayhan, Kirche, and Khumawala 2002; Terziovski, Samson, and Dow 1997). It is one step among many in a constant process of improv- ing productivity, efficiency, and quality. In contrast, firms motivated by external reasons see certification as an end in itself. They believe it will improve their image and allow them to enter new markets,or they may simply be responding to customer pressure. For these firms, certification serves as a signal to communicate desirable organizational attributes to potential customers. Customer requirements are now an increasingly important motivation for certification. Supply chain transactions are becoming more distant and international and it is difficult for buyers to observe the qualifications of suppliers (Brown, van der Wiele, and Loughton 1998; Terlaak and King 2006; Terziovski, Sam- son, and Dow 1997). As other companies acquire certification, firms pur- sue certification to avoid losing their competitive positions. Does certi cation improve organizational performance? Several empirical studies have explored the relationship between ISO 9000 certification and internal benefits in organizational performance, with mixed results. On one side of the debate, some studies have found a positive relationship between certification and organizational performance. A cross- sectional analysis of 649 mostly large firms in China, India, Mexico, and the United States by Rao, Ragu-Nathan, and Solis (1997) found that certified The ISO 9000 Quality Management Standards 101 firms had higher levels of implementation of quality management practices and higher-quality products or processes.2 Firms planning to get registered for ISO 9000 had similar results as firms with no interest in registration. Given that obtaining registration for ISO 9000 would represent a small additional cost relative to the total costs of ISO 9000 implementation, it is reasonable to assume that the firms that are not interested in registration have not already complied with the internal requirements of ISO 9000. This suggests an actual role for certification in yielding internal benefits, as opposed to a simple self-selection of superior firms into certification. In a survey of 288 Spanish companies, Casadesús and Giménez (2000) found that firms gained from certification through a number of internal benefits related to the implementation of a quality system. These cross-sectional studies do not necessarily imply causality. Sharma (2005) was able to infer causality through a time-series study of 70 firms in Singapore over a six-year period. The author found that certi- fication was associated with improvements in operating efficiency, based on profit margins, but also with growth of sales and improved overall financial performance based on earnings per share. The effects of certifi- cation were greater on profit margin than on growth of sales, suggesting that an improvement in overall performance was largely due to improve- ments in internal business processes. On the negative side, other studies have found limited or no effect of certification on organizational performance. Terziovski, Samson, and Dow's (1997) cross-sectional analysis of 858 manufacturing firms in Aus- tralia and New Zealand showed that ISO 9000 certification was a poor predictor of organizational performance and quality. The authors argue that many companies pursue certification to satisfy customer require- ments but revert to traditional practices immediately after certification, thus nullifying any potential beneficial effect of certification. Quazi, Wing Hong, and Tuck Meng (2002) administered Rao and colleagues' question- naire to 93 Singaporean firms and found no relationship between certifica- tion and quality management practices and quality results. Disagreements with Rao and colleagues' results could be due to differences in the profile of the firms in the surveys. Rao and colleagues' survey consisted mainly of large firms in the manufacturing industry, whereas Quazi and colleagues' survey consisted mainly of small and medium enterprises (SMEs) in a broader range of sectors. Recent work by Terlaak and King (2006), based on an 11-year panel study of 19,713 manufacturing facilities in the United States, found that, controlling for inventory size, ISO 9000 did not have a significant effect on operational improvements, proxied by the generation 102 Quality Systems and Standards for a Competitive Edge of process waste (or scrap). However, this study must be interpreted with care because the amount of process waste is only one measure among many of a quality system. This contrasts with the studies by Rao, Ragu- Nathan, and Solis; Casadesús and Giménez;Terziovski, Samson, and Dow; and Quazi, Wing Hong, and Tuck Meng, who included more comprehen- sive measures of the implementation of a quality system. Does certi cation improve nancial performance? From a manager's perspective, what may ultimately matter most is whether certification leads to improved financial performance. Here, empirical results suggest different degrees of performance improvements. An empiri- cal study of 146 large firms in Singapore by Chow-Chua, Goh, and Wan (2003) showed that certification had indeed led to better overall financial performance. These conclusions were based on a survey of perceptions of certification on business performance and an examination of financial data over a 10-year period. In their study of U.S. manufacturing facilities,Terlaak and King (2006) found that production volumes grew faster after certifi- cation, even after controlling for operational performance and inventory. Hence, although certification did have some business value, it was ultimately functioning as a market signal that reduced information asymmetries and helped differentiate between high-quality and low-quality suppliers. Recent work based on firm-level surveys of investment climate factors in develop- ing economies found that ISO 9000 certification had a significant impact on several measures of productivity.Average productivity gains were estimated to be between 2.4 percent and 17.6 percent for three Central American economies, less than 1 percent for four Southeast Asian economies, and 4.5 percent in China (Escribano and Guasch 2005a, 2005b). Other studies have provided less optimistic conclusions about the financial performance implications of ISO 9000 certification. Heras, Dick, and Casadesús (2002) examined sales and profitability of 800 Spanish firms over five years and found that the performance of companies that sought certification was likely to be superior to that of companies that did not seek certification, but that this was already true before certifica- tion; performance did not actually improve after certification. Further- more, firms demonstrating an intent to pursue certification performed as well as certified firms. This seemed to agree with Terlaak and King's (2006) implication that firms with superior performance show a greater propensity to become certified. But in contrast to Terlaak and King's con- clusion, the firms examined by Heras, Dick, and Casadesús (2002) did not benefit from certification. Nonetheless, the Heras, Dick, and Casa- The ISO 9000 Quality Management Standards 103 desús study does not rule out the possibility that the prospect of obtaining certification changes a firm's organizational practices and enhances its performance in the years preceding certification. In a study of 48 manufacturing companies in the United States over the 1990­98 period, Wayhan, Kirche, and Khumawala (2002) found a very limited impact of certification on financial performance, as measured by returns on assets. Moreover, this effect dissipated quickly over time and ISO 9000 did not affect any other indicators of financial performance such as revenue, stockholder equity, or gross profit. This could be linked to Terzi- ovski, Samson, and Dow's observation that firms are unwilling to accept and use the standard operating procedures defined in ISO 9000 once they have completed the certification process. However, Wayhan, Kirche, and Khu- mawaki do note that during the period of the study U.S. suppliers had not yet begun to require their suppliers to be ISO 9000 certified. Of interest to Latin America, Lima, Resende, and Hasenclever (2000) compared certified and noncertified firms in Brazil over the 1992­98 period and concluded that ISO 9000 certification had an effect on only one of five financial per- formance indicators, sales to total assets, which increased with certification. None of the indicators related to operational or net income were affected, perhaps suggesting that certification increases costs as well as revenues. However, Lima, Resende, and Havenclever note that the accounting indica- tors that were used could not measure changes in productive efficiency. In view of the mixed evidence, is there still a role for ISO 9000? In spite of conflicting views on the overall effect of certification, researchers on both sides of the debate acknowledge that ISO 9000 does provide the potential for improving organizational performance and producing higher- quality products and services (Escanciano, Fernández, and Vázquez 2002). Terziovski, Samson, and Dow (1997) argue that the role of certification has been misunderstood. While many managers view certification as an end in itself, either for marketing or for internal reasons, conformance to ISO 9000 alone cannot produce sustainable improvements in organizational perfor- mance. Rather, managers must use it as a means to implement better quality management systems, such as a total quality management (TQM) scheme. TQM is a comprehensive approach to quality leadership in which quality tools and techniques are applied to all functions and all levels of an organization. There seems to be broad agreement that intangible factors that are difficult to transcribe in a standard, such as employee empowerment and capable senior management, have a great influence on a firm's ability to implement TQM. Singels, Ruël, and van de Water 104 Quality Systems and Standards for a Competitive Edge (2001) point out that the procedures described in ISO 9000 guarantee con- stant product quality, a precursor to higher quality. Thus, ISO 9000 is a necessary condition for good product quality, but certified organizations do not automatically have good product quality. This seems to be the opin- ion of managers as well. In a survey of Australian firms by Brown, van der Wiele, and Loughton (1998), a majority of respondents viewed ISO 9000 certification as a first step in implementing a TQM program. In fact, very few companies thought that TQM could be implemented before ISO 9000 certification. A survey of Singaporean SMEs by Quazi and Padibjo (1998) found that ISO 9000 provided a "stepping stone" toward TQM practices. Empirical evidence confirms that a firm will benefit from certification if it is truly working toward a better quality assurance system. Evidence of such a relationship was first unearthed in surveys of British firms by the Science and Engineering Policy Studies Unit (Heras, Dick, and Casa- desús 2002; Dick 2000). The study was limited to descriptive statistics but found that companies that cited consumer pressure as a motive for certification were less likely to report improvements (Terziovski, Samson, and Dow 1997). Jones, Arndt, and Kustin (1997) conducted a survey of 272 certified Australian companies and found that those companies that viewed certification as an instrument to satisfy customer requirements were less likely to report having achieved benefits from certification than companies that pursued certification for internal reasons. Interestingly, this second group of proactive and developmental companies represented a low proportion of respondents. This could provide an explanation for the lack of internal benefits observed in a number of empirical studies. Another analysis of a survey of 160 Australian firms, mostly SMEs, by Brown, van der Wiele, and Loughton (1998) led to the same conclu- sion. Brown, Arndt, and Kustin echoed Jones, van der Wiele, and Lough- ton in noting that the external reasons for seeking certification were far more important than the internal reasons, but that a stronger drive to achieve internal benefits led to a more positive perception of the organi- zational, quality, and financial improvements realized. Singels, Ruël, and van de Water (2001) investigated the relationship between certification motives and benefits by examining the organizational performance of 192 organizations in the Netherlands. The authors found that, on the whole, certification did not lead to statistically significant improvements in organizational performance. However, when motivations for certifica- tion were accounted for, the study found that organizations that pursued certification out of an internal motivation profited in terms of perfor- mance outcomes. The ISO 9000 Quality Management Standards 105 Finally, it is worth noting the contribution of quality adoption, par- ticularly of ISO 9000, to export performance, and the best example is China. Chinese firms understood very early the need to incorporate qual- ity standards (box 6.1). That was critical for the remarkable performance of Chinese export growth during the last two decades. Conclusion Table 6.1 summarizes the literature results discussed in this section. Two important points emerge from the existing evidence. First, firms with superior organizational performance and quality management practices are more likely to obtain certification. This implies that ISO 9000 can be used as a market signal for distinguishing high-quality firms from low- quality firms. Second, a firm's motives for pursuing ISO 9000 certifica- tion have an effect on the internal benefits of certification. Firms that view certification as a genuine tool for improving their quality manage- ment system and as a first step toward TQM are more likely to gain from certification. Given this result, if certification is used to improve organi- zational performance, it should involve follow-up processes that build on ISO 9000 to achieve quality leadership. Internal Firm-Level Barriers to ISO 9000 Certification Despite the potential business value of implementing a quality manage- ment system, a number of internal barriers can hamper the adoption of ISO 9000 by firms. The most frequently cited barriers to certification are lack of quality awareness and resistance to change; difficulty of under- standing and interpreting standards; the time commitment required; lim- ited financial resources; and limited availability of calibration services or equipment (Santos 2002; Schuurman 1997; ITC 2004; Brown, van der Wiele, and Loughton 1998; Karapetrovic, Rajamani, and Willborn 1997). Many of these obstacles are particularly difficult for SMEs to overcome because of their lower levels of human and capital resources and their relative isolation. Lack of awareness and understanding of quality and standards A lack of quality awareness among firm management and staff can con- stitute one of the greatest obstacles to the implementation of a quality management system. This problem is particularly acute in small firms, where managers generally have fewer interactions with external knowl- edge networks and are less familiar with the latest quality management 106 Quality Systems and Standards for a Competitive Edge Box 6.1 China and the Drive for Quality The remarkable economic performance of China since the early 1990s--annual GDP growth rates near 10 percent, extraordinary export sales, huge trade sur- pluses--is often attributed at least partially to its low labor costs. But cheap labor alone cannot explain China's success. Bangladesh has labor costs that are half those of China, yet its performance has been nowhere near China's. While China does indeed have low labor costs, at least relative to many countries, the reasons for its success are more complex. One important and often overlooked factor is China's determined drive to adopt quality standards. If ISO certi cation is used as a proxy for adoption of quality standards, Chi- na's progress and its current predominance become clear. In the late 1980s, the number of rms in China that had adopted ISO 9000 or 9001 was very low, just a few thousand rms. By 2003, however, China was leading all countries in the number of rms with ISO 9000 and ISO 9001 certi cation, as the gure on the following page shows. China had 75,755 rms with ISO 9000, compared with 61,212 such rms in Italy, 38,927 in the United States, and 33,964 in Japan. Similarly, China led in the number of rms with ISO 9001, with 40,997 certi ed rms; the second country in the ranking, Japan, had 16,813, and the United States had only 4,587. Latin American countries lag signi cantly in obtaining certi cation: in 2003 Brazil had 6,120 rms with ISO 9000, Argentina had 4,149, and Colombia had 4,120. Mexico had only 2,508 rms with ISO 9000 and 265 rms with ISO 9001. practices. Furthermore, SME managers tend to be risk averse and may be reluctant to abandon their traditional management practices. When managers do decide to implement a quality management system, their lack of knowledge or confidence in ISO 9000 can deter them from committing sufficient time and resources to the process.This can compro- mise their chance of obtaining certification (ITC 2004; Santos 2002). Top and middle management may be especially reluctant to adopt ISO 9000 if they perceive the requisite changes in power structures as undermining their leadership (Schuurman 1997). Resistance to change may also be found among employees who do not understand the ultimate benefits of reorganizing their tasks or taking on different responsibilities. Employee reluctance appears to be more pronounced in SMEs than in larger firms The ISO 9000 Quality Management Standards 107 China and its rms thus understood very quickly the need to use and adopt quality standards, setting up an e ective national quality system. While the num- bers for ISO certi cation are particularly easy to track, similar trends with respect to China's performance exist for just about any other indicator of quality adoption. Top Ten Countries for ISO 9000 and ISO 9001 Certificates, with Mexico Comparator 80,000 75,755 ISO 9000:2000 certificates 70,000 ISO 9001:2000 certificates 61,212 60,960 60,000 es 50,000 tificat erc 40,997 40,000 38,927 of 35,802 33,964 30,000 28,690 27,135 number 19,870 20,000 16,813 14,733 10,811 9,301 10,299 10,000 8,872 7,024 6,529 5,060 4,587 2,508 265 0 y o Italy es China Japan Spain Stat German Kingdom AustraliaFrance itzerlanded Mexic ed Sw Unit Unit Source: ISO 2005. and was found to be the top obstacle to certification in surveys of certified micro and small enterprises in Brazil (Santos 2002) and of certified SMEs in Australia (Brown, van der Wiele, and Loughton 1998). It is difficult for individuals outside the quality management profes- sion to understand and interpret ISO 9000 standards. Due to greater indivisibilities of labor, personnel in small firms tend to be less specialized than those in larger firms. Small firms may not have the human resources to establish a quality department or to maintain staff specifically dedi- cated to quality. Furthermore, SMEs tend to invest less in training than larger firms, which limits the skill levels of their workers and their abil- ity to interpret ISO 9000 standards. This poses a serious obstacle to the adoption of ISO 9000. In Brazil, a survey of certified firms found that Table 6.1 Summary of Studies Evaluating the Effect of Certification on Firm Performance Study Model Performance measure Effect of certification Rao, Ragu-Nathan, Cross-sectional, mainly Quality management + and Solis (1997) large manufacturing practices and output rms, United States, quality India, China, and Mexico Casadesús and Cross-sectional, Spain Quality management ++ Giménez (2002) practices Sharma (2005) Time-series, 6 years, Operating e ciency ++ Singapore and nancial performance Terziovski, Samson, Cross-sectional, Organizational 0 and Dow (1997) manufacturing rms, performance and Australia and New output quality Zealand Quazi, Wing Hong, Cross-sectional, mainly Quality management 0 and Tuck Meng SMEs, Singapore practices and output (2002) quality Terlaak and King Time-series, 11 years, Operational and 0 (2006) manufacturing rms, quality improvements, ++ United States production volumes Chow-Chua, Goh, Times-series, 10 years, Overall nancial ++ and Wan (2003) large rms, Singapore performance Escribano and 2 years, 9 developing Productivity and ++ Guasch (2005a, countries exports 2005b) Heras, Dick, and Time-series, 5 years, Financial performance 0 (but performance Casadesús (2002) Spain of certi ed rms is already superior pre-certi cation) Wayhan, Kirche, and Time series, 10 years, Financial performance + (e ect dissipates Khumawala (2002) manufacturing rms, over time) United States Lima, Resende, and Time-series, 7 years, Brazil Financial performance + Hasenclever (2000) Jones, Arndt, and Cross-sectional, Australia Organizational Internally motivated Kustin (1997) performance and rms: ++ output quality Others: + Brown, van Cross-sectional, Australia Organizational Internally motivated der Wiele, and performance rms: ++ Loughton (1998) Others: 0 Singels, Ruël, and Cross-sectional, Organizational Internally motivated van de Water (2001) Netherlands performance rms: ++ Others: 0 Source: Authors'compilation. Note: 0 = no e ect, + = slight positive e ect, ++ = signi cant positive e ect. 108 The ISO 9000 Quality Management Standards 109 Table 6.2 Firms Using External Consulting Services for Certification in Two Brazilian Surveys Survey Firm size Proportion of firms using consulting services (%) SEBRAE 1994 Small 80 Medium 45 Large 25 INMETRO 1996 <100 employees 70 All rms 55 Source: Schuurman 1997. 25 percent of the firms had difficulty understanding ISO 9000 (Santos 2002). While this survey represented firms of all sizes, another Brazilian survey of exclusively micro and small enterprises found that 72 percent of them considered their lack of understanding of ISO 9000 to be an obstacle to certification. As a result, smaller firms need to rely on external consulting services to successfully implement a quality management sys- tem (table 6.2). Reluctance to devolve such responsibilities to outsiders can act as a barrier to certification for smaller firms. Limited time and nancial resources Implementing a quality management system requires significant time commitments from all company personnel. Implementation time de- pends on many factors, including the complexity of the company, the current level of quality, the skill level of the personnel, and the degree of management commitment. Companies usually dedicate a first phase of the process to training of management and employees, followed by a second phase in which the company adapts its procedures to ISO 9000. Because ISO 9000 targets all processes in the enterprise, every staff member needs to dedicate some time to implementing the quality system in addition to their existing tasks. This may require extending the total number of person-hours and even hiring new personnel. Table 6.3 shows that in Australia, Canada, and the United States, smaller firms are able to obtain certification faster than larger firms, possibly a reflection of the more complex tasks performed by larger firms. The relationship is the opposite in Argentina, where smaller firms take longer than larger firms. Overall implementation times are also longer in Argentina than in Australia, Canada, and the United States. The financial costs of implementing a quality management system can be a barrier to certification, especially for SMEs. Table 6.3 lists average Table 6.3 Total ISO 9000 Implementation and Certification Costs Registration Total costs as Average total Average costs % of total implementation implementation as % of implementation Country Year Firm size costs (US$) time (months) turnover costs United 1999 small and 72,502 -- -- -- States, medium Canadaa medium- 106,890 -- -- -- large United 1996 small and 71,000 14 1.20 16 States, medium Canadab medium 102,000 15 0.50 15 United 1994 medium 250,000 -- -- -- Statesc Australiad 1995 small 55,000 7­12 -- -- medium 83,000 10­16 -- -- large 144,000 13­18 -- -- Argentinae 1995 small 80,000­100,000 24 -- -- medium -- 21.3 -- -- large 20,000­40,000 18.3 -- -- Brazilf 1999 small 77,323 15 -- -- medium 166,737 -- -- -- large 526,034 -- -- -- Saudi 2003 all 84,104 -- -- 12 Arabiag South 1998 small 13,321 -- 0.30 12 Africah medium 31,937 -- 0.05 17 large 72,795 -- 0.05 13 Sources: a. Wilson, Walsh, and Needy 2002; b. Schuurman 1997; c. Stevenson and Barnes 2001; d. Schuurman 1997; e. Ramos 1995; f. Santos 2002; g. Magd, Kadasah, and Curry 2003; h. Turner, Ortmann, and Lyne 2000. Note: -- = Not available. Firm sizes are de ned as follows: a. Small and medium = less than $25 million annual sales; medium-large = $25 million to $200 million annual sales. b. Small and medium = less than $11 million annual sales; medium = $11 million to $25 million annual sales. c. Medium = $25 million annual sales. d. Small = less than 50 employees; medium = 50­350 employees; large = more than 350 employees. e. No de nitions. f. Small = 20­99 employees in industry and 10­49 in commerce; medium = 100­499 employees in industry and 50­249 in commerce; large = 500 or more employees in industry and 250 or more in commerce. h. Small = less than $18 million turnover; medium = $18 million to $117 million turnover; large = more than $117 million turnover. 110 The ISO 9000 Quality Management Standards 111 total ISO 9000 certification costs by firm size in various countries. The level of those costs for firms in developing countries is quite significant and even more so when converted into purchasing power parity. So, per- haps not surprisingly, one does not see a rush toward certification, espe- cially among small firms. Cross-country comparisons are difficult to interpret due to differ- ing definitions of firm size, but the survey data do not seem to indi- cate substantial variations in costs across countries, except for South Africa, where the costs are much lower.3 Although in most countries costs increase with firm size, a survey of firms in Argentina showed an opposite relationship. Here, small firms had to spend more than twice as much as large firms to become certified. Ramos (1995) attributes this to small firms having weaker quality management systems and having to spend much more of their resources to upgrade them. In general, SMEs are more likely than large firms to view certification costs as a barrier. A Brazilian survey showed that 85 percent of small companies, 75 percent of medium companies, and 55 percent of large firms reported difficulties in obtaining the necessary resources for certification (Schuurman 1997). This is not only because SMEs have less access to finance than their larger counterparts, but also because certification tends to represent a larger share of turnover for smaller enterprises (table 6.3). All that begs the question of whether there is a role for the public to assist in the financing of quality certification. Certification costs depend on a company's size and complexity and its existing level of preparedness for the process. Training is often one of the most important costs involved in certification (figure 6.2). Key staff members need to receive specialized training to maintain the qual- ity system, and all personnel must receive some general training on the operational aspects of ISO 9000. Specialized technical training is par- ticularly needed when a firm introduces new technologies to implement a quality management system. Many firms, and especially SMEs, rely on external training services if they have not already been operating some form of quality management. Also, there can be significant costs associ- ated with the additional salaries that need to be paid to cover the time spent on implementing the quality management system. These costs may be significant if workers are poorly trained and spend a lot of time imple- menting the quality system or if additional workers need to be hired to supervise and maintain the system. Purchases related to new equipment and facilities also account for a major share of costs (figure 6.2), and these costs vary widely across 112 Quality Systems and Standards for a Competitive Edge Figure 6.2 Share of Brazilian Firms Reporting Important Certification Investments in Different Categories training 78 equipment 57 consulting 55 registration 41 personnel 28 calibration 28 facilities 22 documentation 18 0 10 20 30 40 50 60 70 80 90 % of firms reporting investments Source: Santos 2002 using data from Federação das Indústrias do Estado do Rio Grande do Sul. industries, depending on capital intensiveness. In addition to purchasing equipment, companies may also incur costs to have existing equipment calibrated by third-party laboratories. External consultant fees consti- tute another significant cost (figure 6.2 and table 6.4). Consultants can assist firms with any part of the certification procedure or they can be employed to take a company through the entire implementation process. These costs are usually a greater burden for smaller firms that do not have the internal capabilities to implement ISO 9000 on their own. Registration costs account for a minor share of total ISO 9000 adop- tion costs, but they can still act as a barrier to entry for some small firms. To receive an official ISO 9000 certificate and be included in the certi- fication body's registry, firms have to request an audit from a certifica- tion body.The actual ISO 9000 registration process includes auditing and Table 6.4 Typical Consulting Costs for Firms Seeking Certification, Argentina, 2005 Firm size Cost of certification-related consulting Up to 10 employees $8,000­$10,000 + travel expenses 11­100 employees $14,000­$17,000 + travel expenses More than 100 employees At least $23,000 + travel expenses Source: Informal interviews with certi cation bodies in Argentina. The ISO 9000 Quality Management Standards 113 Table 6.5 Typical ISO 9000 Registration Costs in Latin America, 2005 Country Firm size Cost of registration Argentina Up to 10 employees $1,500 + travel expenses 11­100 employees $2,500 + travel expenses More than 100 employees $4,000­$5,000 + travel expenses Ecuador All rms $400­$1,000 including 2 follow-ups Mexico 26­45 employees $3,900 + $1,700 per follow-up 66­85 employees $5,700 + $2,300 per follow-up 172­275 employees $8,000 + $3,100 per follow-up Peru Up to 10 employees $1,400 including 2 follow-ups 86­110 employees $4,400 including 2 follow-ups Source: Interviews with certi cation bodies in Argentina, Ecuador, Mexico, and Peru. administration costs that the certification body determines according to the size and complexity of the enterprise. The costs are affected by the time that auditors spend on the evaluation. Registration costs can vary across countries but generally do not exceed a few thousand U.S. dollars for small firms (table 6.5). ISO 9000 registration is usually valid for three years, during which time the certification body conducts two follow-up evaluations, with additional costs to the firm. The amount paid to the certification body accounts for a minor share of total ISO 9000 adoption costs, usually around 12 to 17 percent (table 6.3), but these fees can still act as a barrier to certification for some firms. Limited supply of services Quality support services necessary for ISO 9000 certification may be difficult to find. Depending on the nature of the firm, the implementa- tion of a quality management system may require the use of external consulting, training, inspection, testing, and calibration services. While this is not usually a problem in industrial countries, these services may be unavailable in developing countries. Consulting and training services may be required to help firms understand the complex technical require- ments of ISO 9000 and relate them to the organizational management of the firm. Testing and inspection can be required to meet customer qual- ity requirements but will constitute a serious barrier to certification if they are not available locally. Quality control requires that measurement- related equipment be traceable to an accepted reference standard. Some- times neither calibration equipment nor services are available to the firm. 114 Quality Systems and Standards for a Competitive Edge In a survey of certified Brazilian firms, 15 percent mentioned that the unavailability of calibration services and equipment was an obstacle to certification, and 10 percent mentioned that the lack of a traceable mea- surement standard was an obstacle (Santos 2002). Finally, firms may find themselves in isolated areas where they do not have access to certification bodies and auditors able to provide them with ISO 9000 registration. National Factors Influencing the Diffusion of ISO 9000 The pattern of ISO 9000 adoption is highly uneven across countries and regions. Figure 6.3 shows that Europe and the Asia Pacific region account for about 85 percent of worldwide certificates, while Central and South America together account for a mere 3 percent of certificates. In addi- tion to firm-level factors, a number of country-specific factors can create opportunities and incentives for ISO 9000 adoption by affecting access to information, expected costs, and expected benefits. These factors can be broadly categorized as market factors, shaped by the characteristics of the individual market actors and the market structure, and the eco- nomic environment, shaped by the institutions, the infrastructure, and the inputs available to market actors. Figure 6.3 Regional Shares of Worldwide ISO 9000 Certificates, 2004 Africa Australia and New Zealand 0.5% 3.0% East Asia and the Pacific Western Europe 25.9% 41.1% Europe and Central Asia 7.8% South Asia Japan 2.0% 7.3% Others Latin America and the Caribbean 0.2% 3.0% North America Middle East and North Africa 7.0% 2.1% Source: ISO 2005. Note: Developing countries are grouped according to the World Bank regional classi cation system. The ISO 9000 Quality Management Standards 115 Market factors International trade. International trade can spur the diffusion of ISO 9000. Firms often cite market pressure as an important reason for seek- ing certification. Many international buyers routinely require ISO 9000 as a contractual condition for suppliers. This is especially true for buyers in countries where certification is widespread. These buyers are likely to understand the need for certification and are integrated in supply chains where certification is the norm. Trade also encourages formal and informal interactions among business actors from different countries. These interactions serve as information channels through which tech- nical knowledge and experiences about organizational practices can be exchanged. Through these channels, firms that export are likely to gain access to information on the implications and benefits of certification. Countries that export a large share of their output have greater incen- tives to become certified, especially when they export to regions where certification is already widespread, such as the European Union. Two panel data studies of certification and bilateral trade by Guler, Guillén, and Muir Macpherson (2002) and Neumayer and Perkins (2005) show that trade with countries that have high certification rates has a positive effect on domestic certification. Using annual ISO 9000 certificates in 34 countries over the 1993­98 period, Guler, Guillén, and Muir Macpherson find that the number of certificates significantly increases when countries have strong trade ties with other countries that have a high number of certificates. Neu- mayer and Perkins find that the number of ISO 9000 certificates in 130 countries over the 1995­2001 period is positively correlated to exports to a market with a high number of certificates (in this case the EU). Grajek (2004) investigates the directions of causality between certifica- tion and trade, using data ranging over 1995­2001 and covering 101 coun- tries.4 His regressions confirm that while ISO 9000 does in fact increase bilateral exports, there is also a positive effect of foreign customers' adop- tion of ISO 9000 on domestic adoptions. Albuquerque, Bronnenberg, and Corbett (2004) use an ISO diffusion model on 56 countries over nine years to examine cross-country influences. They find bilateral trade flows to be a strong determinant of cross-country influences in ISO 9000 adop- tion. Corbett (2003) combines firm-level and country-level data for nine countries in another diffusion model to explore global ISO 9000 adoption mechanisms. Corbett first shows that Europe has the highest proportion of customers requiring ISO 9000 certification, followed by North America and Japan, then Asia, South America, and Africa. The author then demon- strates that in countries considered to be late adopters of ISO 9000, firms 116 Quality Systems and Standards for a Competitive Edge with higher exports to early-adopting countries (that is, the EU) tend to become certified earlier. Anderson, Daly, and Johnson (1999) use financial and accounting panel data for North American firms over a six-year period to show that industries exporting to Europe are more likely to seek certi- fication. According to these authors, this is consistent with the greater role of ISO 9000 as a regulatory and customer requirement in the EU. Type of sector and market structure. Although ISO 9000 is generic and can be applied to any organization, some industrial sectors have shown a greater preference for this standard and have adopted it more rapidly than others. This is in part because quality is a more important com- petition factor in some sectors than in others. For example, sectors like mining and agriculture generally do not rely on quality to compete, but on low production and transportation costs. ISO 9000 certification has been more prevalent in the manufacturing industry, where quality and safety are most important to output, and services are starting to repre- sent an increasing share of certificates (figure 6.4). In 2000, manufactur- Figure 6.4 Top 10 Sectors Registering the Most Worldwide ISO 9000 Certificates, 2004 construction basic metals and fabricated metal products electrical and optical equipment machinery and equipment wholesale and retail trade; repair of motor vehicles, motorcycles, and personal and household goods other services rubber and plastic products chemicals, chemical products, and fibers food products, beverages, and tobacco transport, storage, and communication 0 10 20 30 40 50 60 70 80 thousands of certificates Source: ISO 2005. The ISO 9000 Quality Management Standards 117 ing accounted for 64 percent and services for 25 percent of worldwide certificates (ISO 2001). Neumayer and Perkins's (2005) empirical study confirms the positive relationship between the share of the manufactur- ing sector in the economy and ISO 9000 diffusion rates.5 Figure 6.4 displays aggregate numbers of worldwide certificates by sec- tor. However, these data must be interpreted carefully because they do not appropriately represent the relative importance of certification in a particular sector. A large number of certificates in a sector may simply reflect the economic size of the sector or the degree of atomization. Cer- tification intensity within a sector could be better captured by standard- izing the number of certificates by number of productive facilities or the level of economic output. Unfortunately, global data on output and facili- ties are neither available nor reliable for many countries. Nonetheless, for tradable goods, the number of worldwide ISO 9000 certificates can be standardized by global export volume, as shown in figure 6.5. Again, Figure 6.5 Top 10 Sectors Registering the Most Worldwide ISO 9000 Certificates, Standardized by Global Export Volume, 2004 basic metals and fabricated metal products rubber and plastic products leather and leather products machinery and equipment chemicals, chemical products, and fibers food products, beverages, and tobacco textiles and textile products wood and wood products pulp, paper, and paper products electrical and optical equipment 0 20 40 60 80 100 120 140 number of certificates per US$1 billion of global exports Sources: ISO 2005; UNCTAD 2004. 118 Quality Systems and Standards for a Competitive Edge in terms of traded goods, manufactured products appear to exhibit the highest certification intensity. Some industries may exhibit low rates of ISO 9000 certification because they use other quality system standards. These include standards such as Hazard Analysis and Critical Control Point (HACCP), which addresses food safety, and an increasing number of sector-specific quality management standards based on ISO 9001, such as QS 9000 and ISO/TS 16949 in the automobile industry, TL 9000 in the telecommunications industry, and ISO 13485 for the medical instrument industry. The average size of firms in a sector affects the number of certificates. Micro, small, and medium enterprises generally show less propensity to adopt ISO 9000 than large firms. Not only do small enterprises have fewer financial and human resources to implement quality management system techniques, but they are also generally less aware of quality issues and more resistant to change. Furthermore, ISO 9000 implementation and certification costs represent a larger percentage of total turnover in smaller firms, and it takes longer before benefits are captured (Schuur- man 1997). These factors provide disincentives for certification in sectors dominated by SMEs. Type of market players. The presence of transnational corporations (TNCs) supports the diffusion of ISO 9000 in an economy. TNCs often originate in developed countries where certification is widespread. These firms operate using best-practice organizational forms and their corpo- rate strategies often require them and their subsidiaries to implement ISO 9000 (Schuurman 1997). Thus, the mere presence of foreign affili- ates can be expected to create a local demand for certification. But TNCs also encourage ISO 9000 adoption in other ways. Much like foreign importers, TNCs operate at high international standards of qual- ity and demand that their local suppliers also become certified. Furthermore, TNCs are recognized for their superior management practices and viewed as models by local organizations (Mendel 2000). Local firms will be inclined to mimic the organizational practices of TNCs because these practices are seen as fostering success. Moreover, the interaction of foreign affiliates with foreign peers and local organizations will facilitate the transfer of technical knowledge across borders. Finally, TNC personnel gaining employment in domestic firms or spinning off their own companies will have been acquainted with the culture of qual- ity and will act as channels of knowledge transfer on ISO 9000 quality management systems. The ISO 9000 Quality Management Standards 119 A few empirical studies have demonstrated the effect of TNCs on ISO 9000 diffusion. Using panel data on global ISO 9000 diffusion, both Guler, Guillén, and Muir Macpherson (2002) and Neumayer and Perkins (2005) found a statistically significant relationship between the presence of foreign multinationals and the number of ISO 9000 certificates.6 The presence of foreign multinational corporations was proxied by the value of inward foreign direct investment stock. In many countries, the government has been an important market force for the diffusion of certification. Governments have played the roles of both supplier and producer in many of the industries that provided the initial thrust for certification, such as energy, defense, and telecommuni- cations equipment (Guler, Guillén, and Muir Macpherson 2002). Gov- ernment agencies throughout the world now require their contractors to become certified. Even in the United States, considered a late adopter of ISO 9000, the Department of Defense, the Department of Energy, the Food and Drug Administration, and the National Aeronautics and Space Administration all require ISO 9000 certification from their large con- tract suppliers (Rao, Ragu-Nathan, and Solis 1997; Guler, Guillén, and Muir Macpherson 2002). The economic environment Supply and demand characteristics are not alone in influencing the dif- fusion of ISO 9000. The economic environment in which firms operate plays a significant role in creating quality awareness and shaping incen- tives for certification. Government regulations. Regulations can act as significant incentives for certification, but they can also be a deterrent if they increase business risk. The ISO 9000 standard's rapid adoption in Europe was partly the result of harmonized regulations from the 1992 European Community Single Mar- ket Initiative (Grajek 2004; Guler, Guillén, and Muir Macpherson 2002; Mendel 2000; Anderson, Daly, and Johnson 1999). In these trade directives, the EU mandated minimum quality requirements of products and produc- tion processes affecting public safety. In most cases, ISO 9000 registration was accepted as a valid conformity assessment procedure. These legal rules turned ISO 9000 into a marketing instrument for suppliers, who themselves imposed certification requirements down their supply chains. However, gov- ernment regulations can also hamper certification.An excessive or unpredict- able regulatory burden can be expected to discourage the types of long-term investments needed to implement a quality management system. 120 Quality Systems and Standards for a Competitive Edge Government support programs. Governments can also promote the dif- fusion of certification through awareness and support programs. These may include quality awareness campaigns, training, technical assistance, and financial assistance programs for the implementation of quality man- agement systems. Governments have mounted major national campaigns for ISO 9000 certification in many developed and high-growth econo- mies such as the United Kingdom, the EU, Japan, and countries in the Association of Southeast Asian Nations (Mendel 2000). Prior certifications. The number of past ISO 9000 adoptions has an important impact on the rate of certification. A large domestic base of certified organizations is likely to create a high demand for certification because certified organizations often require their own suppliers to be- come certified as well. A large base of certified organizations also re- duces implementation costs by enhancing the supply of information on ISO 9000 and increasing the number of complementary services such as training programs, quality consultants, auditors, and certification bodies. In fact, Neumayer and Perkins (2005) include past certificates as a lagged dependent variable in their empirical study. They find a bandwagon ef- fect through which certification in one year positively affects certifica- tion in the following year. Education. Education levels are likely to affect the cost of implement- ing a quality management system. Employees need solid basic education skills to benefit from the training programs necessary to implement ISO 9000. A more educated workforce will be able to implement the require- ments for ISO 9000 more quickly and effectively, and they may also have more prior knowledge about organizational practices. Again, Neumayer and Perkins's empirical work shows a statistically significant relationship between secondary school enrollment rates and ISO 9000 diffusion. The Indian software sector provides a good example of the drive by firms to adopt ISO 9000 quality cetification and other standards (box 6.2). In particular, the sector has shown a clear understanding of the need to develop skilled human capital. The ISO 9000 Quality Management Standards 121 Box 6.2 Quality Adoption in the Indian Software Sector: Moving Up the Value Chain The leading Indian rms have moved up the value chain in software services, developing organizational and managerial capabilities that enable them to o er more comprehensive services than merely low-cost programming. One sign of maturity is that the industry increasingly procures xed-price contracts, rather than the time-and-materials contracts of earlier years. With the greater risk of xed-price contracts comes exibility in organizing work, greater management control, and an opportunity to earn higher returns as e ciency improves. Revenue per worker is increased, indicating a move up the value chain-- from an average of $9,000 in scal 1995/96 to $20,500 in 2000/2001--but rev- enues are still lower than what they are in product-based companies. In order to build client value, companies have expanded their capacity to service a wider range of software-development tasks, as well as to move into new services, such as product design and information services outsourcing. Software development includes analysis and speci cation of requirements, software design, writing and testing of software, and delivery and installation. Indian companies are trying to move beyond only writing and testing, which require the least skill and account for only a small portion of the overall project costs, to higher skill levels that require deeper business knowledge of the indus- try for which software solutions are being developed. In their quest to climb the value chain, India's software rms ensured product quality and reliability by adopting internationally recognized standardized work processes. An increasing number of rms have met international certi cation requirements for key quality standards. For many, this was an exercise in brand building, but the processes and procedures put in place left their hallmark on the quality of software products and services. Firms seek certi cation from various sources, beginning with quality man- agement practices that meet ISO 9000 standards to ensure consistent and orderly execution of orders. The next stage focuses on software engineering and certi cation under the People Capability Maturity Model (CMM) framework of the Software Engineering Institute (SEI) at increasing levels of process matu- rity. Another stage focuses on aligning internal practices with the CMM, which is a framework to guide attracting, motivating, and retaining a talented technical sta .The Six Sigma methodology ensures end-to-end quality across all company continued on the next page 122 Quality Systems and Standards for a Competitive Edge Box 6.2 Quality Adoption in the Indian Software Sector: Moving Up the Value Chain--continued operations and focuses on improved customer satisfaction by reducing defects, with a target of virtually defect-free processes and products. As of December 2003, India had 65 companies at SEI CMM Maturity Level 5. In October 2002, the SEI of Carnegie Mellon University published a list of high-maturity organizations as part of its Survey of High Maturity Organizations and High Maturity Work- shop research. The full set of 146 high-maturity organizations includes 72 Level 4 organizations and 74 Level 5 organizations. Of the 87 high-maturity organiza- tions assessed outside the United States, 77 are in India. Because most Indian software rms are export-oriented and serve clients around the world, meeting globally acceptable frameworks and standards has been critical to validating their credentials to new clients, who often demand that vendors adopt ISO and CMM standards. The reasons for the success of the quality improvements can be grouped in three categories: people-based, business-related, and management-related (Jalote 2001). The Indian software industry primarily delivers services, which have embraced globally software-process improvement (SPI) more than those who deliver products. As Indian companies serve worldwide clients who demand that their vendors adopt standards such as ISO and CMM, companies were motivated to certify their credentials and used these frameworks to also deliver real software-process improvement. As companies moved to an o shore model, SPI became a necessity to succeed. Managing subcontracted work typi- cally requires monitoring structures to contain risk. This imposes a degree of formality at the interface between the users and developers--something that is generally hard to achieve with in-house development. For most organizations, software development is their core competency that must be continually improved.Their high-growth trajectory required the infusion of a large number of new engineers every year.Without tightly controlled processes, it would have been impossible to absorb new recruits into the development process quickly. Since the cost of manpower was not very high in India, it was possible for most companies to dedicate a team for its SPI e ort. A survey of high-maturity orga- nizations in India indicated that most companies had dedicated manpower for SPI equal to about 1 to 2 percent of their engineering manpower (Jalote 2001). Most of the software companies in India are very young. Being followers in the software-development process, they could exploit the collective knowledge and The ISO 9000 Quality Management Standards 123 experience of organizations the world over in implementing SPI. Most companies introduced quality systems very soon after they were formed.This ensured that the company had work standards to which each new entrant had to conform. After that, the company, people, and quality systems all matured together. As the people inthecompanyhavecontributedactivelytotheSPImovementsfromtheearlydays, it induced among the practitioners a sense of ownership for the quality system. Software companies attract the best talent from engineering schools. Some of the CMM lead assessors have observed that the scores on the Myers-Briggs personality tests conducted as part of the capability appraisals often indicate that Indian engineers are di erent from their counterparts in the United States. Indian employees are ambitious and look for improvement in the way the orga- nization works, which creates a need for process orientation. The average age of the Indian engineers is in the 20s and that of managers is late 20s to early 30s. Younger professionals are more receptive to change, as they have not invested in traditions and, indeed, want changes. Indian culture is more family-oriented rather than individualistic. This prompts people to conform to established frameworks and systems. Profes- sionals do not mind being measured. There are fewer privacy concerns, and in- house surveys have indicated that most engineers are more concerned about the nature of work and the overall work environment and not so much about being measured. The software background of top managers helps to secure backing from senior management for SPI initiatives. Most of the facilitating factors are based in more general and societal con- texts. Such factors are hard to emulate once the context changes. The govern- ment had little role to play in this movement. India does not have centers along the lines of the U.S. or European software engineering institutes. The Ministry of Information Technology in India did bring in the world's best Software Test- ing and Assessment of Software Maturity through licensing arrangements with the Software Engineering Institute at Carnegie Mellon University. Under this agreement, the Indian Standardization, Testing, and Quality Certi cation (STQC) Directorate of the Ministry of Information Technology undertook the job of cer- ti cation, testing, and training of trainers and assessors in India. The increasing importance of outsourced information technology (IT) ser- vices from developed countries prompted many clients to voice concerns about data protection practices of service providers. Issues of data con dential- ity, integrity, and availability have come to the fore.The latest EU data protection laws are designed to ensure that personal data of EU citizens are not sent to a country that has less stringent legal protection. continued on the next page 124 Quality Systems and Standards for a Competitive Edge Box 6.2 Quality Adoption in the Indian Software Sector: Moving Up the Value Chain--continued Clients are also demanding adherence to security standards to ensure infor- mation security. The government of India and NASSCOM [National Association of Software and Service Dompanies] are working closely to respond to these concerns. The government introduced clauses in its IT Act of 2000 covering privacy, digital signatures, and cyber crime to meet EU requirements. More generally, the gov- ernment strengthened software testing and assessment capabilities in India, in association with some of the leading organizations internationally. The Ministry of Information Technology set up the STQC directorate to train assessors and implement security standards. An Information Security Technology Develop- ment Council has also been set up to promote research in the area of informa- tion security. Research and development To secure ultimate success the e orts on quality and standards need to be com- plemented with appropriate R&D expenditures. Those expenditure amounts have been and continue to be small, with some increase in recent years. Low R&D expenditures can partly be explained by the service (instead of product) focus, which would require greater investment in R&D. The bulk of R&D occurs in subsidiaries set up by multinationals. As a cost reduction strategy, a number of large and medium product compa- nies started captive development centers in India. Other companies have part- nered with Indian rms to set up product development centers, and still others are outsourcing to India functions such as requirement speci cation, design, testing, and maintenance. The availability of capable and low-cost Indian tech- nical expertise, coupled with the deep nancial resources of the multinationals, provides for cost-e ective R&D. However, as multinationals increasingly distrib- ute R&D or product development operations globally, they become less likely to develop whole products in a single place such as India. Indian rms were reluctant to invest in product development because they lacked resources and expertise and, more importantly, because of the di cul- ties in designing products for distant and unfamiliar markets. Even when rms have the resources, they nd it hard to justify the high risks of product develop- The ISO 9000 Quality Management Standards 125 ment. The risks are much lower in providing services than in selling product, in part because of the lower level of skill and nancial risk. In the software industry, product development is a small component of the overall costs of developing and promoting software products. Software rms may spend as much as 50 percent of revenues on advertising and marketing and as little as 10 to 15 percent on product development. There are few examples of successful product development by Indian soft- ware companies. For the industry as a whole, only 1 to 5 percent of the software packages typically succeed in the market. Only recently have Indian companies reached a size and maturity to consider investing in R&D and marketing. One successful example to date consists of products developed by Indian compa- nies for the banking sector. Indian software companies as learning organizations Tschang, Amsden, and Sadagopan (2001) examined the di erent ways in which upgrading takes place in the Indian software industry. They used the R&D classi cation of pure, basic, and applied research to dif- ferentiate di erent rms' technological abilities and functions. They found suf- cient evidence of rms upgrading to the applied-research stage--itself an achievement, since it involves more conceptual work or longer-term e orts at research.The two highest levels of research--pure science and basic research-- are almost nonexistent in India, especially in domestic rms. This shows that the nature of the industry is "applied,""service-oriented," or "incremental" in its innovation. The technologies developed are typically not breakthroughs, but are rather rst implementations, involving"transforming, variating, and reapply- ing"known techniques to the software product under design. The model of upgrading into products is perhaps the most di cult task, given the many reasons for failure. There is a paucity of success stories. The lead- ers of domestic rms who went into products, including those who left Wipro, Satyam, and other service companies to build their new enterprises, all noted that they had a di erent mentality and business objective in mind. Their goal was to build products or to create a fundamentally new service. Their plans would not have materialized if they remained in their former software service rms. Each new start-up may have also been trying to nd a defensible or com- petitive niche within the Indian software industry. Many rms are discouraged from trying the product market because of the distance from the nal market and lack of su cient resources or expertise at continued on the next page 126 Quality Systems and Standards for a Competitive Edge Box 6.2 Quality Adoption in the Indian Software Sector: Moving Up the Value Chain--continued the outset. Service companies have resources, but nd it hard to justify risk tak- ing when they have such nice returns. Ultimately, even rms that break into the product market can stumble and fall. Ramco was an example. The shortage of this risk-taking attitude across the broader industry, coupled with resource and distance-from-market constraints, will make it di cult for more product rms to emerge. Unless all these factors change, it is unlikely that the Indian industry as a whole will change its complexion to one with more diverse models of upgrading. The Indian service companies are clearly following a trajectory laid out by their original competencies, continuing to dominate the larger part of the services value chain all the way back to the requirements analysis and consulting stages. The emergence of the specialized services model, such as Mindtree's con- tract R&D service, shows that the Indian industry does have its own style of promising entrepreneurial capabilities and the resourcefulness to develop a wider variety of areas. Ultimately, the running of multinational subsidiaries on a cost center basis (and the continued tradition of de ning product requirements elsewhere) will constrain these subsidiaries from promoting new ideas or products locally. This pattern is dif- ferent from domestic rms, which run as pro t centers, giving them both heavier responsibility and greater scope for doing challenging work. In summary, both domestic rmsandmultinationalsappeartobeabletoupgradetoappliedresearch, but the business models themselves suggest that the domestic rms have broader scope to do products, if they so choose.Those local rms may engage in a broader range of R&D, though not necessarily the most advanced technologies. The Indian product company Sassken built a large R&D arm to research tech- nologies for making the" rst implementation"of a communication standard (high- level and detailed design). That e ort can be considered to involve both (a)"learn- ing"in applied research and (b) implementation in models of applied research. The same kind of learning and concept modeling in applied research has been done at the CMC, one of India's earliest software companies set up by the national government. CMC had to design systems from scratch, many of which had social objectives, such as India's rst railway reservations system, perhaps the most complicated systems endeavor ever undertaken in India. Source: Excerpted from Bhatnagar 2006. C H A P T E R 7 Standardization in Latin America The issues described in the first half of this book are relevant to all devel- oping countries. They are particularly essential, however, for middle- income countries and for countries that have chosen exports as a key strategy for economic growth. Countries are taking a number of mea- sures to improve their national quality systems, but most often these measures are not systemic and coherent enough to provide the desired result. Rather, they tend to produce fragmented and incomplete systems. This is the case for most countries in Latin America and the Caribbean (LAC). The second half of this book focuses on that region to illustrate what developing countries typically are doing in this area and the issues and problems of the existing systems. The region was chosen for its abun- dance of middle-income countries, most of which are choosing export promotion as a key engine for growth. We begin with an assessment of the national standardization systems in Latin American countries, with a particular focus on industrial volun- tary technical standards. We review the standard-setting institutions in a number of countries in the region, as well as in several benchmark coun- tries outside the region, and compare their structure, role, functions, and resources. We evaluate the performance of these institutions in terms of the number of standards produced, their scope, and their overall potential for diffusion in the economy. While the number of standards may be the most readily available performance indicator, it does not reflect the quality of the standards being adopted.The international integration of a country's standardization system, also covered in this section, provides valuable addi- tional insight on the quality of national standards in Latin America. 127 128 Quality Systems and Standards for a Competitive Edge The discussion in these chapters draws on a survey of national con- formity assessment institutions conducted by the authors in a number of mostly Latin American countries (hereafter referred to as authors' research). Questionnaires were sent to national standarization, accredi- tation, and metrology bodies in Argentina, Brazil, Colombia, Ecuador, Mexico, Peru, and Turkey. The aim was to obtain direct information and a comprehensive understanding of institutional characteristics, quality, and performance by focusing on issues such as scope, governance, autonomy, personnel, finances, activities, facilities, international integration, and rec- ognition, and adherence to best practices. This information was comple- mented by legislative and regulatory research and online information available from national, regional, and international conformity assesment organizations. The Standards-Setting Institutions The organization of standards-setting There is no dominant model for the organization of standardization activities in Latin America. Countries have adopted and combined dif- ferent types of systems, all more or less attached to the public sector with various degrees of decentralization (table 7.1). Where the primary standards bodies are part of the public sector, they usually operate as autonomous agencies. In the cases of Ecuador and Peru, the standards body is attached to a government ministry but operates autonomously. Table 7.1 Characteristics of the National Voluntary Standards Bodies Primary voluntary Degree of centralization Country standards body Legal status (voluntary standards) Argentina IRAM private centralized Brazil ABNT private centralized Chile INN private hybrid Colombia ICONTEC private hybrid Ecuador INEN public centralized Mexico DGN combination decentralized Peru INDECOPI public centralized Korea, Rep. of KATS public hybrid Spain AENOR private centralized United Kingdom BSI British Standards private centralized Source: Authors'research. Standardization in Latin America 129 In Chile, the National Standardization Institute (INN) is a private-law organization but was originally established by CORFO, an autonomous economic development agency attached to the Ministry of Economics. At the other end of the spectrum, the standards bodies in Argentina, Brazil, and Colombia originate in the private sector. They operate as pri- vate nonprofit organizations but are officially recognized as national stan- dards bodies through government decrees or resolutions. In return for official recognition, these bodies must adhere to legislation on national standardization. There is no systematic preference for a centralized or decentralized system in Latin America. In Mexico, the primary national standards body, the General Bureau of Standards (DGN), is deeply integrated in the national government and operates as a department of the central government's Ministry of the Economy. However, Mexico operates on a highly decentralized model of standardization and the DGN is mainly concerned with mandatory standards, developing voluntary standards in exceptional circumstances only. Voluntary standards are developed by private sector organizations registered with the Ministry of the Economy. In contrast, Argentina and Brazil have adopted highly centralized systems much like those of their European counterparts. Chile and Colombia have adopted a hybrid version of standards-setting, with the responsibili- ties for standards development distributed between a primary national body and sectoral organizations. Standards bodies in Latin America are usually involved in a host of activities. None confine themselves to developing voluntary standards (table 7.2). In some countries, the national standards body also develops mandatory standards. Some Latin American countries have a single pub- lic institution performing national standardization along with metrology or accreditation functions or both. Integrating metrology and accredita- tion with national standardization is not the norm in most developed countries but often occurs in small countries or countries where the stan- dards infrastructure is not mature.1 However, many of the standards bod- ies in Latin America perform certification, testing, or calibration tasks, which is also the case in many standards bodies throughout the world. In Mexico, where the standardization system is very decentralized, the primary standards body, the DGN, focuses on mandatory standards; a number of independent standards bodies develop voluntary standards while also participating in testing, certification, and training activities. Most Latin American countries also offer training services in the area of voluntary standards. 130 Quality Systems and Standards for a Competitive Edge Table 7.2 Activities of Standards Bodies Mandatory National Calibration Country standards Accreditation metrology and testing Certification Training Argentina no no no no yes yes Brazil no no no no yes yes Chile yes yes yes no no yes Colombia no no no yes yes yes Ecuador yes no yes yes yes yes Mexico: yes no no no no no DGN Mexico: no no no yes yes yes others Peru somea yes yes yes no no Korea, no yes yes yes yes no Rep. of Spain no no no yes yes yes United no no no no yes yes Kingdom Source: Authors'research. Note: a. Only legal metrology. Table 7.3 presents the share of personnel in standards bodies involved in developing standards. This proportion is low in some countries, such as Colombia, Ecuador, and Peru, suggesting an important role for com- plementary activities in these organizations. Chile is an exception, with 75 percent of INN's personnel involved in standardization activities. Table 7.3 Proportion of National Standards Body Staff Involved in Standardization Activities percent Country Proportion of staff involved in standardization activities Brazil 30 Chile 75 Colombia 14 Ecuador 15 Peru 2 Sources: Authors'research; INN Web site. Note: Uses 2004 data for Brazil, Colombia, Ecuador, and Peru; 2002 data for Chile. Standardization in Latin America 131 Governance and autonomy All of the private bodies and some of the autonomous public bodies are governed by executive councils elected through a general assembly of members (table 7.4). In some cases, a number of seats on the executive council are reserved for government appointees. In the cases of Chile and Peru, there is no general assembly and all council members are nominated by the central government.While Peru has a consultative council to repre- sent the interests of the public, private sector representation is limited to technical committees in Chile. Peru is the only country in Latin America where the executive director is directly appointed by the government. The standards bodies operate with a substantial level of technical autonomy from the government. In general, the bodies have full author- ity over the standardization process, they are free to formulate their own budget, they are not affected by government staffing policies, they may diversify their services (as long as it does not affect their ability to act as a standards body), and they may engage in international negotiations. There are some exceptions, such as in Ecuador, where salaries are subject to public sector restrictions. Regardless of the level of autonomy of the organization, each country has laws and decrees articulating the obliga- tions or activities of its standards bodies. Table 7.4 Private Sector Participation in the Governance of Standards Bodies Share of executive council Executive director General Consultative members appointed by appointed by Country assembly council government (%) government Argentina yes no 0 no Brazil yes no 10 no Chile no no 100 no Colombia yes yes 33 no Ecuador yes yes 71 no Mexicoa yes yes 0 no Peru no yes 100 yes Korea, Rep. of no yes 100 yes Spain yes no 0 no United yes yes 0 no Kingdom Sources: Authors'research; Web sites of the national standards bodies. Note: a. In the case of Mexico, information is given for the private standards bodies (organismos nacionales de normalización). 132 Quality Systems and Standards for a Competitive Edge The standards development process in most Latin American countries is conducted according to international norms. The general public may propose standards development projects to the standards body. Standards drafts are then developed by technical committees through a consensus process. In most cases, standards projects and drafts are approved by com- mittees of experts representing different societal and economic interests. Colombia and Ecuador differ from this practice in that the executive council of the standards body needs to approve each standard. In Chile, the standards must be approved by a relevant government ministry. Resources for standardization Although there is significant variability in the main sources of funding for Latin American standards bodies, most of them depend only minimally on public transfers (figure 7.1).2 In Latin America the public funding devoted to support for standards bodies is practically nonexistent. The other two activities more directly related to standardization, publication sales and membership fees, do not play an important role either. As is typical throughout the world, standards publications account for only a small fraction of the revenues of the standards bodies surveyed, except in Brazil, where they constitute 50 percent of the revenues of the Brazilian Technical Standards Association (ABNT). There is great variability in the extent to which the standards bodies in Latin America generate income through membership fees. In Chile, INN is not a membership organiza- tion and thus derives no revenues from membership fees. In Colombia, membership fees account for a small portion of total revenues, as in Spain and the United Kingdom. In contrast, in Argentina, Brazil, and especially in Peru, membership fees account for a significant portion of income. Standards bodies that offer certification services depend on them for a large share of revenues; this is true of the Colombian Institute for Techni- cal Standards and Certification (ICONTEC) and the Argentine Institute of Standardization and Certification (IRAM). The case of Mexico is more complex because it is essentially a decentralized system. The primary standards body is integrated in the Ministry of the Economy and depends solely on public funding. The private national standards bodies (organis- mos nacionales de normalización) do not receive public funding for stan- dardization purposes. Their funding comes from membership fees and services. Although it is difficult to evaluate the total resources for standards development based solely on the budgets of the primary standards bod- ies, Latin American countries appear to dedicate relatively few funds Standardization in Latin America 133 Figure 7.1 Funding Sources of the National Standards Bodies Colombia (2004) Brazil (2004) Ecuador (2002) standards membership standards publications fees other publications 5% 5% 5% 11% certification other 24% 42% other certification 40% standards certification membership publications 43% 50% fees public 50% 21% transfers 4% Argentina (2002) United Kingdom (2002) Spain (2002) standards publications membership standards 4% fees publications other 16% 15% certification 7% 78% membership fees other 2% 48% certification membership certification 33% fees 73% other public public 5% 2% transfers transfers 2% 5% standards publications 10% Sources: ISO 2003a; authors'research. to this activity. Figure 7.2 shows the budgets of the different standards bodies. The budgets for the United Kingdom (not shown), Spain, and Korea are several times larger than those of Latin American countries. Because most standards affect manufactured goods, standardizing insti- tutional resources by manufacturing value added facilitates cross-country comparisons. In terms of standardized budgets, Colombia and Ecuador dramatically outperform Spain and Korea, while other Latin American countries lag behind. However, the budget of the standards body is not necessarily a good reflection of the resources for standardization activities because an important part of the budget may be dedicated to complementary activi- ties such as certification and testing. Furthermore, in decentralized stan- dardization systems, organizations other than the primary standards body 134 Quality Systems and Standards for a Competitive Edge Figure 7.2 Budgets of the National Standards Bodies 60 900 budget (millions of 2003 US$) budget per US$1 million of 800 50 manufacturing value added, turing (2003) 700 US$) 40 600 manufac 2003 of of 2003, 500 30 400 million added (millions 20 300 US$1 value per budget 200 10 100 budget 0 0 , , , , Spain,2002 . of2002 2004 2002 2004Chile2002 Peru2004 Rep Colombia, Brazil 2004gentina, Ecuador ea, Ar Kor Sources: ISO 2003a; authors'research. may be providing many of the resources for national standards devel- opment. Finally, standardizing the budget by the manufacturing value added allows for a rather imperfect comparison of resources. Even if a country performs little manufacturing activity, its population will still consume manufactured goods and there will still be a need for a certain number of standards. Likewise, it is difficult to compare national resources devoted to stan- dardization by comparing the number of staff in the national standards bodies because some staff members may be involved in activities other than standardization. In some countries, the decentralization of standards setting also involves the use of staff in other organizations. Table 7.5 pro- vides an indication of the resources used in developing national standards by examining staff in the entire standards body and the proportion of staff members dedicated uniquely to standardization activities. In this table, a comparison of the standardization staff, relative to manufacturing value added, brings out the relatively low number of personnel involved in standardization activities in Brazil and Peru. In both of these countries, the development of standards is centralized and one would expect a large number of staff in the national body rather than a low number. Standardization in Latin America 135 Table 7.5 Staff of the National Standards Bodies Total Number of staff involved Number of staff involved in permanent in standardization standardization activities, relative to Country staff activities 2003 manufacturing value added Argentina 170 n.a. n.a. Brazil 70 21 0.39 Chile 53 40 3.45 Colombia 200 28 2.54 Ecuador 89 13 4.46 Peru 149 4 0.36 Korea, 244 n.a. n.a. Rep. of Spain 469 n.a. n.a. United 5,788 n.a. n.a. Kingdom Sources: ISO 2003a; authors'research. Note: 2004 data for Brazil, Colombia, Ecuador, Peru, and the United Kingdom; 2003 data for Spain; 2002 data for Argentina, Chile, and Korea. n.a. = not available. Mandatory standards and technical regulations The share of mandatory standards produced by the national standards bod- ies has decreased in some countries and increased in others.Table 7.6 shows the shares of mandatory standards in the total number of standards devel- oped by the standards bodies in 1990 and 2002. In Brazil and Colombia, the national standards body does not develop mandatory standards. This does not imply that they have no national mandatory standards, but simply that these are the responsibility of a separate governmental entity. In Peru, the only mandatory standards published by the national standards body relate to legal metrology. Although Argentina has a larger share of manda- tory standards than Brazil, Colombia, or Peru, the Argentine standards body, IRAM, does not have a mandate to publish mandatory standards. Rather, the government sometimes incorporates its voluntary standards in national law, effectively rendering them mandatory. Mexico has an even larger pro- portion of mandatory standards, which are the responsibility of the primary standards body, the DGN. In Ecuador and Chile, the proportion of manda- tory standards is quite high. However, while their share has been reduced drastically in Ecuador, from 95 percent in 1990 to 25 percent in 2002, the change has occurred in the opposite direction in Chile, where 40 percent of standards are now mandatory, which is high by international standards. 136 Quality Systems and Standards for a Competitive Edge Table 7.6 Share of Mandatory Standards in Total National Standards, 1990 and 2002 percent Country 1990 2002 Argentina 5 9 Brazil 7 0 Chile 30 40 Colombia 8 0 Ecuador 95 25 Mexico 4 15 Peru 1 1 Korea, Rep. of 0 0 Spain n.a. 20 United Kingdom n.a. 0 Sources: Stephenson 1997; ISO 2003a. Note: n.a. = not available. The Performance of Standards Bodies Standards development activities Most of the surveyed Latin American countries have increased their stock of standards since 1991, but not at a rate comparable to developed countries. As can be seen in figure 7.3, Argentina and Brazil had nearly the same number of standards as Korea in 1990 but were lagging far behind about a decade later. While the stock of standards nearly doubled in Korea between 1990 and 2002, and more than doubled in the United Kingdom between 1990 and 2003, most Latin American countries have seen their standards stocks increase by 16 percent or less during the 1990­2002 period (table 7.7). Colombia and Chile have had larger rela- tive increases than their counterparts, although Chile started out with a very low number of standards in 1991. However, the net growth rate of the standards stock may hide much of the national standardization activity given that old standards may be eliminated while new standards are being created. If many of a country's standards are obsolete and are being eliminated at a high rate, then a low growth of the standards stock during a limited period does not necessarily imply that few standards are being developed. Latin American countries lag far behind Korea, the United Kingdom, and Spain in terms of the number of new standards adopted in 2004. Standardization in Latin America 137 Figure 7.3 Number of National Standards, 1990 and 2002 16,000 15,176 number of national standards in 2002 14,000 number of national standards in 1990 ds 12,000 standar 10,000 9,271 8,000 7,710 national of 6,000 5,570 5,000 3,800 4,000 number 2,583 2,318 2,000 0 Korea, Brazil Argentina Mexico Colombia Peru Chile Ecuador Rep. of Sources: Authors'research; Web sites of the national standards bodies. Table 7.7 Growth of the Standards Stock between 1990 and 2002 percent United Korea, Kingdoma Rep. of Brazil Argentina Mexico Colombia Peru Chile Ecuador 124 79 16 ­2 1 43 15 29 16 Sources: Stephenson 1997; ISO 2004; DTI 2005. Note: a. 1990­2003. Of those three countries, Spain developed the fewest standards in 2004, but still about four times more than Colombia, the best Latin American performer (figure 7.4). Surprisingly, Brazil, the largest economy in Latin America, adopted the fewest new standards in 2004. Latin American countries have far fewer technical committees than the United Kingdom and Korea (figure 7.4). Argentina leads the Latin American group with more committees than Spain, followed by Colom- bia. The number of technical committees represents the number of technical areas in which standards are being developed. Generally, tech- nical committees can develop many standards simultaneously, so more standards per committee may not necessarily be a measure of efficiency but may simply imply that there are more members in the committee. 138 Quality Systems and Standards for a Competitive Edge Figure 7.4 Number of New Standards, 2004, and Active Technical Committees, 2005 3,500 1,600 number of new voluntary 3,000 standards adopted in 2004 1,400 number of active technical committees in 2005 1,200 ees 2,500 ds 1,000 ommittc 2,000 standar 800 of 1,500 echnicalt 600 of number 1,000 400 number 500 200 0 0 ea, . of ed Peru Kor Chile Rep Unitingdom Spain Brazil Colombia Ar gentina K Source: Web sites of the national standards bodies. The scope of technical activities covered by committees may also vary across countries, rendering comparisons of the number of committees less meaningful. For example, while there is one committee for PVC and another for plastic tubes in Chile, these two areas are covered by a single ISO technical committee at the international level. The numbers of new and existing national standards and technical committees give a general idea of the performance of each standards body but cannot on their own lead to strong conclusions about the effec- tiveness of the standards. Standard statistics do not reflect the relevance of standards to the domestic industrial structure, their impact on tech- nical change, and their impact on trade. Standards can actually impede growth if they are ill conceived. Hence, standardization statistics should be interpreted carefully. Di usion of standards The price of written standards documentation in Latin American coun- tries is relatively high. Although in most countries the price of a sin- gle standard would be relatively low for a small or medium enterprise, marginal increases in prices may become prohibitive when a business requires a large array of standards to compete. Figure 7.5 displays the cost Standardization in Latin America 139 Figure 7.5 Price of National Standards Based on ISO 9001:2000 140 0.9 0.8 120 0.7 100 0.6 80 capita 0.5 per US$ 60 0.4 GDP of 0.3 % 40 0.2 20 0.1 0 0 ed o ISO ru Chile Pe Unitingdom Spain Brazil Mexic gentina K Ecuador Ar Colombia price of national standard based on ISO 9001 (US$) price of national standard based on ISO 9001, as a share of GDP per capita Source: Web sites of the national standards bodies. of purchasing a national standard based on ISO 9001:2000 in different countries and from the International Organization for Standardization (ISO) itself. The ISO 9001:2000 standard is used as a proxy for the price of other national standards. The figure shows that while the average price of standards is lower in most Latin American countries than in developed countries such as Spain or the United Kingdom, this no longer holds true when prices are adjusted to reflect differences in per capita income. In this case, only national standards sold in Mexico remain competitive. International Integration Adoption of international standards Most of the surveyed Latin American countries are full members of the organizations responsible for the development of most international stan- dards (table 7.8). Peru is an exception. It is not a member of the Interna- tional Electrotechnical Commission (IEC) and,as a correspondent member of ISO, it cannot take part in the development of ISO standards. 140 Quality Systems and Standards for a Competitive Edge Table 7.8 Membership in International Standards Bodies Country ISO ITU IEC CODEX Argentina yes yes yes yes Brazil yes yes yes yes Chile yes yes no yes Colombia yes yes associate yes Ecuador yes yes no yes Mexico yes yes yes yes Peru correspondent yes no yes Korea, Rep. of yes yes yes yes Spain yes yes yes yes United Kingdom yes yes yes yes Sources: ISO, ITU, IEC, and CODEX Web sites. Note: CODEX = Codex Alimentarius Commission; ITU = International Telecommunication Union. Although progress is being made in some Latin American countries, others have incorporated very few international standards in their national standards. The evolution of international standard adoption rates shown in table 7.9 brings out the major differences in adoption rates across countries. Argentina and Ecuador have barely increased their extremely low share of international standards, while Brazil and Peru have seen dramatic increases. Colombia, the Latin American country with the most Table 7.9 Evolution of the Share of National Standards Based on International Standards percent Country 2002 2004 Argentina 1 1 Brazil 4 30 Chile 25 n.a. Colombia 27 35 Ecuador 1 2 Peru 5 22 Korea, Rep. of 46 n.a. Spain 18 n.a. United Kingdom 45 n.a. Sources: ISO 2003a; authors'research. Note: n.a. = not available. Standardization in Latin America 141 international standards, was still far behind the United Kingdom and Korea in 2002. Participation in international standards development activities Latin American countries are not very active in the development of international standards. Brazil and Mexico, each with a larger economy than Spain's or Korea's, participate in far fewer ISO technical commit- tees than Spain or Korea. When Latin American countries are members of ISO committees, it is mostly as observers rather than as participating members (figure 7.6). In particular, Argentina, Mexico, and Brazil are observer members of roughly 35 percent of the 734 ISO technical com- mittees and subcommittees.3 Few international standards originate in Latin American countries. In the region, Brazil and Colombia take the lead in developing new inter- national standards, as suggested by their higher numbers of technical committee secretariats and working group governorships (figure 7.7). However, they are for the most part leading working groups, which have responsibility for fewer standards than technical committees or subcom- mittees. Brazil, with the largest number of secretariats in Latin America, Figure 7.6 Number of Participant and Observer Memberships in ISO Technical Committees and Subcommittees, 2005 700 participant observer 600 500 400 memberships of 300 number200 100 0 o Peru ea,. of ed Chile Brazil Spain gentina Kor Ecuador Mexic Colombia Ar Rep Unitingdom K Source: ISO Web site. 142 Quality Systems and Standards for a Competitive Edge Figure 7.7 Participation in ISO Standards Development, 2004 14 number of technical committee secretariats number of working group governorships 12 ernorshipsv 10 go or 8 etariats 6 secr of 4 2 number 0 o ea,. of Mexic Peru Chile Brazil gentina Kor Spain Ecuador Rep Ar Colombia Source: ISO Web site. has only two, as compared to 10 for Spain and 99 for the United King- dom. Apart from Brazil, Colombia, and Mexico, other Latin American countries in the survey do not have a single governorship or secretariat. Regional coordination as a standards upgrading strategy Latin American countries participate in a number of overlapping regional standards coordination schemes involving harmonization and compatibility strategies. The overwhelming majority of these standards- related measures originate in preferential or regional trade agreements (table 7.10). Most of the trade agreements, especially the bilateral agree- ments or extraregional agreements, refer to the provisions of the World Trade Organiation Technical Barriers to Trade (WTO TBT) Agreement and promote standards compatibility by adopting international standards where possible. The five trade agreements and organizations with the most ambitious mandates for standards upgrading are listed in table 7.11 and described below. APEC. APEC's main approach is to align the standards of member econo- mies with international standards in a number of priority areas. Four orig- inal priority objectives were designated in 1996.4 Developed economies Standardization in Latin America 143 Table 7.10 Membership in Organizations and Preferential Trade Agreements with Standards-Related Measures Country Regional or preferential trade agreements Total Argentina Mercosur; Chile-Mercosur; COPANT 3 Brazil Mercosur; Chile-Mercosur; COPANT 3 Chile Chile-Mercosur; APEC; Chile-Mexico; Central America-Chile; 8 Chile-EU; Chile-Korea; Chile-U.S.; COPANT Colombia Andean Community; Group of 3; Colombia-CARICOM; COPANT 4 Ecuador Andean Community; APEC; COPANT 3 Mexico NAFTA; APEC; Group of 3; Bolivia-Mexico; Chile-Mexico; Costa 13 Rica­Mexico; Mexico-Nicaragua; Mexico­Northern Triangle; Mexico-EU; Mexico-Israel; Mexico-Nicaragua; Mexico-Uruguay; COPANT Peru Andean Community; COPANT 2 Sources: Aldaz-Carroll 2006; OAS Foreign Trade Integration System (SICE) Web site. Table 7.11 Principal Regional Agreements and Organizations with Standards-Related Measures Agreement/ Standards organization Countries Description policy objective APEC Australia, Brunei Darussalam, Intergovernmental Harmonization (Asia-Paci c Canada, Chile, China, Hong Kong organization Economic (China), Indonesia, Japan, Rep. and free trade Cooperation) of Korea, Malaysia, Mexico, New agreement Zealand, Papua New Guinea, Peru, Philippines, Russian Federation, Singapore, Taiwan (China), Thailand, United States, Vietnam COPANT Argentina, Barbados, Bolivia, Brazil, Private nonpro t Harmonization (Pan American Canada, Chile, Colombia, Costa association of Standards Rica, Cuba, Dominica, Dominican national standards Commission) Republic, Ecuador, El Salvador, bodies Grenada, Guatemala, Guyana, Honduras, Jamaica, Mexico, Nicaragua, Panama, Paraguay, Peru, Saint Lucia, Trinidad and Tobago, United States, Uruguay, Venezuela Andean Community Colombia, Bolivia, Ecuador, Peru, Customs union Harmonization Venezuela Mercosur (Southern Argentina, Brazil, Paraguay, Uruguay Customs union Harmonization Common Market) NAFTA (North Canada, Mexico, United States Free trade Compatibility American Free Trade agreement Agreement) 144 Quality Systems and Standards for a Competitive Edge were given until 2000 to harmonize their standards in these areas, and developing member economies until 2005 (APEC 2004a). Since then, three additional priority objectives and alignment timetables have been defined. Not all APEC countries have shown interest in voluntary har- monization, and in 2004 only 15 members submitted progress reports. Of these 15 economies, 10 had reached their alignment targets, including Peru. Chile had aligned in 76 percent of the original priority areas. Mexico did not submit a progress report. Fourteen of these countries, including Peru and Chile, had achieved alignment in the additional priority areas (APEC 2004b). Despite the successful upgrading of standards in Chile and Peru, the impact of the APEC "voluntary action plan" is limited. In Chile, the priority areas concern a total of 56 national standards out of 2,300, and in Peru they concern as few as 17 standards out of 3,797. Another part of APEC's harmonization strategy is to encourage the participation of member states in the development of international stan- dards by ISO and the IEC, in anticipation of the subsequent alignment of member economies to these standards. Here again, APEC has selected a limited number of priority areas and member participation in the stan- dards development process is entirely voluntary. COPANT. COPANT promotes the harmonization of standards in the Americas with an emphasis on the adoption of international standards, or the development of regional standards when there are no relevant or appropriate international standards. COPANT was created in the 1940s and now comprises 28 standards bodies. As of March 2003, 12 technical committees were responsible for developing "Pan American standards." Their adoption as national standards in member countries is voluntary. Each of the surveyed Latin American countries participated in at least half of the technical committees, except for Ecuador, which was not in- volved in a single technical committee (figure 7.8). COPANT has not been very active in the past few years. Based on the most recent available information, only two Pan American standards were adopted in 2002. The Andean Community. The Andean Community has provided the re- gion with a forum for standards harmonization.Since 1997,its approach has been to adopt regional standards and gradually upgrade them toward more demanding international standards. The Andean Community prioritizes the use of international standards to develop "Andean standards," but when international standards are unavailable or unsuitable for regional needs, there is a preference for regional standards (such as COPANT or European Standardization in Latin America 145 Figure 7.8 Participation in COPANT Standards Development Activities 14 number of technical committee participant memberships number of technical committee secretariats 12 etariats 10 8 6 memberships/secr of 4 number 2 0 Mexico Argentina Peru Colombia Chile Brazil Ecuador Source: COPANT Web site. Committee for Standardization standards) rather than national standards. The Andean standards are developed through an association representing the national standards bodies, the Red Andina de Normalización, which is administered by a regional coordination body (Ente Coordinador). When Andean standards projects are based on international standards they must be approved by at least three countries; otherwise they must be approved unanimously.5 The national standards bodies of member states must adopt the Andean standards as national standards but their application remains voluntary. There are currently 50 Andean standards. Mercosur. Mercosur's approach also involves the gradual upgrading of regional standards. Standards harmonization in this trade bloc is led by the Mercosur's Standardization Committee, which was officially created in 1991 in a common market resolution6 and represents the national standards bodies of the four members. The Mercosur Standardization Committee consists of a directing council of representatives from each standards body, along with 18 sectoral committees focusing on different industries. Brazil and Argentina dominate the sectoral committees, Brazil holding 11 secretariats and Argentina six. The directing council approves regional standards by consensus. Like the Andean Community, Merco- sur prioritizes harmonization with international standards, followed by 146 Quality Systems and Standards for a Competitive Edge regional standards and national standards. There are currently 493 Mer- cosur voluntary standards. NAFTA. The provisions on standards in the 1994 NAFTA treaty closely follow those of the WTO TBT Agreement but place an added emphasis on compatibility. The member economies are encouraged to use interna- tional standards, except when these are inappropriate to fulfill legitimate objectives. Chapter 9 of the treaty states that the three countries should "to the greatest extent practicable, make compatible their respective standard-related measures." A NAFTA Committee on Standards-Related Measures (CSRM) oversees the implementation of chapter 9. The treaty requires the CSRM to establish working groups and subcommittees in four key sectors, and subsequently in other sectors as needed. The North American Trilateral Standardization Forum was formed by the three standards bodies of the member states to work with the government- only CSRM. A particular feature of the NAFTA treaty is that it allows interested parties of member states to participate directly in the develop- ment of new standards on the same basis as domestic firms in the NAFTA countries. Figure 7.9 Number of TBT Notifications on Technical Regulations and Conformity Assessment Procedures, 2004 40 35 34 32 30 30 25 23 notifications T 20 19 TB 16 of 15 10 number 5 4 4 0 0 y o . of Peru European Brazil Chile gentina Mexic Rep Ecuador Ar Colombia ea, Communit Kor Source: WTO Web site. Standardization in Latin America 147 The implementation of WTO TBT and SPS agreements The surveyed countries have complied with the basic obligations of the WTO TBT Agreement, but some countries have submitted few notifica- tions. Each country has established a national enquiry point for standards, conformity assessment, and technical regulations and has submitted noti- fications on technical regulations and conformity assessment procedures (figure 7.9). In 2004, a few countries submitted almost as many notifica- tions as the European Community for the regional standards developed for EU and European Free Trade Association countries, and more than Korea. This either implies that they are fulfilling the terms of the agree- ment or that they are very active in the development of technical regula- tions and conformity assessment procedures.Three countries, Chile, Peru, and Ecuador, submitted very few notifications relative to their regional counterparts. The Latin American countries surveyed have all accepted the TBT Code of Good Practice. C H A P T E R 8 Certi cation in Latin America Certification is a critical area in Latin American countries, but their prog- ress in this regard has been slow and uneven. The institutional structure is not very friendly and financial/cost issues remain. Moreover, a deficient investment climate does not encourage firms to move forward. Efforts by countries have been tepid at best; both greater emphasis on certification and proper incentives are needed. Certification Activities As a region, Latin America registers a low number of ISO 9001 qual- ity management systems certificates, in both relative and absolute terms. Figure 8.1 displays the number of ISO 9000 certificates in Latin Amer- ica and the Caribbean, including earlier versions of the ISO 9000:1994 family (up to 2003) and ISO 9001:2000 (as of 2001). On average, the total number of certificates in Latin America has increased over the past decade, but in 2004 this number represented a smaller share of global certificates than it did in 2001. This may be due to new sector-specific quality management system certificates that have replaced ISO 9000 in the past few years. The dip in the number of ISO 9000 certificates in 2003 can be explained by the deadline for the transition from ISO 9000:1994 to ISO 9000:2000 and the fact that withdrawn certificates were not reported in some countries. In Latin America, the number of ISO 14000 environmental man- agement systems certificates has been steadily increasing since 1994 (figure 8.2). The region accounts for a growing share of the world's total, 149 150 Quality Systems and Standards for a Competitive Edge Figure 8.1 ISO 9000 Certificates in Latin America and the Caribbean, 1993­2004 25,000 3.5 3.0 es 20,000 2.5 tificat 15,000 otalt erc 2.0 orld of w 10,000 1.5 of 1.0 % number 5,000 0.5 0 0.0 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 LAC ISO 9000 certificates as a share of world total (%) LAC ISO 9001:2000 certificates as a share of world total (%) LAC ISO 9000 certificates LAC ISO 9001:2000 certificates Sources: ISO 2001, 2003b, 2005. Figure 8.2 Number of ISO 14000 Certificates in Latin America and the Caribbean, 1995­2004 4,000 4.0 3,500 3.5 es 3,000 3.0 tificat 2,500 2.5 otalt erc 2,000 2.0 orld of w 1,500 1.5 of % 1,000 1.0 number 500 0.5 0 0.0 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 LAC ISO 14000 certificates LAC ISO 14000 certificates as a share of world total (%) Sources: ISO 2005. Certification in Latin America 151 holding 3.8 percent of certificates at the end of 2004. Latin American countries are performing better in relation to ISO 14000 than in relation to ISO 9000. Latin American and Caribbean countries have far fewer ISO 9000 and 14000 certificates than the United Kingdom or Spain, but there are wide differences in certification within the Latin American and Carib- bean region (figure 8.3). At the end of 2004, all countries in Central and South America registered at least one ISO 9001 certificate, and all coun- tries, apart from Belize and Surinam, registered at least one ISO 14001 certificate. However, some countries, like Ecuador and Peru, registered extremely low numbers of management system certificates. In 2004, the growth rate of ISO 9000 certificates in most Latin Ameri- can and Caribbean countries was lower than their expected growth rate, given world averages. The fitted line in figure 8.4 represents average global growth rates from 2001 to 2004 for a given number of certificates. Given their current number of ISO 9000 certificates, only El Salvador, Nicaragua, and Guyana, each with an extremely low number of certifi- cates, and Chile, had growth rates higher than their expected values. One country, Belize, had a negative growth rate. Figure 8.3 Number of ISO 9001:2000 and ISO 14001 Certificates, 2004 60,000 IS0 9001 50,884 IS0 14001 50,000 40,972 es 40,000 tificat erc 30,000 of 20,000 number 12,558 12,416 10,000 6,120 4,337 4,149 4,120 3,391 3,134 924 299 205 105 57 34 25 12 9 0 o IndiaRep . of ysia Spain Chile Peru Rica emala Kingdom Brazilala M gentina Ecuador Salvador Jamaica ea, Ar ColombiaMexicIndonesia Venezuela Costa ed El Guat Honduras Kor Unit Source: ISO 2005. 152 Quality Systems and Standards for a Competitive Edge Figure 8.4 Growth Rates of ISO 9000 Certificates in 2004 vs. Number of ISO Certificates; World Average Annual Growth Rates, 2001­04 2 world average growth rate El Salvador 9000 1 Guyana Chile ISO log, Nicaragua of es Bolivia Mexico Argentina 0 Ecuador wthorg Colombia tificat Costa Rica Uruguay erc Panama Venezuela, Brazil ­1 Guatemala Peru R. B. de annual Paraguay ­2 0 5 10 15 ISO 9000 certificates, log Source: ISO 2005. Note: The world average growth rate was calculated using data from 170 countries. Belize had a negative growth rate and is not shown in the gure. Roughly half of Latin American and Caribbean countries have ISO 14000 growth rates above their expected value, according to world aver- ages. The fitted line in figure 8.5 shows average global growth rates for all countries from 1998 to 2004 for a given number of certificates. ISO 14000 adoption has grown faster in Latin America and the Caribbean than in the rest of the world. Some countries, such as Ecuador and Chile, are performing particularly well. Nonetheless, a few countries have had negative or zero growth rates. Relative ISO 9000 certification rates vary widely in Latin America and the Caribbean, but most countries are not performing to global standards. Because most ISO 9000 certificates are obtained in the manu- facturing and service sectors, standardizing the number of certificates by the total value added of these two sectors allows for more informative cross-country comparisons. Ideally, certification data would be standard- ized by the number of firms in a given country, but due to the difficulty in obtaining reliable and comparable firm-level data for a broad range of countries, certification rates are standardized by value added instead. While a better indicator than crude certification, this method has a few shortcomings. A country with a very concentrated market structure (where a small number of large firms produce most of the value added) will appear to have a lower standardized certification rate than a country Certification in Latin America 153 Figure 8.5 Growth Rates of ISO 14000 Certificates in 2004 vs. Number of ISO Certificates; World Average Annual Growth Rates, 1998­2004 3 world average growth rate Ecuador 2 14000 ISO log, 1 of es Guatemala Chile wthorg tificat 0 Bolivia erc Colombia Brazil Costa Rica ­1 Argentina Peru annual Uruguay Mexico ­2 0 2 4 6 8 10 ISO 14000 certificates, log Source: ISO 2005. Note: Belize, Guyana, Honduras, Panama, Paraguay, and República Bolivariana de Venezuela had negative or zero growth rates and are not shown in the gure. with a very atomized market structure (where value added is distributed among a large number of small firms), even if the actual number of certi- fied firms is greater in the former than in the latter. Hence, standardizing the number of certificates by value added should only be used as a proxy for the diffusion of quality standards in the economy, not as an accurate indicator. As shown in figure 8.6, differences in certification performance are not merely due to differences in the value added of manufacturing and services across countries. Colombia leads the Latin American and the Caribbean region and, even by global standards, registers an extremely high number of ISO 9000 certificates given the size of its manufacturing and service sectors. In contrast, Mexico, which registers a high absolute number of certificates, does not perform well considering the size of its manufacturing and service sectors. Figure 8.7 shows the certification performance of Latin American and Caribbean countries around a line representing predicted certification rates based on world averages. While a few countries, such as Colombia and Uruguay, are performing better than world averages, most countries display lower certification rates or lag far behind. This contrasts with the more uniform situation presented in East Asia, where most countries are performing better than world averages. 154 Figure 8.6 ISO 9001 Adoption Standardized by Value Added in Manufacturing and Services, 2004 120 90 85 100 80 75 ISO 9001 80 value added in certificates per 70 manufacturing US$1 billion and services value added in 60 65 (% GDP) manufacturing and services 60 (constant 2000 US$) 40 55 50 20 45 0 40 y e ru o y C C C Pe B. de Rica . of Guyana Tobago Chile enada Bolivia Spain Gr ParaguaBelizHondurasBarbadosAr Lucia emala Brazil LA LA LA India Colombia gentina R. SurinameEcuadorJamaica Costa Salvadoraribbeanlar Panamasmall ge Mala ysiaChina Indonesia Thailand Rep ea, Kingdom and Saint GuatNicaragua RepublicUrugua Mexic El C average Kor ed Trinidad Venezuela, Dominican average average average Unit ISO 9001 certificates per US$1 billion value added in manufacturing and services (constant 2000 US$) value added in manufacturing and services (% GDP) Sources: World Bank 2005c; ISO 2005. Note: LAC = Latin America and the Caribbean. Certification in Latin America 155 Figure 8.7 Number of ISO 9000 Certificates and Value Added in Manufacturing and Services, 2002 14 log, Latin America Caribbean es 12 East Asia predicted 10 tificat Brazil erc 8 Uruguay Colombia Mexico 6 Panama Argentina 9000 Paraguay Peru Chile Venezuela, R. B. de 4 ISO Nicaragua Costa Rica Ecuador of 2 Guyana Guatemala Belize Bolivia 0 El Salvador Suriname Honduras number­2 18 20 22 24 26 28 30 value added in manufacturing and services (current US$), log Sources: World Bank 2005c; ISO 2005. Note: Predicted values were computed using data from 116 countries. The number of ISO 9000 certificates standardized by the number of establishments in the major manufacturing sectors confirms the large dif- fusion disparities across Latin American and Caribbean countries. For some countries, reliable data are readily available on both the number of establishments and the number of certificates in certain manufacturing sectors. This allows for a standardization of the number of certificates by the number of establishments in the 10 major manufacturing sectors that account for 46 percent of global ISO 9000 certificates. This exercise reveals again Colombia's leadership position in adoption of quality man- agement systems, with almost 28 percent of firms certified (figure 8.8). Colombia performs better than Spain, Korea, the United Kingdom, and the other Latin American countries. At the other extreme, Ecuador lags far behind all other countries, with only 2 percent of firms certified in the 10 sectors of interest. Latin American and Caribbean countries perform relatively poorly in terms of ISO 14000 certification rates. As shown in figure 8.9, most of the Latin American and Caribbean countries performing better than expected from their value added in manufacturing and services are coun- tries with very few ISO 14000 certificates, such as Belize and Guyana. The bulk of countries, with more mature environmental certification markets, underperform. The comparison with East Asian countries is striking. Most countries in that region exhibit higher ISO 14000 certifi- cation rates than would be expected. 156 Quality Systems and Standards for a Competitive Edge Figure 8.8 Aggregate Number of ISO 9001 Certificates Standardized by Number of Establishments in 10 Major Manufacturing Sectors 35 30 27.7 25 tified erc 20 14.6 firms 15 of 11.2 % 10 7.9 7.7 5 4.1 2.0 0 . of o Spain Brazil Colombia Rep Mexic Ecuador ea, Kingdom Kor ed Unit Sources: ISO 2005; UNIDO INDSTAT 4 (Industrial Statistics Database), Rev. 3, 2005. Note: Manufacturing sectors include basic metal and fabricated metal products; chemicals, chemical products, and bers; concrete, cement, lime, plaster, etc.; electrical and optical equipment; food products, beverage, and tobacco; machinery and equipment; pulp, paper, and paper products; rubber and plastic products; textiles and textile products (except Brazil); leather and leather products (except Brazil). Certi cate data are December 2004. Establishment data are 2000 (Colombia, Spain, Mexico), 2001 (Korea, the United Kingdom), and 2002 (Brazil, Ecuador). Figure 8.9 Number of ISO 14000 Certificates and Value Added in Manufacturing and Services, 2002 log, 10 es Latin America Caribbean 8 East Asia predicted Brazil tificat erc 6 Mexico Colombia Argentina Costa Rica 4 Chile 14000 Uruguay Peru Venezuela, R. B. de ISO 2 Guyana Belize Bolivia of Honduras 0 Ecuador number­2 20 25 30 value added in manufacturing and services (current US$), log Sources: World Bank 2005c; ISO 2005. Certification in Latin America 157 The number of ISO 14000 certificates standardized by the number of establishments in the 10 major manufacturing sectors gives a much more positive picture of the performance of certain Latin American and Carib- bean countries in relative terms. In manufacturing, Mexico, Colombia, and Brazil have adopted nearly as many certificates on a per-firm basis as Korea, and twice as many as Spain and the United Kingdom (figure 8.10). This can be explained by the fact that the vast majority of environmental management system certificates in Latin America and the Caribbean are adopted in the manufacturing sectors, as opposed to the service sector. Again, at the other extreme, there are still countries that do not register a single ISO 14000 certificate in the major manufacturing sectors. In the automobile sector, Latin American countries are performing near or above world averages in terms of their adoption of sector-specific quality management system certification. Examining sector-specific cer- tification rates allows us to correlate certification data with data con- cerning the number of firms, because the number of firms in the sector Figure 8.10 Aggregate Number of ISO 14000 Certificates Standardized by Number of Establishments in 10 Major Manufacturing Sectors 3.0 2.5 2.4 2.3 2.1 2.0 tified erc 1.5 1.4 firms 1.0 of 1.0 0.9 % 0.5 0.0 0 . of o Rep Mexic Brazil Spain Ecuador ea, Colombia Kingdom Kor ed Unit Sources: ISO 2005; UNIDO INDSTAT 4 (Industrial Statistics Database), Rev. 3, 2005. Note: Manufacturing sectors include basic metal and fabricated metal products; chemicals, chemical products, and bers; concrete, cement, lime, plaster, etc.; electrical and optical equipment; food products, beverage, and tobacco; machinery and equipment; pulp, paper, and paper products; rubber and plastic products; textiles and textile products (except Brazil); leather and leather products (except Brazil). Certi cate data are December 2004; establishment data are 2000 (Colombia, Spain, Mexico), 2001 (Korea, the United Kingdom), and 2002 (Brazil, Ecuador). 158 Quality Systems and Standards for a Competitive Edge is more readily available than the total number of firms in an economy. Figure 8.11 displays the numbers of QS 9000 and ISO/TS 16949 cer- tificates in different countries compared with the number of establish- ments in the automobile industry. Both of these standards incorporate the requirements of ISO 9000 in addition to some industry-specific requirements.1 Colombia, Ecuador, and Brazil are close to world aver- ages relative to the number of establishments in the auto industry. Mexico registers many more certificates than would be expected for its number of manufacturers, and roughly as many certificates as Canada, Germany, or Japan.2 Figure 8.11 Number of QS 9000 and ISO/TS 16949:2002 Certificates vs. Number of Enterprises in the Automobile Sector 10 Korea, Rep. of log, 8 United Kingdom es Germany Mexico Canada India tificat Japan Spain erc Brazil Australia 6 France Italy Singapore Malaysia Turkey Sweden Hungary Belgium 16949:2002 Austria Netherlands S Iran, Islamic Rep. of Portugal 4 Indonesia Colombia ISO/T Finland Norway and Egypt, Arab Rep. of Ecuador New Zealand 9000 2 Morocco Romania Malta QS of Ukraine Oman number 0 Lithuania ­2 0 2 4 6 8 10 number of establishments in the automobile sector, log Sources: ISO 2005; QSU Publishing Co. Web site; UNIDO INDSTAT 4 (Industrial Statistics Database), Rev. 3, 2005. Note: 2005 data for QS 9000; 2004 data for ISO/TS 16949; 2000, 2001, and 2002 data for number of enterprises. Certification in Latin America 159 The Supply of Certification Services Domestic firms can have their quality management systems certified by either national or foreign certification bodies, but foreign certification bod- ies dominate the certification markets of Latin American and Caribbean countries. Figure 8.12 shows the number of foreign affiliates represented in different countries of the region, along with the number of national certification bodies. Foreign certification bodies with no local representa- tion may also provide certification services to local firms, but they are not accounted for in the figure. Only certification bodies accredited nationally or abroad are considered. Most countries, even those with small certifica- tion markets like Peru and Ecuador, have a reasonable supply of certifica- tion services, but with the information available it is difficult to confirm whether there is sufficient competition in the certification industry. Available market share information in Argentina, Brazil, and Mexico shows that the market is relatively concentrated. As shown in figure 8.13, Mexico's four largest certification bodies, with size measured by Figure 8.12 Accredited Domestic Certification Bodies and Foreign Affiliates Offering Quality Management Systems Certification, 2005 30 26 25 25 firms or 20 19 es 17 15 14 affiliat of 10 10 9 number 5 0 o Mexic Brazil Chile Peru gentina Ecuador Ar Colombia number of foreign affiliates not accredited by the national body number of foreign affiliates accredited by the national body number of domestic companies accredited by the national body Sources: Web sites of the national accreditation bodies and certi cation bodies. 160 Quality Systems and Standards for a Competitive Edge Figure 8.13 Market Shares of Top Five Certification Bodies in Brazil, Argentina, and Mexico Brazil Argentina Mexico others BVQI IQS* others BVQI SGS others ATR* TUV Rheinland UL DNV FCAV* DNV SGS IRAM* Mexiko* CALMECA C* ICS BRTÜV Sources: Web sites of the Ministry of the Economy of Mexico, the Brazilian National Institute of Metrology, Stan- dardization and Industrial Quality (INMETRO), and the Argentina Accreditation Bureau (OAA). Note: Market share is measured by the number of ISO 9001:2000 certi cations issued. * denotes a domestic certi cation body, as opposed to a foreign a liate. 2005 data for Brazil and Mexico; 2003 data for Argentina. the number of ISO 9001 certifications issued, hold close to 50 percent of the market. The dominant bodies in Mexico tend to be domestic. In Brazil and Argentina the top four certification bodies control a much larger share of the market, above 70 percent. Most of them are affiliates of foreign certification bodies; in fact, in these two countries there is only one national certification body among the top five certifiers. Most Latin American countries, especially those registering few ISO 9000 certificates, have virtually no domestic certification bodies. This is the case of Ecuador and Peru. Colombia is an exception; in 2004, 42 percent of ISO 9000 cer- tifications were granted by ICONTEC, a national certification body. The average number of ISO 9000 registrations per certification body shows wide cross-country disparities in the development of the certifica- tion market. Figure 8.14 displays the average number of certifications by certification body operating within a country. Foreign bodies with no per- manent local representatives or affiliates were not included in this analy- sis. Certification bodies in Colombia are the most active, having registered on average 294 quality management systems in 2004. Those in Ecuador are the least active, having registered only six firms each in 2004. The quality of certification services in Latin America and the Carib- bean depends on the quality of the auditing staff. There are relatively few lead auditors with internationally recognized credentials in this region. The International Register of Certificated Auditors (IRCA) is the larg- est international certification body for auditors of management systems. IRCA standards are high and IRCA certification instills confidence in an auditor's qualifications. When value added in the manufacturing and service sectors is accounted for, Latin American countries have few lead Certification in Latin America 161 Figure 8.14 Number of ISO 9001:2000 Certifications in 2004 Standardized by Number of Accredited Certification Bodies Operating Domestically in 2005 350 per 300 294 es 245 250 tificat erc body 207 200 of 150 130 number ertificationc 100 54 erageva 50 21 6 0 o Brazil Chile Peru Colombia gentina Mexic Ecuador Ar Source: ISO 2005. Figure 8.15 Number of Lead Auditors Certified for ISO 9001:2000 with IRCA in 2005, Standardized by Total Value Added in Manufacturing and Services in 2004 fo es 10 vic 9 billion ser 8.2 8 and US$1 7 per turing 6 5.4 ors 5 audit 4 manufac in 3 lead 1.9 of 2 1.5 1.4 added 1 0.8 0.1 0.1 0.1 0.1 0 number value . of o Spain Brazil Peru Chile Kingdom Rep gentina Mexic Ecuador ea, Ar Colombia ed Kor Unit Sources: IRCA Web site; World Bank 2005c. auditors certified for ISO 9001 with the IRCA compared to the United Kingdom or Korea (figure 8.15). Brazil is the top performer in Latin American and registers nearly as many IRCA-certified auditors, relative to its national manufacturing and services value added, as Spain. 162 Quality Systems and Standards for a Competitive Edge Possible Factors Influencing ISO 9000 Diffusion As discussed in chapter 6, national characteristics can significantly affect ISO 9000 adoption rates. A comparison of national market and trade char- acteristics, and of institutional and business environments, can provide some context for understanding cross-country differences in ISO 9000 diffusion. Market and trade factors Trade destination does not appear to be a prime determinant of certification in Latin America. The European Unionn (EU) and Japan are high adopters of ISO 9000 and are likely to impose certification requirements on their international suppliers. Countries with higher shares of exports to the EU and Japan should be expected to have more incentives for certification. However, heterogeneity in ISO 9000 adoption in Latin America does not appear to be determined by trade patterns. Countries where exports to the EU and Japan play a more important role in the economy do not display noticeably higher certification rates than others (figure 8.16). An extreme case is Colombia, which displays the highest certification rate despite send- ing relatively few exports to the EU or Japan. One explanation is that coun- Figure 8.16 Exports to the EU and Japan as a Share of GDP and Certification Rates 12 80 es vic GDP 70 of 10 ser billion % 60 and as 8 50 US$1 per turing Japan 6 40 es and 30 EU 4 tificat manufac ot 20 erc in ts 2 10 9001 added expor 0 0 ISO o value Chile Peru Brazil Ecuador gentina Mexic Ar Colombia exports to EU and Japan as % of GDP ISO 9001 certificates per US$1 billion value added in manufacturing and services Sources: UNCTAD 2004; IMF 2005; World Bank 2005c; ISO 2005. Note: ISO 9001 data are 2004; export data are 2003. Certification in Latin America 163 Figure 8.17 Inward FDI Stock as a Share of GDP and Certification Rates 70 80 es vic 60 70 ser billion 60 50 and GDP US$1 of 50 % 40 per turing as 40 es 30 ock 30 st tificat manufac 20 in FDI 20 erc 10 10 9001 added ISO 0 0 o ru value Chile Brazil Pe Ecuador gentina Mexic Ar Colombia FDI stock as % of GDP ISO 9001 certificates per US$1 billion value added in manufacturing and services Sources: UNCTAD 2004; IMF 2005; World Bank 2005c; ISO 2005. Note: ISO 9001 data are 2004; FDI data are 2003. tries trading heavily with the EU and Japan may not be exporting goods in sectors where quality is necessary. Another explanation is that other factors affect ISO 9000 adoption to a greater extent than trade destination. In Latin America, the presence of transnational corporation (TNC) sub- sidiaries is not a prime determinant of ISO 9000 adoption. The situation with respect to TNCs is very similar to that of trade destination: there is no obvious pattern of higher certification rates in countries with a greater presence of TNCs. Figure 8.17 uses inward foreign direct investment (FDI) stock as a proxy for the presence of foreign affiliates. As for trade destina- tion, either the TNC subsidiaries are not in sectors where quality is impor- tant, or other factors play a larger role in the diffusion of ISO 9000. The presence of export sectors that are conducive to certification appears to be moderately related to ISO 9000 adoption. As discussed in chapter 6, some export sectors are associated with higher global certifica- tion rates than others. As of December 2004, the top eight global sectors in terms of ISO 9000 concentration were (ISO 2005) 1. Construction 2. Basic metals and fabricated metal products 164 Quality Systems and Standards for a Competitive Edge 3. Electrical and optical equipment 4. Machinery and equipment 5. Wholesale and retail trade; repair of motor vehicles, motorcycles, and personal and household goods 6. Rubber and plastic products 7. Chemicals, chemical products, and fibers 8. Food products, beverages, and tobacco These sectors register high certification rates, either because quality is considered to be an important competitive factor in these sectors or because ISO 9000 is particularly effective in increasing organizational efficiency in these sectors. Countries where exports in these quality-focused global sectors account for a large share of GDP should also display high certification rates. This pattern seems to hold for Latin America, except for Mexico, where a high share of exports are destined to the United States, and Colombia, where ISO 9000 diffusion is exceptionally high (figure 8.18). It is not clear that the size of manufacturing firms is influencing ISO 9000 diffusion in Latin America. Smaller firms are generally less aware of Figure 8.18 Exported Goods in Global Quality­Focused Sectors as a Share of GDP 20 80 18.3 es 18 vic 70 ser 16 billion 14.4 60 14 13.3 and (%) US$1 12 50 per turing GDP 10 9.1 40 of es e 8 30 6.0 5.9 tificat shar 5.8 manufac 6 20 erc in 4 2 10 9001 added 0 0 ISO o value Chile Mexic Brazil Peru gentina Ecuador Ar Colombia share of GDP of exported goods in top 10 sectors with highest ISO 9000 concentration ISO 9001 certificates per US$1 billion value added in manufacturing and services Sources: UNCTAD 2004; World Bank 2005c; ISO 2005. Note: ISO 9001 data are 2004; export data are 2003 for all countries except Argentina, 2002. Certification in Latin America 165 the benefits of ISO 9000 and the implementation of a quality management system is more costly to them. Thus, countries where small and medium enterprises (SMEs) account for a high share of firms should display rela- tively fewer certificates per firm. According to figure 8.19, Latin Ameri- can countries do not seem to have a particularly high share of SMEs in their economies, as proxied by the share of manufacturing employment in SMEs. East Asia, where SMEs account for a larger share of manufactur- ing employment, have attained significantly greater certification rates than Latin America and the Caribbean.3 Figure 8.20 displays SME shares of GDP in different countries, based on data from the manufacturing sector. Here, no clear trend emerges, possibly because certification data are standardized by output and not by the number of establishments. Nonetheless, firm size and certification data in Mexico (figure 8.21) do reveal a generally higher tendency to adopt ISO 9000 in larger firms. The economic environment Latin America's poor business environment may be reducing incentives for certification.A business climate characterized by uncertainty and high busi- ness costs may discourage any long-term investments in a quality manage- ment system. An aggregate indicator constructed by Beck, Demirgüç-Kunt, Figure 8.19 Regional SME Sector Share of Manufacturing Employment, 1990­99 Averages 80 74 70 (%) 63 62 60 56 55 yment 50 emplo 40 of e 30 28 shar 20 16 SME10 0 Asia Europe and and CIS es Africa East AmericaStat Canada Zealand Latin ed Australia New Unit Source: Beck, Demirgüç-Kunt, and Levine 2003. Note: SMEs are de ned as having fewer than 250 employees. CIS = Commonwealth of Independent States. 166 Quality Systems and Standards for a Competitive Edge Figure 8.20 SME Sector Share of GDP and Certification Rates 100 80 (%) 65 GDP 60 56 54 51 of e 46 39 40 shar SME 20 20 0 . of Spain Peru gentina Ar Kingdom Rep ea, Colombia Ecuador ed Kor Unit SME share of GDP (%) ISO 9001 certificates per US$1 billion value added in manufacturing and services Sources: Ayyagari, Beck, and Demirgüç-Kunt 2003; World Bank 2005c; ISO 2005. Note: ISO 9001 data are 2004 for every country except 2003 for Korea, 2002 for Spain, 2001 for the United King- dom. National SME de nitions are used. SME sector share is based on 1990­99 averages in the manufacturing sector. Figure 8.21 ISO 9001 Adoption and Firm Size in Mexico Distribution of ISO 9001:2000 Firms by size certificates by size of firm Small Medium 3.1% 0.9% Large Large 18.67% 0.3% Medium 45.78% Micro 16.64% Micro Small 95.7% 18.83% Sources: Mexican Ministry of the Economy Web site; Contactopyme Web site. Note: Percentages may not add to 100 due to rounding. Certification in Latin America 167 Figure 8.22 Average Business Environment Indicator, by Region 3.0 2.5 2.2 2.0 index 1.5 1.3 1.0 onment 0.5 0.5 vir 0.2 en 0.0 ­0.5 ­0.5 ­1.0 ­0.7 business ­1.0 ­1.5 ­2.0 and and Asia CIS Latin es Canada ZealandEurope Africa East America Stat ed Australia New Unit Sources: Beck, Demirgüç-Kunt, and Levine 2003; World Bank 2005c; ISO 2005. Note: Based on data from 64 countries. CIS = Commonwealth of Independent States. and Levine (2003) includes information on the degree of private property rights protection, the cost of contract enforcement, the cost of entering the market, and the efficiency of the bankruptcy system. Average regional business environment indicators reveal a particularly poor business climate in Latin America (figure 8.22). A comparison of national business environ- ment indicators and certification rates shows a tendency for countries with sounder business climates to exhibit higher certification rates (figure 8.23). Latin America's poor institutional environment may also serve to reduce incentives for certification. Ineffective and unpredictable institu- tions can discourage the implementation of a quality management sys- tem, because firms are unable to predict the returns on their investments. An aggregate indicator constructed by Beck, Demirgüç-Kunt, and Levine (2003) rates the institutional environment in which firms operate. This indicator contains information on voice and accountability in the politi- cal system, the quality and consistency of regulations and regulatory enforcement in the country, political stability, rule of law, lack of offi- cial corruption, and effectiveness of the government bureaucracy. Again, Latin America performs poorly according to this indicator (figure 8.24). With respect to individual countries, there is a tendency for countries with better institutional environments to have greater ISO 9000 diffu- sion rates (figure 8.25). Colombia is an exception here, as it displays the 168 Quality Systems and Standards for a Competitive Edge Figure 8.23 Business Environment Indicator and Certification Rates, by Country 2.5 100 es vic 2.0 ser 80 billion index 1.5 and US$1 60 1.0 per turing onment es vir 0.5 en 40 tificat manufac 0.0 erc in 20 business­0.5 9001 added ISO ­1.0 0 . of o value Spain Chile Brazil Peru Kingdomea, Rep ColombiaAr gentina Mexic ed Kor Unit business environment indicator ISO 9001 certificates per US$1 billion value added in manufacturing and services Sources: Beck, Demirgüç-Kunt, and Levine 2003; World Bank 2005c; ISO 2005. Note: ISO 9001 data are 2004 for every country except 2003 for Korea, 2002 for Spain, and 2001 for the United Kingdom. Figure 8.24 Average Institutional Development Indicator, by Region 2.0 1.5 1.5 1.4 index 1.0 0.8 elopment 0.5 0.4 dev 0.0 0.0 ­0.5 institutional ­0.5 ­0.7 ­1.0 and and Asia CIS Latin Zealandes Canada Europe Africa East America Australia Stat New ed Unit Source: Beck, Demirgüç-Kunt, and Levine 2003. Note: CIS = Commonwealth of Independent States. Figure 8.25 Institutional Development Indicator and Certification Rates, by Country 2.0 100 es vic 1.5 ser index 80 billion and 1.0 US$1 60 per turing elopment 0.5 es dev 40 0.0 tificat manufac erc in 20 ­0.5 institutional 9001 added ISO ­1.0 0 . of o value Spain Chile Brazil exic Peru Kingdom Rep gentina M ea, Ar EcuadorColombia ed Kor Unit institutional development indicator ISO 9001 certificates per US$1 billion value added in manufacturing and services Sources: Beck, Demirgüç-Kunt, and Levine 2003; World Bank 2005c; ISO 2005. Note: ISO 9001 data are 2004 for every country except 2003 for Korea, 2002 for Spain, and 2001 for the United Kingdom. Figure 8.26 Secondary School Enrollment and Certification Rates 120 100 es (%) e 90 vic rat 100 ser 80 billion and 70 80 ollment US$1 60 enr per turing 60 50 es net 40 40 tificat manufac school 30 erc in y 20 20 9001 ondar 10 added ISO sec 0 0 . of o value Spain Chile Brazil Peru Kingdom Rep gentina Mexic ea, Ar ColombiaEcuador ed Kor Unit secondary school net enrollment rate (%) ISO 9001 certificates per US$1 billion value added in manufacturing and services Sources: World Bank 2005c; ISO 2005. Note: ISO 9001 data are 2004 for every country except 2003 for Korea, 2002 for Spain, 2001 for the United Kingdom. School enrollment rates are for 2002. 169 170 Quality Systems and Standards for a Competitive Edge worst institutional environment in the sample but the highest certifica- tion rate in Latin America. Latin America's low educational levels hinder the implementation of quality management systems. ISO 9000 implementation costs and time are lower when the involved personnel have received at least minimum levels of basic educational training. Latin American countries are gener- ally characterized by low levels of educational attainment. In particular, de Ferranti et al. (2003) observe that most Latin American countries have massive deficits in net enrollment in secondary schools. These enrollment rates are much lower than what would be predicted based on per capita GDP. Figure 8.26 points to a strong correlation between ISO 9000 adop- tion rates and net secondary education rates in Latin America. Countries with higher education rates tend to have higher ISO 9000 diffusion rates, Colombia, as always, being an exception. It follows that low educational levels could contribute to Latin America's general difficulty in imple- menting ISO 9000. C H A P T E R 9 Accreditation in Latin America International recognition is critical for access to markets, and while Latin America has been making progress, much more is required to support the aggressive export strategies embraced by most countries in the region. Accreditation bodies need increased private sector representation--it is now minimal--and much more autonomy to enable them to quickly respond to sectoral needs and firms' demands. Accreditation Institutions The organization of accreditation Latin American accreditation bodies vary in legal status. As is the case in the rest of the world, there are both private and public accreditation bodies in the region (table 9.1). When they are part of the public sector, accreditation bodies tend to operate as autonomous government agen- cies, sometimes affiliated with a specific government ministry. Examples include Peru's INDECOPI (National Institute for the Defense of Compe- tition and the Protection of Intellectual Property) and Brazil's INMETRO (National Institute of Metrology, Standardization and Industrial Quality). Mexico has recently moved its accreditation system from the Ministry of the Economy to a private nonprofit organization. In the case of Chile and Argentina, the National Standardization Institute (INN) and the Argentine Accreditation Bureau (OAA) are private organizations but were created by the government. In all countries surveyed, the functions and obliga- tions of both public and private accreditation bodies are governed through national legislation, which gives full authority over voluntary standards to a single accreditation body. Many accreditation systems in Latin American countries are relatively new and were formed in the past decade. 171 172 Quality Systems and Standards for a Competitive Edge Table 9.1 Legal Status of National Accreditation Bodies Country Name Status Argentina OAA private Brazil INMETRO public Chile INN private Colombia SIC public Ecuador OAE public Mexico EMA private Peru INDECOPI public Korea, Rep. of KAB private Korea, Rep. of KATS-KOLAS/KAS public Spain ENAC private Sources: Web sites of the national accreditation bodies. Note: EMA = Mexican Accreditation Body; ENAC = National Accreditation Body; KAB = Korea Accreditation Board; KATS-KOLAS/KAS = Korea Agency for Technology and Standards, Korea Laboratory Accreditation Scheme, Korea Accreditation System; OAE = Ecudorian Accreditation Body; SIC = Superintendency of Industry and Commerce. Accreditation bodies in Latin American countries tend to be involved in more activities than their developed-country counterparts (table 9.2). In some Latin American countries, the accreditation body is responsible for accrediting organizations against regulatory standards in addition to voluntary standards. This is also the case in Spain. In countries where the accreditation body operates as part of the public sector, it is often respon- sible for other activities related to voluntary conformity assessment. In Bra- zil, Chile, Colombia, and Peru, the accreditation body is also the national metrology institute (NMI). This may be problematic if the accreditation and metrology functions are not clearly separated, because the NMI offers calibration services that could theoretically compete with the services of potential candidates for accreditation. In Chile and Peru, the national accreditation body is also the national standards body. In these cases, it is important that the NMI, itself potentially involved in the development of standards, not influence the standardization process disproportionately. In developed countries such as Spain and the United Kingdom, metrology and standardization functions are usually performed by different organiza- tions. None of the accreditation bodies in Latin America currently offer commercial certification services. Governance and autonomy In Latin America, some accreditation bodies have very little private sector representation (table 9.3).This is often the case in public sector bodies,where Table 9.2 Activities of National Accreditation Bodies Accreditation against regulatory National Calibration Country standards Standards metrology and testing Certification Argentina yes no no no noa Brazil yes no yes yes no Chile no yes yes no nob Colombia yes no yes yes no Ecuador yes no no no no Mexico yes no no no no Korea, Rep. of: no no no no yesc KAB Korea, Rep. of: no yes yesd no yes KATS-KOLAS/KAS Peru no yes yes yes no Spain yes no no no no United Kingdom no no no no no Sources: Web sites of the national accreditation bodies. Note: KAB = Korea Accreditation Board; KATS-KOLAS?KAS = Korea Agency for Technology and Standards, Korea Laboratory Accreditation Scheme, Korea Accreditation System. a. By law, OAA is entitled to certify auditors, but it has never exercised this right. b. INN certi ed auditors until 2004. c. Certi cation of management system auditors. d. Legal metrology only. Table 9.3 Private Sector Participation in the Governance of Accreditation Bodies Private sector Private sector Share of executive Executive participation participation council members director in general in consultative appointed by appointed by Country assembly council government (%) government Status Argentina yes no 15 yes private Brazil no yes 100 yes public Chile no no 100 no private Colombia no no 100 yes public Ecuador yes yes 50 no public Mexico yes no 28 no private Peru no yes 100 yes public Spain yes yes 0 yes private United yes yes 0 no private Kingdom Sources: Web sites of the national accreditation bodies; authors'research. 173 174 Quality Systems and Standards for a Competitive Edge there is no formal membership mechanism for the private sector, such as in the Superintendency of Industry and Commerce (SIC) in Colombia and INDECOPI in Peru. But this is also the case in Chile, where the INN was created by the government as a private foundation. In the case of Colombia and Chile, there is neither a general assembly nor a consultative council to represent the interests of stakeholders. In contrast, most private accredita- tion bodies have a general assembly of members who elect an executive council or board of directors. Nonetheless, in all of these cases a share of the executive council members are appointed by government entities to ensure that they are adequately represented. The case of Argentina is fairly similar to that of Spain: there is heavy private sector representation in the executive council, but the executive director is appointed by the central government. Accreditation bodies operating in both the private and public sectors are far from fully autonomous. An autonomy index, showing an organization's ability to exercise authority over its financial, technical, administrative, and human resources, can be calculated.1 While this index reaches unity for most Latin American standards bodies, it is much lower for accreditation bodies, Figure 9.1 Accreditation Body Personnel, 2005 90 1.9 2 1.8 79 80 value es 70 67 billion vic ser staff 60 1.3 1.3 1.2 US$10 and 1.1 50 per 1.0 1 turing permanent staff 40 of 30 manufac in number 20 permanent of 11 10 added 10 8 5 3 0 0 number o Brazil Mexic Chile Peru gentina Ecuador Ar Colombia number of permanent staff number of permanent staff per US$10 billion value added in manufacturing and services Sources: Authors'research; Web sites of the national accreditation bodies. Accreditation in Latin America 175 reaching 0.250 for Peru's INDECOPI and Colombia's SIC, 0.625 for Brazil's INMETRO, and 0.875 for the Ecuadorian Accreditation Body (OAE). Resources for accreditation There are important differences in the number of staff working in the different accreditation bodies, but these differences are smaller if the size of the industrial economy is taken into account. Brazil's INMETRO and the Mexican Accreditation Body (EMA) have a large number of person- nel, but this is mostly due to the size of their manufacturing and service sectors (figure 9.1). When the number of personnel is standardized by value added in manufacturing and services, most accreditation bodies have roughly the same number of personnel. The number of evaluation staff supporting the accreditation process is highly unequal across countries. Standardizing the number of lead assessors by value added in manufacturing and services highlights important dispari- ties in this regard (figure 9.2). Ecuador and Argentina have an exceptionally Figure 9.2 Number of Qualified Lead Assessors, 2005 140 6 126 es 120 5.0 5 billion vic ser 100 US$10 and 4 per US$) assessors 80 turing staff lead 2.7 3 2000 59 of 60 2.3 53 manufac 2 in onstant (c number 40 1.5 permanent 1.3 of 1.1 added 1 20 8 7 0.4 6 number value 2 0 0 o Mexic Brazil Chile Peru gentina Ecuador Ar Colombia number of lead assessors number of permanent staff per US$10 billion value added in manufacturing and services (constant 2000 US$) Sources: IAAC 2004; World Bank 2005c; Superintendencia de Industria y Comercio of Colombia; authors'research. Note: 2004 data for Argentina, Chile, and Mexico. 176 Quality Systems and Standards for a Competitive Edge high standardized number of lead assessors, while Peru has very few. Lack of evaluation staff can constrain the speed of accreditation. The picture changes when comparing the number of technical experts. Technical experts are usually contracted to offer expertise in areas where assessors are not specialized. Hence, a low number of technical experts will limit the number of technical areas in which the accreditations can be performed. Brazil and Colombia have the largest number of technical experts in Latin America (figure 9.3). When this number is standardized by value added in manufacturing and services, Colombia and Peru have the highest numbers of technical experts, while Mexico has the lowest num- ber.Technical experts may be especially useful where a low number of lead assessors limits their ability to specialize in different technical areas. Latin American accreditation bodies are generally slower at delivering accreditations than counterpart bodies elsewhere. It takes 50 percent to Figure 9.3 Number of Qualified Technical Experts, 2005 300 30 26 value 250 241 25 es billion vic ts ser exper 200 20 US$10 and US$) 160 per 15 ts 150 15 turing 2000 echnicalt of exper 100 89 10 manufac onstant (c 76 7 71 in number echnicalt 5 5 5 50 5 of added 2 25 12 0 0 number o Brazil Peru Chile Colombia Mexic gentina Ecuador Ar number of technical experts number of technical experts per US$10 billion value added in manufacturing and services (constant 2000 US$) Source: Authors'research. Note: 2004 data for Argentina, Chile, and Mexico. Accreditation in Latin America 177 Figure 9.4 Minimum Time Required for the Laboratory Accreditation Process 12 10 10 8 8 6 6 6 5 months 4 4 4 2 0 ed Peru Unitingdom Spain Brazil Colombia Argentina K Ecuador Source: Authors'research. 150 percent longer to accredit a laboratory in Latin America than it does in the United Kingdom, the exception being Peru (figure 9.4). This could produce a disincentive to accreditation. Accreditation Activities Most Latin American national bodies have accredited a significant num- ber of quality system certification bodies (figure 9.5). Relative to value added in manufacturing and services, some Latin American countries have accredited as many as or more such certification bodies than Spain. Three countries, Brazil, Chile, and Colombia, have accredited more qual- ity system certification bodies than the United Kingdom, relative to value added in manufacturing and services. Peru and Ecuador are exceptions and lag behind, having accredited very few certification bodies. When the number of accredited inspection bodies is standardized by value added in manufacturing and services, Chile and Brazil perform much better than the United Kingdom, Korea, or Spain, and Colombia performs slightly better (figure 9.6). However, these comparisons can be misleading because they do not show the scope of accreditation.Accredi- tations issued by Latin American bodies cover a limited number of tech- nical areas. In Chile, the vast majority of inspection bodies are accredited 178 Quality Systems and Standards for a Competitive Edge Figure 9.5 Number of Accreditations of Management System Certification Bodies 140 40 120 35 value es vic 30 billion 100 ser 25 US$10 and US$) editations 80 cr per turing 2000 ac 20 of 60 15 editations manufac onstant cr (c number 40 in 10 ac of 20 added 5 number 0 0 o ed . of Chile Peru UnitingdomRep Brazil Spain Mexic ColombiaAr gentina K Ecuador ea, Kor number of accreditations of management system certification bodies (ISO 9000, QS 9000, and ISO 14000) number of accreditations per US$10 billion value added in manufacturing and services (constant 2000 US$) Sources: Web sites of the national accreditation bodies; World Bank 2005c. Note: 2003 value added for the United Kingdom, Korea, and Spain; 2004 for other countries. in the areas of gas installations and electrical products. In Colombia, most are accredited for gas installations. In both these countries, there are few or no inspection bodies accredited in other areas such as electrical instal- lations or machine safety. Three countries, Argentina, Peru, and Ecuador, have not accredited a single inspection body in any area. This can inhibit certification, because some certifications require inspections. Most Latin American countries have accredited a relatively high number of testing and calibration laboratories (figure 9.7). Comparing standardized numbers of accreditations, Latin American countries either exceed Spain's and Korea's performance or are not far behind. Nonethe- less, it is difficult to assess the performance of accreditation bodies in terms of accredited calibration and testing laboratories. There may be cross-country differences in scope of accreditation. The scope of some Accreditation in Latin America 179 Figure 9.6 Number of Accreditations of Inspection Bodies 250 12 value 10 es 200 vic billion ser 8 150 US$10 and editations cr per ac 6 turing of 100 4 editations manufac cr number in ac 50 of 2 added number 0 0 ed . of BrazilUnitingdom Rep Spain Chile Peru K Colombia Ar gentinaEcuador ea, Kor number of accreditations of inspection bodies number of accreditations of inspection bodies per US$10 billion value added in manufacturing and services Sources: Authors'research; World Bank 2005c. Note: 2003 value added for the United Kingdom, Korea, and Spain; 2004 for other countries. accreditations may be quite narrow, covering only one calibration or test area, while in other cases an accreditation may cover a wide range of measurements and tests in different fields. While an examination and comparison of individual accreditations would be an option, this is com- plicated by the fact that most Latin American countries do not follow internationally accepted guidelines in defining the scope of laboratory accreditation. In some Latin American countries, the national accreditation system appears to be either unrecognized domestically or not performing ade- quately. Evidence is provided by the number of foreign certification bod- ies that operate in Latin American countries but are not accredited by the national accreditation body (figure 9.8). In Peru and Ecuador, none of the nine quality system certification bodies operating nationally are 180 Quality Systems and Standards for a Competitive Edge Figure 9.7 Number of Accreditations of Testing and Calibration Laboratories 3000 120 value 2500 100 es vic billion ser 2000 80 US$10 and editations cr per ac 1500 60 turing of 1000 40 editations manufac cr number in ac of 500 20 added number 0 0 ed . of Chile Peru Unitingdom Spain Brazil Rep K gentina Ecuador ea, Colombia Ar Kor number of accreditations of laboratories number of laboratory accreditations per US$10 billion value added in manufacturing and services Sources: Web sites of the national accreditation bodies; World Bank 2005c; authors'research. accredited by the national accreditation body. Instead, they are accred- ited in foreign countries. This points either to a disregard for national accreditations or to the incapacity of the national accreditation body to offer competitive accreditation services. International Integration Adoption of international standards for accreditation Accreditation bodies in Latin American countries base their accredita- tion requirements on international standards. All of the accreditation bodies require conformance with standards based on ISO/IEC Guides 62, 66, and 65 for system and product certification bodies, and ISO/IEC 17025 for laboratories. In terms of inspection bodies, all countries base their requirements on ISO/IEC 17020, except for Chile, where INN uses a national standard, and Ecuador, where OAE does not offer accredi- tation for inspection bodies. Apart from Chile, all accreditation bodies Accreditation in Latin America 181 Figure 9.8 Proportion of Foreign Affiliates of Quality System Management Certification Bodies Accredited by the National Accreditation Body 100 93 80 72 national body 60 yb 60 ed 43 edit editation 40 cr cr ac ac 21 % 20 0 0 0 o Mexic Brazil Chile Peru Colombia Ar gentina Ecuador Source: Authors'research. theoretically also offer accreditations for personnel certification bodies, based on ISO/IEC 17024. However, few countries have delivered any accreditations in this area, and only Mexico and Brazil have accredited certification bodies in areas related to auditor certification. Not all accreditation bodies fully conform to international best prac- tice. Table 9.4 shows that Argentina is the only country to meet the requirements of ISO 17011:2004. This new international standard was published in September 2004 and establishes the general requirements for accreditation bodies accrediting conformity assessment bodies in the areas of testing, inspection, management system certification, personnel certification, product certification, and calibration. Members of the Inter- national Laboratory Accreditation Cooperation (ILAC) and the Interna- tional Accreditation Forum (IAF) agreed to meet the requirements of ISO 17011:2004 by January 2006. This will become a prerequisite for joining the ILAC Mutual Recognition Arrangement and the IAF Multi- lateral Recognition Arrangement (MRA), effectively replacing the previ- ous requirements established in Guide 58 for laboratory accreditation systems, Guide 61 for certification body accreditation systems, and ISO/ IEC 17010 for inspection body accreditation systems. Currently, only INMETRO in Brazil and EMA in Mexico conform to these three sets of general requirements. Other accreditation bodies partially meet the general requirements. 182 Quality Systems and Standards for a Competitive Edge Table 9.4 Implementation of ISO Guides and Standards for Accreditation Bodies ISO Guide 61 ISO 17010 ISO 17011 (all ISO Guide 58 (certification (inspection conformity Country (laboratories) bodies) bodies) assessment bodies) Argentina T T P T Brazil T T T P Chile P P P P Colombia P P P P Ecuador T P N P Mexico T T T P Peru T T P P Sources: Authors'research; IAAC Web site. Note: N = none, P = partial, T = total. Uses 2004 data for Argentina, Chile, Ecuador, and Mexico; 2005 data for Brazil, Colombia, and Peru. Regional and international cooperation and mutual recognition agreements Accreditation bodies in Latin America are involved in a number of MRAs at the subregional, regional, extraregional, and international lev- els. Table 9.5 shows the principal regional and international agreements and organizations operating MRAs involving Latin American countries. Each MRA is limited to specific categories of accreditations, including certifiers of products, of quality management systems (QMSs), of envi- ronmental management systems (EMSs), of persons, testing laboratories, calibration laboratories, and inspection bodies. While some Latin American accreditation bodies are fully cooperating with regional and international organizations to establish MRAs, others remain isolated (table 9.6). Membership in an international or regional organization is a first step toward demonstrating credibility, obtaining technical assistance, and participating in MRAs, but Colombia's accredi- tation body, the SIC, is not a member of a single international or regional accreditation organization. As a member of the InterAmerican Accredita- tion Cooperation (IAAC), Peru's INDECOPI remains slightly less isolated than the SIC, but it is not a member of either the IAF or ILAC. Colombia, Peru, and Ecuador are only parties to the Andean Community's incipient MRA, which itself is limited to product certifications. In contrast, coun- tries such as Mexico, Argentina, and Brazil are extremely well connected and internationally recognized, having secured intraregional, extraregional, and international MRAs covering most of their trade partners. Accreditation in Latin America 183 Table 9.5 Principal Regional and International Organizations Operating Mutual Recognition Agreements for Accreditation Within, or in Cooperation with, Latin American Countries Agreement or MRA or MLA organization Description Member countries signatories Mercosur (Southern Customs Argentina, Brazil, MRA under Common Market) union Uruguay, Paraguay development CAN (Andean Customs Colombia, Bolivia, Product: 5 countries Community) union Peru, República Bolivariana de Venezuela, Ecuador IAAC (InterAmerican Regional 23 countries in the QMS: 4 countries Accreditation organization Americas Calibration and Cooperation) testing: 5 countries APLAC (Asia-Paci c Regional 21 countries in the Calibration, testing, Laboratory Accreditation organization Asia Paci c region and inspection: Cooperation) 15 countries PAC (Paci c Regional 16 APEC countries + QMS: 15 countries Accreditation organization accreditation bodies EMS: 10 countries Cooperation) of Islamic Republic Product: 4 countries of Iran, Pakistan, and India EA (European Regional Members and Calibration and Co-operation for organization cooperative testing: 28 countries Accreditation ) agreements: 36 QMS: 21 countries European countries EMS: 21 countries + 11 non-European Product: 22 countries countries Inspection: 17 countries Persons: 19 countries IAF (International International 44 countries + 4 QMS: 36 countries Accreditation Forum) organization regions + 2 regions (EA and APLAC) EMS: 29 countries + 2 regions (EA and APLAC) Product: 24 countries + 2 regions (EA and APLAC) ILAC (International International 71 countries + 5 Calibration, testing, Laboratory Accreditation organization regions and inspection: 40 Cooperation) countries + 2 regions (EA and APLAC) Source: Authors'research. 184 Quality Systems and Standards for a Competitive Edge Table 9.6 Membership in International Organizations and Mutual Recognition Agreements Country IAF ILAC Regional MLA or MRA Argentina yes yes IAAC IAF, ILAC, IAAC Brazil yes yes IAAC, EA IAF ILAC, IAAC, EA Chile yes associate IAAC IAF Colombia no no no CAN Ecuador no a liate IAAC CAN Mexico yes associate IAAC, APLAC, PAC IAF, ILAC, IAAC, PAC Peru no no IAAC CAN Korea, Rep. of yes yes PAC, APLAC IAF, ILAC, PAC, APLAC Spain yes yes EA IAF, ILAC, EA United Kingdom yes yes EA IAF, ILAC, EA Sources: IAF, ILAC, IAAC, EA, APLAC, and PAC Web sites. C H A P T E R 1 0 Metrology in Latin America The Latin American region has a long history in metrology, but limited capacity to date. There is an urgent need for modernization of the rel- evant institutions so they can keep pace with new market demands and also respond more quickly to users' needs and demands. Emphasis should be given to erasing conflicts of interest and facilitating the accreditation for calibration of most metrology laboratories. The National Metrology Institutions Organization of scienti c and legal metrology institutions Apart from Chile, all Latin American countries in the survey have public scientific metrology institutions (table 10.1). This is the norm in most countries throughout the world. Although Chile's INN is technically a private nonprofit organization, it is largely subject to public sector con- trol, thus rendering it a quasi-public institution. Legal metrology institu- tions in all countries are always public. Scientific and legal metrology functions are performed by a single institu- tion in most Latin American countries. This is an accepted practice found in many other countries, although a separation of functions is also common. In Brazil, Colombia, Ecuador, and Peru, the National Metrology Institute (NMI) acts both in scientific and legal metrology.Argentina presents a hybrid situation where the NMI is jointly responsible for legal metrology with the Secretariat of Technical Cooperation within the Ministry of the Economy and Production. In Chile and Mexico, legal metrology is the responsibility of another institution. This is also the case in the Republic of Korea. 185 186 Quality Systems and Standards for a Competitive Edge Table 10.1 Organization of Scientific and Legal Metrology Institutions National National Legal status of Number of national scientific legal national scientific depositories metrology metrology metrology of reference Country institution institution institution standards Argentina INTI Technical public 2 Cooperation Secretariat, Ministry of Economy and Production/ INTI Brazil INMETRO INMETRO public 3 Chile INN SERNAC private 6 Colombia SIC SIC public 1 Ecuador INEN INEN public 1 Mexico CENAM DGN public 2 Peru INDECOPI INDECOPI public 1 Korea, Rep. of KRISS KATS public 1 Spain CEM CEM public 7 United Kingdom NWML NWML public 4 Source: Authors'research. Note: CEM = Spanish Metrology Center; CENAM = National Metrology Center; DGN = General Bureau of Stan- dards; INDECOPI = National Institute for the Defense of Competition and the Protection of Intellectual Property; INEN = Ecuadorian Standardization Institute; INMETRO = National Institute of Metrology, Standardization and Industrial Quality; INTI = National Institute of Industrial Technology; KATS = Korean Agency for Technology and Standards; KRISS = Korea Institute of Standards and Science; NWML = National Weights and Measures Labora- tory; SERNAC = National Consumer Service; SIC = Superintendency of Industry and Commerce. National metrology standards are relatively centralized in most Latin American countries. In many Latin American countries, a single national metrology institution maintains all of the national reference standards (table 10.1). Even in the case of Mexico and Argentina, where two insti- tutions are responsible for national reference standards, the system is rather centralized. In both of these countries, the primary NMI holds the totality of the reference standards, with the exception of the radia- tion standards held in a separate national nuclear laboratory. Chile is an exception in that its NMI does not maintain any of the national standards. Instead, six independent laboratories, which include for-profit compa- nies, are designated by government decree. This system is similar to that of the United Kingdom, where a combination of for-profit and nonprofit laboratories maintain the national reference standards. Metrology in Latin America 187 Table 10.2 Activities of the National Scientific Metrology Institutions Voluntary Intellectual Country standards Accreditation property Certification Argentina no no no yes Brazil no yes no no Chile yes yes no no Colombia no yes yes no Ecuador yes no no yes Mexico no no no no Peru yes yes yes no Korea, Rep. of no no no no Spain no no no no United Kingdom no no no yes Sources: Web sites of the national metrology institutions. In some cases, the NMI is also involved in activities that may create conflicts of interest. NMIs in the United Kingdom, Spain, and Korea con- centrate most of their activities on metrology. In Spain and Korea, they restrict their scope of operation to metrology and research and develop- ment. In the United Kingdom, the National Weights and Measures Labo- ratory (NWML) also performs other activities, but these are limited to certification services (table 10.2). By contrast, a number of Latin Ameri- can NMIs are housed in multifunctional institutions. In Chile, Ecuador, and Peru, the NMI is also the national standards body. Chapter 7 dis- cussed how this could affect the impartiality of the standards develop- ment process if the two functions are not properly separated. In four Latin American countries, the NMI is also the national accredi- tation body. This could create conflicts of interest because the accredita- tion body could require candidate laboratories to be calibrated with the NMI, even though this would be less reliable or more costly than with a globally recognized NMI in another country. The accreditation body could also discriminate against candidate laboratories that compete with the NMI in the market for industrial calibrations. Even when there are no competing private laboratories, the accreditation arm of the NMI could face pressures from its metrological arm to accredit its own laborato- ries to enhance their credibility. Another potential conflict of interest is found in Colombia and Peru, where the SIC and the INDECOPI serve as the NMI as well as the intellectual property registration agency. Metrol- ogy institutions conduct research and development, which may at times 188 Quality Systems and Standards for a Competitive Edge involve patentable technology. Any dispute over intellectual property rights can be unfairly resolved in favor of the NMI if the two functions are not sufficiently separated. In Argentina and Ecuador, the NMI also offers certification services. This should not pose conflicts of interest and may enable the NMI to raise additional income. Scienti c and technical capabilities Most national metrology laboratories are not accredited for calibration (table 10.3). The calibration services provided by the national metrology laboratories are the starting point for the national chain of traceability and should meet the highest standards. Accreditation of an NMI's labo- ratories can help create national confidence in the institution, increase its customer base, and facilitate the signing of MRAs. Chile stands out from the other Latin American countries, because five of its six metrol- ogy laboratories are accredited. However, these laboratories are accredited by INN, which itself is the NMI and coordinates the national metrology laboratory network. It is unclear whether this dual role has influenced the INN's impartiality. The only other NMI with laboratories accredited for calibration is in Peru, but only one of its six labs has accreditation. None of the other countries have accredited their national metrology laboratories for calibration. As mentioned in chapter 4, it is not universal practice to require national metrology laboratories to be accredited, but this can be of great benefit. Two of Spain's six national laboratories are accredited, and all four of the United Kingdom's national laboratories are accredited. While some NMIs offer national traceability in a wide range of sci- entific areas, others have a very narrow range of technical capabilities Table 10.3 Accreditation of the National Metrology Laboratories Country Number of national metrology laboratories accredited for calibration Argentina 0 Brazil 0 Chile 5 of 6 Colombia 0 Ecuador 0 Mexico 0 Peru 1 of 6 Korea, Rep. of 0 Spain 2 of 6 United Kingdom 4 of 4 Source: Authors'research. Metrology in Latin America 189 Figure 10.1 Number of Measurement Areas in the National Metrology Laboratories 25 eas 20 ar 20 19 19 18 16 ement 15 13 10 measur 10 9 of 8 5 5 number 0 o ea, Mexic . of ed Brazil KorRep Peru Chile UnitKingdomSpain gentina Ecuador Ar Colombia Sources: Web sites of the national metrology institutions; authors'research. Note: Possible measurement areas are accelerometry, acoustic, dimensional, electrical, force, ow, hardness, humidity, magnetic ux, mass, optical and photometric, pressure and vacuum, radiological (including ionizing radiation), surface texture, thermal, time and frequency, vibration, viscosity, and volume. (figure 10.1).The scientific capabilities of an NMI are limited by the accuracy, precision, and range of its measurements, but also by the number of measure- ment areas it covers. Countries with technologically advanced and diversified industrial sectors require metrological services in more scientific areas. The most basic calibration services, which include dimensional, mass, force, and pressure measurements, are offered by all of the surveyed Latin American NMIs. The Ecuadorian Standardization Institute (INEN) offers services in these four areas as well as volume measurements, but none in other slightly more complex activities, such as flow and thermal measurement. Argentina, Chile, Colombia, and Peru have capabilities in a wider range of measurement areas than Ecuador, but these tend to be in technologically simple activities. Only NMIs in Mexico and Brazil are involved in the full spectrum of metro- logical activities found in the United Kingdom's and Korea's NMIs. Metrological Activities Calibration services There are important disparities in the number of calibration services offered by NMIs in different Latin American countries. Figure 10.2 shows the number of calibration services offered by the NMIs of different countries, standardized by manufacturing value added. The figure shows Peru with a 190 Quality Systems and Standards for a Competitive Edge Figure 10.2 Number of Calibration Services Offered by NMIs, Standardized by Manufacturing Value Added, 2005 0.8 per added 0.7 es vic value 0.6 ser 0.5 turing 0.4 calibration 0.3 manufac of 0.2 million 0.1 number US$1 0 Poland Peru Turkey Brazil Colombia Ar gentina Sources: Web sites of the national metrology institutions; authors'research. Note: 2003 data on manufacturing value added. much larger number of calibration services than Colombia, Argentina, or Brazil. One possible explanation for the very large number of calibrations displayed in some countries is that the private market for calibration is underdeveloped in those countries. In such cases the NMIs may be serving both the primary market (calibration laboratories) and secondary market (industrial users), while in other countries with more developed calibra- tion markets the NMIs are only serving the primary market. Veri cation There are deficiencies in the enforcement of legal metrology in certain countries in Latin America. While there were more than 12 million mea- suring instrument verifications in Brazil in 2004, there were only 1,500 in Peru and 50 in Colombia during that year. It is clear from these num- bers that the legal metrology infrastructure is at very different stages of development in different countries. While Brazil's system is well estab- lished, Colombia's is still in its infancy. International Cooperation and Recognition of Metrology Institutions International laboratory comparisons Latin American countries are involved in few international and regional programs for the comparison of laboratory measurement and calibration. Metrology in Latin America 191 Inter-laboratory comparison programs conducted at the regional level by the InterAmerican Metrology System (SIM) and at the international level by the International Bureau of Weights and Measures (BIPM) are the most widely accepted references for the technical capabilities of metrology laboratories. Without such references, it is difficult to assess the accuracy and precision of a laboratory's measurements. Latin Ameri- can countries have been involved in few inter-laboratory comparisons. This holds true for both the total number of comparisons and the aver- age number of comparisons for a given measurement area offered by the national laboratory. Mexico and Brazil, the Latin American countries that have participated in the most comparisons, participated in less than half as many as Korea between 1995 and 2005 (figure 10.3). Peru, Colombia, Chile, and Ecuador are the worst performers in the region, having only participated in two to five comparisons during that period, in contrast to 216 in the United Kingdom. Figure 10.3 International Laboratory Comparison Activities, Absolute Number and Standardized by Number of Measurement Fields Offered by the NMI, 1995­2005 250 14 eas ar 12 200 ement 10 omparisonsc omparisonsc y y or 150 8 or measur of 6 er-laborat 100 er-laborat number int int yb of 4 of ed 50 diz 2 number number standar 0 0 . of o Spain Brazil Peru Chile Kingdomea,Rep Mexic gentina Ecuador Ar Colombia ed Kor Unit number of international inter-laboratory comparisons, 1995_2005 number of international inter-laboratory comparisons, 1995_2005 divided by number of measurement areas in the NMI Source: National Institute of Standards and Technology International Comparisons Database, based on appen- dixes B, C, and D of the International Committee of Weights and Measures Mutual Recognition Arrangement. 192 Quality Systems and Standards for a Competitive Edge Table 10.4 Membership in International Organizations Country CGPM Metre Convention CIPM MRA OIML CITAC Argentina yes yes corr. yes Brazil yes yes yes yes Chile yes yes no no Colombia no no no no Ecuador associate yes no no Mexico yes yes corresponding yes member Peru no no corr. no Korea, Rep. of yes yes yes yes Spain yes yes yes no United Kingdom yes yes yes yes Sources: BIPM, OIML, and CITAC Web sites. Note: CITAC = Cooperation in International Traceability in Analytical Chemistry. Membership in international organizations At an institutional level, only Brazil is fully integrated in the international metrology system (table 10.4). Brazil is a member of the two main inter- national metrology organizations, the International Bureau of Weights and Measures (BIPM) and the International Organization of Legal Metrology (OIML), and of the Cooperation on International Traceability in Analyti- cal Chemistry (CITAC), and it has signed the CIPM MRA. While most other Latin American countries are members of the Metre Convention of the General Conference on Weights and Measures (CGPM), Brazil is the only country in the region that is a full member of the OIML. Colombia and Peru are the two most isolated countries in terms of membership in international organizations and MRAs. C H A P T E R 1 1 Existing Policies and Support Programs A number of support programs have been implemented in Latin American countries to facilitate the diffusion of standards and quality in the productive sector. These are the result of efforts by governments, international donors, and the private sector. Most programs aim to correct the specific market failures affecting small and medium enterprises (SMEs) and limiting their adoption of standards, many of which were mentioned in chapter 6. Brazil: Center for Quality, Safety and Productivity In Brazil, the Center for Quality, Safety and Productivity (QSP) is a non- profit organization created by a group of enterprises, academics, and con- sultants in 1991. Its objective is to help firms implement and improve their management systems and become sustainable. The QSP focuses on the main international and Brazilian standards, which include ISO 9001, ISO 14001, and OHSAS 18001. A $36 monthly membership fee provides firms with access to QSP benchmarking instruments, informa- tion on quality and certification, training manuals and interactive training guides, networking opportunities, mutual assistance schemes, and semi- nars. There are currently 135 member firms. In addition, QSP offers consulting and internal auditing services for the implementation of management systems and has assisted more than 400 firms in this area. The center also offers training in quality manage- ment practices and safety practices, including personalized courses for 193 194 Quality Systems and Standards for a Competitive Edge enterprises. Specialized assistance is available to SMEs. So far, more than 600 firms have participated in a specialized SME consulting program. A cooperation agreement with Banco do Brasil also provides low-cost financial services to SMEs participating in the training, consulting, and auditing programs. In this scheme, firms with revenues of up to $2.2 mil- lion are eligible for financing of up to 80 percent of project costs. Colombia: Quality and Environmental Management Program Much of Colombia's recent success with the diffusion of management system certification can be attributed to several support programs funded by the government, international donors, and private organizations. In particular, the Quality and Environmental Management Program (CYGA) has been cred- ited for playing an important role in increasing certification in Colombia.1 CYGA operated from 1999 to 2003 under the auspices of two domes- tic organizations: ICONTEC, the national standards organization, and SENA, the National Training Service.Approximately half of the $2.4 mil- lion program budget was funded through a grant from the Inter-American Development Bank. The objective of the program was to increase the competitiveness of SMEs by strengthening concern for quality assurance and environmental conservation. CYGA focused on introducing quality assurance systems based on ISO 9000 and ISO 14000 standards. The program was based on a holistic approach to certification and had four complementary components: 1. Raising SME awareness of buyer requirements 2. Training auditors and consultants in quality and environmental systems 3. Supporting the implementation and certification of quality and envi- ronmental management systems in SMEs 4. Providing marketing and know-how transfer to ICONTEC The first component consisted of conferences and training activities. Free conferences were organized in seven cities to sensitize SMEs to the advantages of implementing quality management systems. Courses on quality assurance and environmental management were offered to SME personnel in the different cities. SME candidates were selected for the training courses based on a number of requisites, including size, turnover, export potential, and willingness to participate in the program. The second component consisted of offering high-level technical train- ing for management system auditors and consultants (trainers for SMEs) Existing Policies and Support Programs 195 Table 11.1 CYGA: Participation and Subsidies Subcomponent 1999­2003 participation Subsidies (%) Internal auditor training 1,283 auditors 0 Consultant training 682 consultants 0 Enterprise training 503 enterprises 50 Consulting 526 enterprises 50 Pre-audits 337 enterprises 50 Audits 924 enterprises Not speci ed Source: CYGA. who would participate in later stages of the program (component 3). Auditors and consultants were expected to bear the full cost of the course, based on market price, the rationale being that they would recoup these costs through their participation in the SME advisory program. The third component provided financial support for participating SMEs to hire technical advisory services, training services, and pre- auditing in ISO 9000 or ISO 14000. Firms that had not participated in the initial training courses could still participate in component 3 if they demonstrated that they had at least one internal quality system auditor on their payroll. CYGA also subsidized the actual auditing and registra- tion process, conditional upon available funding from the program. Oth- erwise, CYGA had undertaken agreements with two commercial banks to provide loans for this process. Some details on the results of compo- nents 1 to 3 are shown in table 11.1. Component 4 consisted mainly of training the staff of ICONTEC and other related entities in developing technical assistance methods for SMEs and marketing their program. Other quality promotion programs in Colombia Besides CYGA, two other programs have been credited for contribut- ing to the success of certification in firms in Colombia. These are the National Program for Quality Assurance (PNAC) and the Colombian Fund for the Modernization and Technological Development of Micro, Small and Medium Enterprises (FOMIPYME). PNAC was designed by the National Training Service (SENA) to offer specialized training and technical assistance to exporting firms and firms with export potential to help them implement and become certified for ISO 9000, ISO 14000, and the Hazard Analysis and Critical Control Point (HACCP) food safety standard. SENA cofinances 50 percent of the value of the project, and the firm covers the remaining 50 percent. 196 Quality Systems and Standards for a Competitive Edge SENA contributed more than $4.7 million in cofinancing from 1999 to 2005. During that period, 561 firms participating in the program received certification, and 132 are in the process of becoming certified. Roughly 57 percent of the program participants are small firms, 37 percent are medium firms, and 6 percent are large firms. FOMIPYME was created by national legislation in 2000. Its objective was to cofinance programs, projects, and activities to promote the tech- nological development of SMEs and to provide technical assistance to these firms. FOMIPYME cofinanced up to 65 percent of the total value of each project. From 2001 to 2004, FOMIPYME financed 155 projects involving the implementation of a quality system. Of these beneficiaries, 72 percent were microenterprises and 28 percent were SMEs. Peru: CERTIpyme, PYMEvaluación, and PRECISO In Peru, there are three recent private sector initiatives to increase qual- ity control and certification in SMEs. Together, these programs provide various certification schemes that are tailored to the needs and means of small business. CERTIpyme aims to gradually increase the competitiveness of micro and small enterprises by taking them through various certifica- tion schemes. The program is operated by COPEME, a consortium of nongovernmental organizations dedicated to the development of SMEs; Quality Consulting, a quality management systems consulting firm; and CERPER, a private certification body. There are two certification levels that are designed to progressively take the firm toward a quality man- agement certification scheme. Level I, "organization certification," estab- lishes that a firm complies with all official requirements to operate in the formal sector. Level II, "process certification," includes the requirements from Level I and is based on a firm's productivity and its adoption of a basic quality system. The program includes all of the activities to support certification, including consulting, training, and auditing. The costs are $80 for Level I certification and $120 for Level II certification. SGS, an international certification body, operates two programs for SMEs. PYMEvaluación is a certification program in which SGS evaluates various aspects of a company's technical capacity and its ability to ful- fill customer requirements. These aspects include logistics management, commercial management, human resources, productive facilities and capacity, quality control, and financial condition. The objective of the evaluation is to provide SMEs with a means to evaluate their strengths Existing Policies and Support Programs 197 and weaknesses. The certificate, when obtained, can be used to enhance buyer confidence in a supplier. However, this certificate is tailor-made for the Peruvian market and is not widely recognized internationally. A second SGS program, PRECISO, is designed to offer low-cost ISO 9001:2000 certification to SMEs. Prices are adapted to the specific needs and structure of each firm. PRECISO is open to firms with fewer than 25 employees, and to firms of between 26 and 45 employees subject to a preliminary evaluation. Mexico: National Committee on Productivity and Technological Innovation In Mexico, the National Committee on Productivity and Technological Innovation (COMPITE) was established in 1997 as a nonprofit organiza- tion to promote productivity, quality, and social responsibility in SMEs. There are currently six COMPITE offices throughout Mexico that offer subsidized training, courses, workshops, and consulting services and orga- nize congresses. Although COMPITE offers services to firms of all sizes, micro and small enterprises constitute most of those receiving workshops and consulting services (table 11.2). COMPITE receives funding from the Ministry of the Economy to offer subsidized services for SMEs. In 2004 the ministry provided COMPITE with $1.74 million. SMEs and individual participants may only receive subsidies for one service from COMPITE annually, but they can combine this funding with other state or municipal funding. A number of courses are offered in the areas of ISO 9000, quality cul- ture, managerial skills, and productivity. These courses are designed for entrepreneurs, business employees, and consultants. Also, COMPITE's Table 11.2 Distribution of Enterprise Size in Two COMPITE Programs, 2004 Program Enterprise size Share of total participating firms (%) Workshops Micro 72 Small 19 Medium 5 Large 5 Consulting Micro 40 Small 39 Medium 15 Large 6 Source: COMPITE. 198 Quality Systems and Standards for a Competitive Edge Table 11.3 COMPITE Programs: Participation and Subsidies Program Participation,2001­04 Subsidies (US$) Training courses 20,698 enterprises Entrepreneurs: up to 50%, $885 maximum 291,812 hours Consultants: up to 70% 54,298 participants (individuals) Workshops 8,289 enterprises Microenterprise: up to 70%, $3,628 maximum 189,664 hours Small enterprise: up to 50%, $3,097 maximum Medium enterprise: up to 30%, $2,212 maximum Consulting 2,269 enterprises Microenterprise: up to 70%, $5,752 maximum 78,477 hours Small enterprise: up to 50%, $5,310 maximum Medium enterprise: up to 30%, $4,867 maximum Source: COMPITE Web site. four-day workshops provide quick solutions to problems related to man- ufacturing processes.2 Workshops are offered to individual firms, but are also available to groups of firms in the case of microenterprises. COMPITE's consulting services are designed to help SMEs implement the ISO 9000 quality management system. Consulting services are offered to individual companies, but also to groups of firms in the case of microen- terprises. Consulting services include a diagnostic service that establishes the degree of implementation of a quality system in the enterprise; an implementation consulting service, which consists of 10 hours per month of assistance in the implementation of a quality system; and revision of the quality system by a different consultant. The participation and subsidies of the training, consulting, and workshop programs are shown in table 11.3. Finally, COMPITE organizes an annual International Congress on Quality for Micro, Small, and Medium Enterprises. The objective of the congress is to promote an exchange of experiences and knowledge between SME entrepreneurs. The cost of the congress is subsidized for SME entrepreneurs and personnel. Attendance varies each year and has ranged from 35 to 1,800 SMEs in the past three years. Chile: Chilean Economic Development Agency In Chile, a central government agency, the Chilean Economic Develop- ment Agency (CORFO), has been supporting quality and certification through two funding instruments. The Technical Assistance Fund, estab- lished in 1992 for SMEs, and the Enterprise Management Support Pro- Existing Policies and Support Programs 199 gram, established in 1999 for large enterprises, have both provided grants to subsidize consulting services, including, but not restricted to, services for implementation of quality management systems. In 2004, CORFO contributed 3,505 million pesos to 2,188 enterprises through these pro- grams (CORFO 2005). Both of these programs are being phased out and replaced by a new funding instrument explicitly dedicated to quality. The Quality Promotion instrument consists of a grant to subsidize consulting services for SMEs wishing to implement a management system based on a verifiable or certi- fiable standard. Eligible consultants must be listed in the national registry of consultants maintained by INN, Chile's national standards institute. Firms may seek certification against standards such as ISO 9001, ISO 14001, NCh 2909 (the national standard for SME management systems), PABCO (Animal Premises Under Official Control), OHSAS (Occupational Health and Safety Assessment Series), and HACCP. The grant can be used for the entire process leading to the certification or verification, including the conformity assessment procedure itself. The grant covers 70 percent of the implementation costs and 90 percent of the certification costs for the NCh 2909 standard, and 50 percent for implementation and certification of ISO 9001, ISO 14001, OHSAS, and HACCP. The maximum disburse- ment depends on the standard against which the firm is being certified. Argentina: Standards and Quality Accreditation Program (ProCal) In Argentina, the objective of the Standards and Quality Accreditation Program (ProCal) is to improve the competitiveness of SMEs by pro- moting management, process, and product quality. This program, oper- ated through the Argentine Accreditation Bureau (OAA) and partially financed by a grant from the Inter-American Development Bank, aims to strengthen both the supply and demand of accreditation services. On the supply side, ProCal provides financial assistance to the OAA for train- ing and program development. On the demand side, ProCal provides financing to prepare testing and calibration laboratories to qualify for ISO/IEC 17025 accreditation. It also promotes secondary demand for accreditation by holding informative seminars for SMEs to explain the commercial advantage of testing, calibrating, and certifying products of accredited entities. The budget for the demand component of ProCal totals $1,077,000 over three years, of which roughly half is provided by the Inter-American Development Bank. C H A P T E R 1 2 Policy Recommendations and Conclusion This closing chapter presents recommendations for developing countries that are committed to implementing and modernizing effective national quality systems to facilitate trade, successfully integrate in the global economy, and enhance competitiveness. While this book has highlighted the structure and peformance of selected Latin American countries, the recommendations presented here apply not only to them but to all devel- oping countries. The objectives, themes, issues, and problems related to quality and standards are quite similar across countries. The main differ- ences tends to be in the countries' initial conditions. The chapter begins by setting forth a list of general principles and actions needed for successful reform. It next examines the process of con- solidating the national quality system and looks at each of four specific functions of the system in turn--accreditation; metrology; standardiza- tion; and certification, testing, and calibration. For each of these func- tions, recommendations are presented in three areas: the organization of the function, the type and structure of government support, and efforts to enhance international integration. General Principles for Implementation and Reform of National Quality Systems A successful national system of quality and standards that can effectively support trade and competitiveness initiatives will embrace a number of 201 202 Quality Systems and Standards for a Competitive Edge key principles and elements. Countries that set out to build such a sys- tem should · Ensure political commitment to the task. This commitment will be reflected in a clearly stated and integrated vision, in appropriate resource allocation, and in the development or reform of the right type of institutions. · Assign a single government agency or unit to be responsible for the coherence, coordination, and oversight of the national quality system and for proposing appropriate government interventions. It may be more effective to create a new agency by restructuring existing units with diverse or overlapping jurisdictions. · Involve the private sector from the start, and establish a clear under- standing of the relative roles and jurisdictions of the public and pri- vate sectors in this public-private undertaking. An advisory council on quality is one option, and often a best practice. · Make a special effort to integrate small and medium enterprises (SMEs) in the process of adopting all relevant types of standards. · Pursue a decentralized approach that brings service delivery as close as possible to the users and beneficiaries. · Offer a well-planned information and advocacy program to show the need for and benefits of the use and adoption of quality standards. Such a program should be targeted particularly to SMEs. · Provide adequate budget support and capacity to carry out the relevant functions assigned to the public sector. · Develop and nurture the necessary human capital through the training of personnel who will administer the quality services of the system. · Strengthen the international projection and recognition of the national quality system in selected countries, as references to establish bench- marking, levels of compatibility, competitiveness, and affectivity. This will help facilitate its recognition by other countries with similar sys- tems, through bilateral or multilateral agreements, so as to benefit domestic industry and avoid unfair competition and technical trade barriers. Steps should be taken to establish national procedures and processes for the accreditation of laboratories and certification institu- tions that bring international recognition and compliance. · Establish national reference systems for weights and measurements. · Legislate appropriate sunset provisions for the government to withdraw from certain roles and tasks as the system evolves and the private sec- tor can take over. Policy Recommendations and Conclusion 203 · Place an overall emphasis on facilitating competition, among domestic firms and between domestic and foreign firms, as a key motivator for the use and adoption of standards. · Observe the rule of law and property rights, including international property rights, in all cases. · Design and implement effective programs to increase awareness of the need for and impact of adopting modern quality management systems, and implement policies and incentives to build demand for services of metrology, norms, testing, and quality certification by private firms, organizations, and leading government institutions. Consolidating the National Quality System Most countries in Latin America and the Caribbean, and many else- where, have begun taking the first steps in the process of creating an effective national quality system. Many have an adequate--albeit not perfect--legal and regulatory framework of laws and decrees that defines the system of metrology, accreditation, normalization, and certification. A number of countries also have institutions at the national level with a formal mandate to carry out those tasks. Quite often, however, application of the established system of norms and standards is insufficient or flawed. The level of development of the activities and the capacity of the institutions and their international rec- ognition are highly variable and frequently low. In practically all coun- tries there are serious coordination problems and significant overlap in the functions and jurisdiction of the relevant institutions. This is indeed a grave problem with a large adverse impact, especially given the scarcity of public resources assigned to the national quality systems. In the pro- cess of building up and improving the supply of services in the system, it is critical not to lose focus on the need to tailor those services to the needs of industry. The first step is to carry out an extensive and systemic evaluation of the current normativity to see whether it meets the changing needs of the country in the current context of increasing globalization. In the process of updating the country's normativity, it is essential to set up a proper coordinating mechanism among institutions to remedy the sys- temic incoherence that is a problem practically everywhere. In a num- ber of countries, this has been accomplished through the creation of a national quality council comprising members of the different institutions with jurisdiction in specific areas of the national quality system. The suc- 204 Quality Systems and Standards for a Competitive Edge cess of such councils has been moderate. While the level of representa- tion has often been high--in some cases the council reports directly to the president or prime minister of the country--councils sometimes lack an executive secretariat and a unit with resources to implement policies and decisions and monitor progress toward their implementation. Accreditation Organization Ideally, there should be a single accreditation system in which the respon- sible institution, the national accreditation agency, is technically, opera- tionally, and financially independent. Where accreditation functions are spread over several institutions, firms incur additional costs because the requirements for accreditation typically vary across institutions. However, many countries do have more than one accreditation institution to be able to cover not only the industrial sector but also the phytosanitary, environ- mental, health, energy, and other sectors. In such cases it is important to establish the mandate, jurisdiction, and functional scope of each accredi- tation body and establish procedures for coordinating their actions. The system for accreditation of the certification institutions should have a clearly defined scope, which should be consistent with a stan- dard system of industrial classification and can be extended periodi- cally according to the needs of national industry. In some countries the scope of accreditation has remained limited to the quality system with ISO 9000 norms. In other countries it has been extended to include, for example, QS 9000 and ISO 14000 standards. The extension of accredita- tion to areas of increasing demand requires that appropriate regulations be issued to cover the new areas, that the structure and capacity of the national accreditation institution be enlarged and strengthened, and that proper training be given to the auditors. The operating norms of the national accreditation institution should follow international best practices. Since 2006, for example, the ISO/ IEC 17011 norm must be adopted by all accreditation institutions that seek international recognition. Governments should provide the financial resources and properly qualified staff needed to meet those institutional standards. Accreditation bodies in developing countries should incorporate pri- vate sector representatives in their governance structure. Some public accreditation bodies are mainly governed by public sector representatives or appointees, but the productive sector is often best placed to articulate Policy Recommendations and Conclusion 205 the needs of industry. Including private sector representatives in a general assembly or on consultative committees helps ensure that accreditation bodies reflect the views of the private sector, increasing the effectiveness of the national accreditation system. Some countries may need to increase the pool of technical staff used in accreditation activities. When an accreditation body has access to very few assessors capable of offering specialized technical services, the pace of accreditation and the range of sectors in which the accreditation body can work are limited. Accreditation bodies should provide training pro- grams to ensure that they have the necessary personnel. Also, their struc- ture should not be too rigid so that they can operate efficiently and adapt to changing market requirements. Government support In the early stages of developing an accreditation system, the national accreditation body should receive government support. Without an established customer base, the accreditation body will find it difficult to be self-sustainable through accreditation fees; in particular, it needs government financial support to train technical personnel. Once a critical mass of customers has been reached, the government should gradually phase out its financial contributions to the national accreditation body. Countries with insufficient resources should consider creating regional accreditation bodies. In some cases a country's economy may be so small that it is difficult to justify the establishment of a national accredita- tion body. Countries in Central America and the Caribbean, for instance, where there are few potential candidates for accreditation, should pool their resources and establish a regional accreditation body that is recog- nized in each economy. Countries with limited accreditation activities should investigate whether support is needed on the demand side or the supply side of accreditation. Countries may register very low numbers of accreditations for a variety of reasons. On the supply side, low activity could be due to the limited capacity of the accreditation body, which may not have the technical capacity to perform some accreditations or may not have enough trained personnel to step up the pace of accreditation. In general, Latin American and Caribbean countries are slow in delivering accredi- tations, and in such cases the accreditation body must invest in training more personnel. An underdeveloped accreditation market may also be due to a prob- lem on the demand side. In this case, there are several possibilities: 206 Quality Systems and Standards for a Competitive Edge (a) the conformity assessment market does not value accreditation by the national body because it is not well recognized; (b) there are not enough conformity assessment bodies to create adequate demand; or (c) there are technical and financial barriers to accreditation that potential customers are not able to cope with. In the case of a demand-side prob- lem, a government should create programs that diffuse information on the advantages of accreditation. It should also work to gain international recognition of its accreditation system and should provide technical assis- tance to organizations that wish to obtain accreditation. The national accreditation institution should diffuse information on accreditation to ensure that industrial firms understand the benefits of accreditation.The accreditation process is largely unknown in many Latin American countries. Currently, a number of ISO certificates and testing services in various countries are delivered by nonaccredited certification bodies. Promoting the benefits of accreditation will increase the demand for accreditation and reduce the risk of fraudulent or low-quality confor- mity assessment services. It would be useful to seek and gain recognition through the use of a single accreditation label or symbol that follows rules similar to those established by the main accreditation institutions in other countries. That label or symbol should include a single and unique accreditation number, allowing users to know the extent of accreditation and the date of accreditation. The government should also provide support for the accreditation of testing and calibration laboratories and for proficiency testing programs to ensure competition in these markets. In many countries the public sector holds a large share of these markets because of historical govern- ment support for its activities. New entrants will be able to compete in offering these services if they receive initial technical and financial sup- port for accreditation. In particular, proficiency testing programs, often required for many types of laboratory accreditations, are very expensive and demand a high level of technical expertise. Providing support for proficiency testing programs would help increase the number of accred- ited laboratories available in the country. As long as the national accreditation agency is not internationally rec- ognized--as is the case in many countries--the government should con- sider supporting the accreditation of certification institutions by entities that are internationally recognized. Governments should consider pro- grams offering partial financial support, such as matching grants, to help cover the high costs of international accreditation of the certification entities operating in the country. Certifications issued by duly accredited Policy Recommendations and Conclusion 207 entities will facilitate trade and lower its costs, allowing domestic pro- ducers to compete more effectively in external markets. International integration Developing countries should seek full membership in the main international accreditation organizations, the International Accreditation Forum (IAF) and the International Laboratory Accreditation Cooperative (ILAC), and should sign their mutual recognition agreements. A number of countries still have not joined these organizations.As a result, their accreditation systems can lose credibility at home and abroad, which impedes the development of a market for accreditation. Latin American countries' most important trade partners recognize the MRAs of both the IAF and the ILAC, and joining these treaties would significantly decrease technical barriers for Latin American exporters. Countries that have significant regional trade with signatories of the InterAmerican Accreditation Cooperation's (IAAC's) MRA should seek to join this agreement. Several of the large Latin American countries have already done so. Countries that face technical barriers in joining the ILAC or IAF agreements should join the IAAC agreement if this is more in line with their technical capabilities. At the same time, the IAAC should con- tinue its efforts to have its MRA recognized by the IAF and ILAC. Metrology Organization At the national level it is essential to have a single metrology institution that can coordinate the reference laboratories as well as the testing and calibration laboratories. In some countries the primary or secondary measurement stan- dards are offered by various institutions, allowing the provision of calibration services at moderate cost. To ensure that the development of measurement services keeps pace with the growth of industry, an institution should be established to coordinate the activities of the industrial calibration laboratories with those of the testing laboratories and facilitate the linkage (traceability) of such measures to the primary measurement standards. The coordination of actions and strategies among public and private institutions will help ensure strictly impartial behavior and build technical and scientific capacity. Countries should ensure that their national metrology institutions (NMIs) are not associated with activities that introduce conflicts of interest. There are a few cases in which the NMI is part of an institution that conducts activities in the areas of standardization, intellectual property registration, and accreditation. This can threaten the impartiality and credibility of the 208 Quality Systems and Standards for a Competitive Edge NMI. Countries where a single institution conducts work in these areas should create clear financial and administrative barriers between the differ- ent functions and ensure that the decision-making process is transparent. Current regulation, especially regarding legal metrology, needs to be updated. Countries should ensure that their metrology framework reflects the current national and international environment. In some countries, the national calibration system is barely outlined in applicable regulations. Developing countries should accredit the laboratories in their NMIs. An internationally recognized laboratory accreditation will increase the quality and credibility of calibration services provided by the NMI. This should be especially useful in countries where the NMI has not estab- lished a large customer base, or where the country is preparing to join the International Committee of Weights and Measures (CIPM) MRA. Government support Policies to strengthen the existing metrology system should first identify deficiencies in the system. This should include an evaluation of the mea- surement instruments and other metrology equipment that may become necessary in the future. The national metrology system requires a signifi- cant investment program and thus competes with other areas that require public financing. It is therefore essential to set priorities according to the current needs of the productive system and its potential for expansion in external markets. It is most important to do a cost-benefit analysis of new investments in domestic laboratories to support new metrology fields or to establish regional or local legal metrology centers. Countries not able to provide traceability to the productive sector in enough technical areas should collaborate with other countries to invest in new metrological infrastructure.Although the larger countries have devel- oped technical capabilities for a wide range of measurement areas, NMIs in smaller countries only offer measurement services in a limited num- ber of spheres. Those countries should consider expanding the number of measurement services by engaging in collaborative arrangements with neighboring countries and specializing in certain measurement areas. Specialized human capital is needed to increase the metrology capacity of the country and guarantee precise measurements and a supply of reli- able services. Accurate measurements depend not only on well-equipped laboratories but also on the presence of competent and well-paid staff. Scientific personnel must have specialized training and research experi- ence. Likewise, technical staff have to be prepared to construct testing modules and to maintain measurement instruments. Policy Recommendations and Conclusion 209 International integration Developing countries should participate in more international inter- laboratory comparisons to expand the international recognition of their measurements. Latin American countries currently participate in far fewer comparisons than countries in the developed world. As manu- facturing quality becomes more important for exports, Latin American countries need to provide their domestic industries with a national infra- structure that allows for proper traceability of measurements in the most relevant economic sectors. Some developing countries remain internationally isolated and should seek membership in the main international metrology organizations and work toward signing an MRA. NMIs can increase the credibility of their metrology infrastructure and receive technical support by joining inter- national metrology organizations and signing the MRA of the CIPM. Without an MRA in metrology, countries may face technical barriers to trade that hinder exports and the development of the productive sector. Once the CIPM MRA has been signed, the scope of the measurements it covers should be progressively extended. Developing countries should become more internationally integrated in legal metrology to further reduce technical barriers to trade. Without harmonized means and procedures for verifications and tests, it is dif- ficult to establish whether metrological control is equivalent in differ- ent countries. Membership in the International Organization of Legal Metrology (OIML) helps countries harmonize policies regarding trade in products and services with a commercial value based on measurements, as well as trade in measuring instruments. International cooperation in metrology should be strengthened. In recent years international technological cooperation and technology transfer in scientific and industrial metrology has played a major role in the start-up of the national metrology system in many countries. That type of collaboration should be continued, making use of the funds that often exist in developed countries for the implementation of research and development projects in this area. Standardization Organization Every country should ensure that the private sector is included in the governance of the standards bodies and that its participation is balanced. In general, the standardization process is open and transparent in Latin 210 Quality Systems and Standards for a Competitive Edge America, but there are some exceptions to this rule. In some countries the agency that issues the standard is biased toward a particular sector; as a result, the overall interests of society suffer. In other countries, such as Peru and Ecuador, standards bodies are subject to excessive govern- ment influence and lack proper private sector representation in their governance structure through a general assembly of members or at least through a consultative committee. The participation of the private sector in determining the overall objectives, strategy, and structure of the stan- dards body will enable the institution to more closely serve the needs of industry and will increase transparency. Active participation of scientists and technologists trained as standards specialists should be emphasized. Consultative committees and stan- dards bodies do not have strong enough ties to science and technology institutions. Even when the law requires the participation of academic specialists, their participation frequently does not affect the relevant standard. Academia has demonstrated an increasing interest in standards, but there are currently few incentives to encourage such involvement. Moreover, there are no training programs for standards specialists. Most people learn about standards development by actually participating in the process, which can be an inefficient way to learn. Countries should ensure that they are dedicating sufficient resources to standardization. Standards development requires a number of admin- istrative personnel to coordinate all of the technical committees, review the standardization process, interact with international organizations, and ensure the diffusion of standards in the economy.There are large discrepan- cies in the amount of resources dedicated to standards development across countries. Some countries, for example Peru, have very few staff and a very small budget for standardization relative to the size of their economies. Whenever standards bodies raise revenues through commercial activi- ties, the government should ensure that these activities are administra- tively and financially separated from its standardization activities. This will help prevent conflicts of interest in the normal operation of the national standards body. Also, it will be useful to examine the financing mechanisms that other standards bodies use to raise revenues through standardization activities. Standards bodies in other regions raise more income than those in Latin American countries through sales of standards publications and membership fees. These sources of revenue should be increased to make the standards bodies less dependent on certification and testing activities, because revenue from the latter may decrease once there is more competition in the market. Policy Recommendations and Conclusion 211 Standards development activities and international integration Developing countries should increase their standards adoption rates while removing obsolete standards. Latin American countries are not increas- ing their standards stocks as fast as developed countries, even though their standards stocks are small. Of course, the objective should not be to produce as many standards as possible, because inadequate standards can actually inhibit growth, but to produce as many useful standards as pos- sible. The low number of standards in countries such as Ecuador, Chile, and Peru seems to indicate potential for substantially increasing the size of the standards stock. Countries should ensure that mandatory standards do not account for a growing share of standards. Although mandatory standards are on the decline in many countries, there are countries in which their share of total standards has actually increased in the past 15 years. Too many obligatory standards can place constraints on the productive sector that stifle innovation and prevent industries from becoming competitive. Countries should systematically review their national standards to reduce and update their standards stock. A mechanism to conduct cost-benefit analyses of mandatory standards has not been developed in most Latin American countries. As a result, the review of standards often occurs without any information about their impact and becomes no more than an administrative exercise. Developing countries should combine parallel approaches to upgrade their standards toward international standards. Many Latin American countries have already made remarkable progress in the adoption of international standards in the past 15 years. Countries should select a strategy of unilateral upgrading toward international standards in sectors where the domestic technological conditions are adequate and there is a high potential for conducting trade with countries outside the region. In sectors where standards upgrading would be technologically infeasible or would place countries at a regional competitive disadvantage, countries should follow a gradual, coordinated approach and adopt regional stan- dards. This approach has the most potential inside regional trade areas such as Mercosur and the Andean Community. Creation of regional standards bodies in Latin America (and in other regions) would not only facilitate coordinated standards upgrading but would enable countries to pool their limited resources for standardiza- tion activities. Developing standards can be expensive in terms of time as well as financial and human resources. Regional standards bodies would allow Latin American countries, especially the smaller economies, to 212 Quality Systems and Standards for a Competitive Edge divide standardization tasks among countries with different comparative advantages and to accelerate the standards development process, espe- cially in areas of common regional interest. Developing countries should increase their participation in the devel- opment of international standards. Toward that end, their governments should support membership in more International Organization for Stan- dardization (ISO) and International Electrotechnical Commission (IEC) technical committees and subcommittees. Few international standards originate from Latin American countries, or from the countries of any other developing region. If developing countries do not contribute to the development of standards used in global trade, the countries that domi- nate the process, mostly developed countries, may impose unfavorable specifications with which developing countries will then have to comply to access markets. Latin American countries should consider forming col- lective blocks to influence the development of ISO and IEC standards. Governments should support participation of the private sector in international standardization activities. High travel costs can make it dif- ficult for firms, especially SMEs, to represent national interests in ISO or IEC technical committees. Private sector participation in these organiza- tions should be supported by government funding. Supporting the Demand for Certification, Testing, and Calibration Most developing countries should increase their adoptions of quality and environmental system standards.The multiplicity of factors affecting ISO 9000 diffusion should provide a case for carefully assessing cross-coun- try differences before evaluating a country's quality management perfor- mance. Some low certification rates may simply reflect market factors, such as industry structure, and may not reflect a poor quality culture as such. The economic environment is responsive to government policy but should also be closely examined before policies are prescribed, because a deficiency in one factor, such as education, cannot always be compen- sated for by addressing another factor, such as quality awareness. Nonetheless, the governments of developing countries should con- tinue supporting quality management certification programs for SMEs. Although ISO 9000 certification is starting to become widespread in the manufacturing sector in many countries of the region, this is not the case for SMEs. Programs targeting SMEs have proven to be successful, even in countries with unfavorable economic environments. Technical and finan- Policy Recommendations and Conclusion 213 cial support should be focused on SMEs because they are usually less aware of quality issues and face more barriers than larger firms. Finan- cial support should not be exclusively limited to the registration costs because these account for only a small share of total certification costs. Support covering training, technical assistance, and even upgrading of equipment and facilities should also be considered. In general, programs that stimulate demand--through matching grants or low-cost loans that allow firms to freely choose providers--have been more successful than programs that subsidize one or several privileged providers. Governments should also support quality management certification in the service sector. Enterprises and organizations in the service sector have adopted far fewer certificates on a per-firm basis than those in the manufacturing sector. However, in many Latin American countries the service sector provides valuable input to the manufacturing export sector and should not be neglected. Countries should promote the availability of lead auditors and support the creation of a national registry of quality management consultants and auditors. There are relatively few lead auditors with internationally rec- ognized credentials in Latin America. The quality of certification services in the region depends on the quality of the auditing staff. In addition, it is difficult for firms and organizations to identify skilled consultants who can help them prepare for certification. A national registry or a business association of consultants could promote the development of this market and improve quality adoption in firms. It is useful to have a national-level registry of firms that have obtained process or product certifications. Such a registry, if used by consumers and major buyers in the public and private sectors, can facilitate trade by enhancing the reliability of firms and products. Governments should lead by example, adopting good quality practices in the ministries and in related government institutions and agencies. A program to foster the improvement of institutional quality should be launched at the highest level of government, with a realistic but ambi- tious calendar and with the proper assignment of resources. Such a pro- gram would help develop a culture of quality and efficiency in the public sector. Moreover, to facilitate the development of quality management and adoption of quality practices in public sector procurement, a train- ing program for personnel responsible for implementation at the national and subnational levels should be implemented. Governments should implement programs that provide incentives for providers and suppliers to adopt quality practices. Cooperation 214 Quality Systems and Standards for a Competitive Edge among large buyers and their suppliers is rare in many countries in Latin America. Governments should establish programs that encour- age large firms to provide incentives to their suppliers to adopt quality improvement measures and to develop groups or a registry of quality- minded suppliers. Firms within the club work together for the develop- ment of quality processes required by the large firm, with the assistance of a quality consultant, the cost of which is shared by the members of the club. To address coordination failures and take advantage of economies of scale, governments should support the strengthening of the quality infra- structure. In most countries, most firms do not have their own proper facilities for testing, metrology, and calibration, and the availability of external facilities is very limited. In many sectors and regions the existing market demand does not provide sufficient incentive for independent firms to set up testing and calibration services and cover their costs. How- ever, the industrial development of some regions requires the availability of such services, so it becomes necessary to assist their creation. This can be done either by providing the services directly in universities or public laboratories or by providing support to consortia of firms, industrial asso- ciations, or chambers of commerce or industry to set up and administer the services. When demand reaches the level where provision of the ser- vices becomes financially sustainable, public support should be phased out and the services run on a purely commercial basis. Finally, it is necessary to systematically advocate and communicate to firms the critical impact that quality and quality management can have on competitiveness and business success, particularly in terms of increasing exports.This is all the more true for SMEs. Large companies-- especially subsidiaries of multinationals and joint ventures, both national and foreign--usually have ample know-how and opportunities for tech- nology transfer and adoption of quality systems. Large firms also tend to have easy access to the services of the national quality system. Medium companies, on the other hand, tend to depend on external funding sources for securing quality standards, metrology, calibration, testing services, and training. Small companies usually have limited information and resources and often do not place high priority on adopting quality standards. Many managers of SMEs in developing countries perceive quality-- and the processes related to the system of norms, certification, testing, and calibration--as an additional burden and not as a source of poten- Policy Recommendations and Conclusion 215 tial cost savings and competitive advantage. That attitude inhibits the investment in laboratories and capital equipment and in improved qual- ity processes and also reduces investment in relevant training. A main reason for that attitude is that the clients of SMEs seldom demand com- pliance with specific quality standards and norms. Most consumers in developing countries respond primarily to price and only secondarily to product quality. Similarly, governments as buyers of goods and services often do not demand that their suppliers meet and comply with relevant quality standards. Periodic information campaigns should be undertaken to explain quality concepts and techniques to SMEs, with references to domestic companies that have adopted best practices and experienced good results. Such a demonstration effect can have a significant impact on firm behavior. Governments should systematically enforce the requirement that firms comply with the relevant quality standards and practices to be eli- gible to participate in any public procurement process or auction. Such a policy should be phased in on a reasonable timetable to give firms the chance to qualify and comply with the new requirements. In the case of all these recommendations, the budget requirements and timetable for implementation will vary according to initial condi- tions. Some of the recommendations concern regulatory or organiza- tional changes that can be implemented in most countries without large resource allocations and without delays because they are not tied to the cycle for approval of the annual national budget. Their implementation can probably be completed in a relatively short period, less than a year. However, recommendations related to improving the supply of services in the areas of metrology, norms, testing, and quality system management do require new resources. Thus, it is essential to go through an exercise of identifying industry needs and country priorities before undertaking such efforts. Last, recommendations related to advocacy--to promote awareness of quality adoption, to increase firms' demand for such ser- vices, and to secure international recognition of the institutions of the national quality system--will require a longer timetable and should be incorporated in medium-term national strategies. In all cases it is essential to ensure periodic oversight and evaluation of the progress and impact of the policies and programs throughout the implementation period. This will allow for midcourse adjustments to ensure that the policies and programs are effective and fulfill their objectives. 216 Quality Systems and Standards for a Competitive Edge Conclusion As increased competition among developing countries in labor-intensive manufactures erodes economic returns, higher-quality markets and high- value goods are increasingly important to maintaining dynamic com- petitive advantage. Globally integrated production networks, typically governed by buyers from developed nations, have raised competitiveness to the top of developing countries' policy agendas. Countries need to offer the high-quality products demanded by consumers and global sup- ply chains and deliver them to markets to meet just-in-time production schedules. The reform of quality systems has been one of the key missing pillars of the reform program in most developing countries. The history, at least in most Latin American countries, of operating as closed economies for so many years is partly to blame. Yet most countries have realized or are beginning to realize that the rules of the game in this new, highly globalized context have changed and that these changes are likely to be irreversible. In response to this new environment, many reform-minded governments have launched initiatives to enhance competition and trade and to make exports a key engine for growth. Providing for a balanced and effective quality infrastructure has to be an integral part of that pro- gram. An effective and coherent national quality system is essential if a country is to access markets, particularly external ones, and become competitive. The pace of reform is accelerating in a number of countries around the world; other countries cannot afford unnecessary delays in their reform programs, lest they fall further behind. The findings presented here suggest that a well-executed reform pro- gram and an effective national quality system, consistent with interna- tionally accepted and required practices, would have a quite significant impact, particularly in terms of improving trade and access to markets and overall competitiveness. Moreover, there are a number of mostly pos- itive by-products and spillovers. Improvements in quality, standards, and so on are often entry points into innovation and technological improve- ments, an area where most countries, particularly in Latin America and the Caribbean, are lagging behind. Thus, initiatives to upgrade quality could help jump-start or accelerate the push for innovation. Moreover, when properly targeted, quality improvement programs for SMEs can facilitate their mainstreaming into value and export chains and in this way improve the livelihood of small producers that are so important to developing countries. Policy Recommendations and Conclusion 217 If reforms to enhance quality and adopt standards are to be effective, they must be implemented coherently and broadly. Most such reforms are uncontroversial and relatively easy to implement, and the resource requirements are quite manageable. The potential benefits for economic growth and poverty alleviation are significant and sustainable, enabling quality and standards reforms to win wide public support. A P P E N D I X A An Evaluation of Mexico's National Quality System Mexico's institutional quality framework has four basic elements: · A standardization system through which activities in the public and private sectors are regulated. There are standards related to health; the environment; consumer safety; commercial information; and indus- trial, labor, and trade practices, among other areas. In Mexico, official standards are generated by 11 secretariats and the process is overseen by the General Bureau of Standards (DGN) in the Secretariat of the Economy. · A metrology institute that maintains appropriate measurement stan- dards at the national level. In Mexico, this is done by the National Metrology Center (CENAM). · A national accreditation system to accredit bodies that assess conformity-- testing labs, calibration labs, certification bodies, and verification or inspection bodies. The Mexican Accreditation Body (EMA) performs this function. · Certification, testing, calibration, and inspection bodies, collectively known as conformity assessment bodies, that certify businesses in aspects relating to metrology and quality. These four elements are closely linked. Calibration laboratories ensure that the measurements performed by testing laboratories and inspection bodies are reliable. Calibration laboratories themselves must demonstrate 219 220 Quality Systems and Standards for a Competitive Edge the accuracy and precision of their instruments, based on their traceability to reference measurement standards held at CENAM, the national metrol- ogy institute.The competence of calibration and testing laboratories and of certification and inspection bodies must be evaluated by a national accred- itation body, the EMA. All aspects of these activities rely on standards. Standards provide the basis for evaluation of all conformity assessment bodies and define the requirements against which conformity assessment is performed. In sum, conformity assessment provides the vital link between standards and the products, processes, and services themselves. Standardization Standards in Mexico can be mandatory (normas oficiales mexicanas), ref- erential, or voluntary (normas mexicanas). The Federal Metrology and Standardization Law was passed in 1992 and reformed in 1997 to accom- modate the North American Free Trade Agreement (NAFTA). Manda- tory (official) standards can be issued by 11 federal government agencies and reference standards can be issued by the Federal Electricity Commis- sion (CFE) and PEMEX, the state-owned oil company. As of June 2005, there were 800 official standards in force and 198 more in progress (table A1). At that time also, there were 5,651 voluntary standards in place and 150 in progress. In principle, anyone can propose a voluntary standard, which will then be submitted to one of 30 technical standards commit- Table A1 Official Mexican Standards by Agency, June 2005 Agency In force In progress SE 110 20 Sectur 8 0 Sagarpa 139 9 Semarnat 107 15 STPS 38 3 Sener 87 23 Sedesol 0 0 SS 195 24 SCT 115 104 Segob 1 0 SSP 0 0 Total 800 198 Source: CENAM. An Evaluation of Mexico's National Quality System 221 tees (the committees are validated by the DGN and supported by eight national standards bodies) for review. There are a number of problems with the standards institutional framework. Standards that require physical or chemical measurements frequently do not include adequate consideration of measurement characteristics--for example, traceability and uncertainty--to produce the required level of confidence. As a result, the standards can become nonviable. This is especially the case with voluntary standards. Consultative committees and standards bodies do not have strong enough ties to science and technology institutions. Even when the law requires the participation of academia, this participation frequently does not affect the relevant standard. Academics have demonstrated an increasing interest in standards, but there are currently few incentives to encourage their involvement. There are no training programs for standards specialists. Most people learn about standards development by actually participating in the pro- cess. This lack of training programs creates inefficiencies in the process. In some cases, the agency that issues the standard can be biased toward a particular sector. As a result, the overall interests of society suffer. A mechanism to conduct cost-benefit analyses of standards has not been developed. As a result, the mandatory review of standards every five years often occurs without any information about the impact of stan- dards, and the review becomes merely administrative. Finally, the DGN does not have sufficient resources to carry out all of its mandated functions. There is a tendency on the part of both produc- ers and sectoral authorities to generate standards that can serve as trade barriers. DGN's role is to prevent this, but the agency is not effectively performing that function. Metrology and Calibration Mexico has developed a robust set of measurement standards, but there are some gaps. CENAM has developed 63 national measurement stan- dards to date, which are used as references to ensure that all measure- ments in the country are uniform, reliable, and in line with international standards. The National Nuclear Research Institute (ININ) is in charge of another three standards. Although these standards collectively cover a large part of the measurement requirements of the country, there remain some areas that require development. One example is the measurement of high gas flows in view of the increasing importance of natural gas.Another 222 Quality Systems and Standards for a Competitive Edge is the production of certified reference materials; these have a long way to go. Unfortunately, budgetary limitations make it difficult for CENAM to respond flexibly and rapidly to new measurement challenges. Calibration involves determination of the relationship between an instrument's input and the magnitude or response of its output. It serves to establish the accuracy and precision of a measuring instrument. Sec- ondary calibration laboratories constitute the primary market for cali- bration from the national institute. They calibrate their own equipment to the national institute's and diffuse their measurements to the down- stream market for calibration, which includes industrial producers, test- ing laboratories, inspection bodies, research laboratories, universities, and other final users. Many conformity assessment bodies require that equip- ment and measurement reference systems be calibrated, or traceable, to other widely accepted metrological references before they issue a prod- uct or system certificate. Mexico's metrology and calibration services serve as a public good, and the country needs to develop them further. Because the costs to develop high-level metrology services significantly outweigh the income received from providing such services in Mexico, public funding is necessary. How- ever, low-level metrology services can be attractive to the private sector. In general, the higher the level in the chain of calibration services, the greater the involvement of the public sector because profitability decreases as one moves up the chain (figure A1). In Mexico, this chain of services is not yet fully developed because there is a shortage of participants in different segments, and there are technical calibration deficiencies. In March 2005, 299 calibration labs were registered with the EMA. Secondary calibration services have insufficient capacity in Mexico. As mentioned above, secondary metrology and calibration services (one step down from CENAM) are not profitable, due to the technical require- ments of labs, specialized equipment, and highly trained personnel. Sec- ondary laboratories in Mexico include research center labs, centers in the Figure A1 Metrology Services Chain in Mexico secondary tertiary 6th-level CENAM calibration calibration calibration labs labs labs greater public participation greater private participation Source: CENAM. An Evaluation of Mexico's National Quality System 223 SEP-CONACYT network overseen by the Secretariat of Public Educa- tion and the National Council on Science and Technology, academic labs (instrument centers of the National Autonomous University of Mexico and the National Polytechnic Institute, among others), the materials test- ing lab of the CFE, and other private sector labs. The secondary service capacity is adequate in some areas, such as elec- tricity, while more development is required in others, such as volume and firmness. Geographic coverage of secondary services is also lacking.About 44 percent of calibration labs are located in Mexico City, Jalisco, and Nuevo León (figure A2). Currently, some fifth- or sixth-level institutions are forced to calibrate their measurements directly with CENAM because there are no secondary or tertiary labs in their area that are accredited or have the necessary level of measurement accuracy and precision. This substantially increases the time and costs of the calibration procedures, causing consumers to suffer and ultimately eroding competitiveness. More inter-laboratory comparisons and proficiency tests are needed. To have a reliable Mexican metrology system, inter-laboratory compari- sons and proficiency tests with calibration labs that are integrated in the traceability chain are required. This is a common practice in all developed countries; it consists of appointing a recognized, prestigious lab or regional Figure A2 Concentration of Calibration Labs in Mexico City, Jalisco, and Nuevo León impact 100 reference materials 90 humidity 60 torsion norms 60 density 57 hardness 50 volume 50 dimensional 44 mass 44 temperature 43 pressure 41 force 41 time and frequency 40 flow 36 electric 30 acoustic 0 optics 0 viscosity 0 0 25 50 75 100 % of national total Source: CENAM. 224 Quality Systems and Standards for a Competitive Edge organization to circulate between different labs to test their measurements against known standard values. In Mexico, this activity is not sufficiently established. Inter-lab comparisons and proficiency tests occur only infre- quently, meaning that in some cases the performance of accredited labs is not reliable. More comparisons are required to maintain the confidence of third parties and comply with international standards. Participation in international inter-laboratory comparisons is a require- ment for maintaining membership in mutual recognition agreements (MRAs). An MRA in metrology provides for the formal recognition of national measurement standards and calibration capabilities. Through measurement comparisons, an MRA establishes the degree of equivalence of national measurement standards maintained by national metrology institutes and thereby guarantees acceptance by the other MRA signato- ries of measurement results that are traceable to the national metrology institute. Mexico is a member of international accreditation organiza- tions such as ILAC (International Laboratory Accreditation Cooperation) and APLAC (Asia-Pacific Laboratory Accreditation Cooperation), which require the EMA to demand that laboratories participate in national and international inter-lab comparisons and aptitude tests to be accredited. Policies to strengthen the existing Mexican metrology system should do the following: · Identify deficiencies in the metrology system, including an evaluation of the measurement instruments and other metrology equipment nec- essary for the future. Once detected, priorities to correct the deficien- cies will have to be determined. · Create an interinstitutional body to unite the Mexican metrology sys- tem and make uniform the criteria that apply to legal metrology. · Prepare a legal metrology framework that reflects the current national and international environment. · Develop the national calibration system, which is barely outlined in current applicable regulation. Accreditation Accreditation is defined as the procedure by which an authoritative body gives formal recognition that an organization or person is competent to carry out specific tasks. Accreditation is sought on a voluntary basis as a proof of competence in a given area. Accreditation provides certification and inspection bodies, as well as testing and calibration laboratories, with An Evaluation of Mexico's National Quality System 225 a means to signal that they are conducting their work to appropriate stan- dards and that they are able to provide reliable services to support quality in firms. The accreditation body evaluates the personnel and supporting management system of the candidates for accreditation and can request practical tests for laboratories when relevant. These tests take the form of proficiency testing schemes through which the measurement results of dif- ferent laboratories are compared (that is, inter-laboratory comparison). The Mexican Accreditation Body is the first and, so far, the only pri- vate organization in Mexico with responsibility for accrediting organiza- tions that assess conformity. The EMA accredits testing and calibration labs and certification, verification, and inspection bodies. Before the cre- ation of the EMA, the DGN had responsibility for accreditation. Reforms to the Federal Metrology and Standardization Law in 1992 and 1997 opened the door for a private, third-party organization to take over these responsibilities. EMA began operating in January 1999. Mexico's market size and subsidies to the EMA have discouraged the emergence of competitors. The EMA is a nonprofit organization whose principal financing sources are fees for accreditation services, dues from associates, use of the EMA brand, training courses, and donations from federal and state agencies and from regional and international organiza- tions. While most other developed countries have a single public institu- tion in charge of accreditation, Mexico has adopted the U.S. accreditation model in which a number of private accreditation bodies fiercely com- pete on the basis of price and service quality. However, subsidies from the public sector give the EMA a significant advantage over potential com- petitors; hence it has remained a monopoly. It is unlikely that another accreditation body will emerge because of Mexico's market size, sizeable investment requirements, and opposition from the EMA. High accreditation costs have led to unaccredited competitors. Costs associated with accreditation from the EMA are substantially higher than U.S. and international prices, and this has led to the emergence of testing establishments that are not accredited but compete with EMA- accredited labs. Such testing establishments are authorized or recognized by different agencies, including the Secretariat of Agriculture and the Secretariat of Health. As a result, they act as de facto accredited labs. The EMA's mix of public and private origins has led to a complex governance structure and slow decision making. The board of directors comprises four sectors, each of which has nine votes: the federal govern- ment (various secretariats); business (represented by various groupings of private sector organizations); users and clients (labs, certification and 226 Quality Systems and Standards for a Competitive Edge inspection bodies); and academia and consumer organizations. Given that these four sectors often have conflicting interests, the decision-making process can be slow, and there have been delays in reaching agreements that would ultimately benefit the end users, that is, businesses that need conformity assessment services. However, the involvement of the differ- ent sectors leads to greater transparency and impartiality. Besides the board, the EMA has a general assembly, executive com- mission, executive management, and 11 departments. The departments include certification, inspection, testing and calibration labs, technical and relationships, administration, systems, quality assurance, new proj- ects, operations, promotion and diffusion, and human resources. The organization's size and complexity is another deterrent to the emergence of potential competitors. To minimize technical barriers to trade, national conformity assess- ment procedures should be harmonized across countries. Complying with a standard or a technical regulation is only useful if compliance can be demonstrated to the buyer or the government at reasonable cost. Demonstrating compliance through conformity assessment itself is only useful if the testing and certification requirements are similar in the exporting country and the importing country. If testing laboratories are not recognized abroad, tests on products carried out in an export- ing country have to be repeated by a recognized laboratory in each of the importing countries. An adverse test report in the importing country can result in the rejection of an entire shipment. Likewise, if certifica- tion in one country is not recognized abroad, domestic firms requiring quality system and environmental management certification for export purposes--for example, ISO 9000 and ISO 14000 registration--need to be certified by organizations in each of the importing countries. To pro- mote the recognition of conformity assessment procedures, many accred- itation bodies base their organizational structure and their accreditation decisions on well-recognized international guides and standards. Mutual recognition of accreditation procedures promotes trade by decreasing transaction costs and eliminating technical barriers. When accreditation in one country is recognized by other countries, the work per- formed by that country's certification bodies, inspection bodies, and cali- bration and measurement laboratories will be accepted in other countries. The EMA has been recognized by the main international accredi- tation bodies, IAF and ILAC. Membership in these two organizations enhances an accreditation body's prospect of gaining international cred- ibility. Accreditation bodies must demonstrate that they operate at high An Evaluation of Mexico's National Quality System 227 international standards to join the IAF or ILAC. In this sense, IAF and ILAC membership is a form of assurance that accreditation bodies are competent to undertake their work and are not subject to conflicts of interest. Furthermore, these international organizations facilitate tech- nology transfer in areas related to quality assessment and provide a forum for learning from other experienced accreditation systems. The EMA is a signatory to various MRAs. For full recognition, accredi- tation bodies must establish agreements with other countries, based on mutual evaluation and acceptance of each other's accreditation systems. Membership in an MRA at the bilateral level (with one other country) or at the regional or international level (with many other countries) is critical to guaranteeing the credibility of the national conformity assess- ment system. MRAs are based on peer evaluation processes through which signatories evaluate each other's compliance with the agreed-upon requirements and evaluate the performance of assessment staff. They usu- ally cover specific accreditation types (for example, quality management system certifiers) or products. Mexico has signed the IAF's Quality Man- agement System (2001), Environment Management System (2004), and Product (2004) multilateral recognition agreements. The EMA needs to take the following steps in the short term: · Obtain international recognition for the accreditation of laboratories and verification units, with the aim of having tests conducted in Mexi- can labs (country of origin) recognized in the destination country. · Fulfill the new ISO/IEC 17011 standard. Beginning in 2005, the new 17011 standard was applied to all accreditation entities in the world, and its completion became obligatory in 2006. · Increase domestic promotion of the EMA. The accreditation process is largely unknown in Mexico, hence priority should be given to promot- ing the importance of accreditation throughout the country. Strategic alliances with all three levels of the government and with the private sector in all regions will be necessary to ensure compliance with the appropriate standards. · Broaden and strengthen the EMA's structure so that it can provide accreditation services in areas of high future demand. Certification, Testing, and Inspection Third-party certification consists of the provision of assurance by an independent body that a product, service, system, process, or material 228 Quality Systems and Standards for a Competitive Edge conforms to one or more standards or specifications. For many standards, especially quality and safety standards, the incentives for self-enforcement are low and producers have much to gain by claiming that a noncompli- ant product or process adheres to a certain standard. Furthermore, the highly technical content of some standards may render it difficult for producers to know whether they have appropriately complied with a standard. If there is no means of differentiating products that conform to a standard from those that do not, standards are of limited use. Certification provides benefits for producers of goods and services, consumers, and government regulators and for international trade. Manu- facturers and service providers can have their products or their manage- ment systems certified to particular standards to distinguish themselves from less reputable suppliers. Buyers benefit from certification because it allows them to compare and differentiate products and services in terms of quality, safety, or other desirable characteristics. It avoids the costs of having each buyer independently validate the characteristics of the prod- ucts and services of a supplier and reduces the risk of purchasing faulty products or services. By reducing search costs for global suppliers, certi- fication reduces technical barriers to trade. Certification by third-party organizations is increasingly included in trade contracts (Schuurman 1997). Finally, regulators benefit from certification because it provides them with a system to enforce governmental health, safety, and envi- ronmental legislation. In March 2005, there were 38 certification bodies accredited by the EMA. Testing and inspection help demonstrate that a product or process satisfies certain technical requirements. Testing involves determining the characteristics or performance of a product or process according to a specified procedure. Inspection is another form of assessment that relies on less sophisticated instruments than does testing. Independent testing laboratories and inspection bodies can be contracted by a firm to obtain a test or inspection certificate as evidence that a product or process con- forms to certain characteristics. In certain cases, testing and inspection are required for firms to implement a quality control system, such as ISO 9000. The first testing labs in Mexico were created roughly 35 years ago, and by March 2005 there were 362 accredited testing labs and 791 accredited inspection bodies. There are internationally accepted standards guiding the operation of certification entities, testing laboratories, and inspection bodies.1 To play a credible role in the conformity assessment system, testing laboratories and inspection bodies must display many of the same characteristics as An Evaluation of Mexico's National Quality System 229 Figure A3 Concentration of Testing Labs in Mexico City, Jalisco, and Nuevo León electric 47 food 42 metal/mechanical 39 chemistry 27 phytosanitary 21 0 20 40 60 % of national total Source: CENAM. certification bodies, notably impartiality, objectivity, and confidentiality. Objectivity relies heavily on the procedures guiding the evaluation pro- cess, the equipment used, and the skills and qualifications of staff. There is a high geographic concentration of testing labs and inspec- tion bodies in Mexico. Nearly 40 percent of Mexico's testing labs are located in Mexico City, Jalisco, and Nuevo León (figure A3). This has translated into price differences between services in the big metropolitan zones and those in the rest of the country. Inspection bodies are also concentrated, with 50 percent of them located in Mexico City, Estado de México, Jalisco, and Nuevo León. The only exception is inspection bodies associated with energy efficiency and natural gas, which are linked to CFE and PEMEX distribution locations. The concentration of inspec- tion bodies creates particular problems in areas such as auto transport, industrial parks, hydraulics, animal health, telecommunications, tourism, and others, where 100 percent of inspectors are located in the above four regions while demand is spread throughout the country. Prices for conformity evaluation services are relatively high. Given that there are relatively few accredited firms in relation to demand, prices are essentially determined in a monopolistic fashion, with price differentials based on type of client and service. Normally travel costs are paid by the client, which puts businesses located outside of Mexico City, Guadalajara (capital of Jalisco), and Monterrey (capital of Nuevo Léon) at a disad- vantage. Even though accredited domestic firms consider themselves to be cheaper than international competitors within Mexico, conformity evaluation prices are three to four times greater on average than prices in the United States and the European Union. High costs are linked to accreditation costs for certification and inspection bodies. 230 Quality Systems and Standards for a Competitive Edge Conformity evaluation bodies are regulated by several organizations.Cer- tification, inspection, and testing bodies are regulated by EMA, CENAM, and the DGN through the accreditation process. A great number of busi- nesses are also certified by the Standards Board and other entities such as ANCE (Association for Standards and Certification, a domestic group) and AENOR (Spanish Association for Standardization and Certification). There are several problems in Mexico's conformity evaluation process. First, current legislation allows people who are being evaluated to be part of the accreditation decision-making process and to be responsible for their own evaluation activities. This creates an inherent conflict of inter- est. The situation was justified by the lack of qualified people when the accreditation system was initiated, but now there are plenty of experts from academia, research centers, and CENAM. Hence, the functional separation of the evaluators and those evaluated is viable and necessary. Second, those responsible for evaluation activities often do not have suf- ficient technical competence in a given specialty, meaning that their results lack credibility. This weakness has been identified and is being addressed. Finally, federal agencies have not been very active in demonstrating the conformity assessment criteria for the standards that they issue. Demand for conformity assessment services is relatively narrow, mean- ing that Mexico has a low number of ISO certificates. Demand has been limited to the following four sectors: · Mexican exporters that need a recognized certificate of quality for their products · Foreign firms that export products to Mexico · Providers of goods and services to the public sector in Mexico · Giant firms and firms listed on the Mexican stock market This universe encompasses a very small share of the total formal busi- nesses in the country. For example, only 34,000 export-oriented com- panies are officially registered, which represents barely 2 percent of the country's total. One manifestation of this is the relatively low, albeit increasing, number of ISO 9000 certificates in Mexico (figure A4). Mexi- co's ISO 9000 certificates as a share of the Latin American and Caribbean region's total--roughly 17 percent in 2004--has not grown appreciably over the last decade (figure A5). As a reference, Mexico's share of total GDP in the region was roughly 30 percent in 2004. Mexico's certificates as a share of the world total increased during the 1990s but then stag- nated during the early 2000s (figure A6). An Evaluation of Mexico's National Quality System 231 Figure A4 ISO 9000 Certificates, 2004 140,000 132,926 120,000 es 100,000 84,485 tificat erc 80,000 of 50,884 60,000 48,989 40,972 37,285 40,000 26,654 number 27,101 12,558 20,000 17,365 3,391 0 France y o Italy es China Japan Spain Stat India exic Kingdom M ed German Australia ed Unit Unit Source: ISO 2004. Figure A5 Mexican ISO 9000 Certificates and Share of Latin American Total 4,000 25 Mexico certificates as a share 3,500 of Latin American total (%) es 20 3,000 otalt tificat 2,500 15 erc 2,000 of American 1,500 10 Latin 1,000 number 5 of 500 % Mexico ISO 9000 certificates 0 0 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 Source: ISO 2001, 2003a, 2005. 231 232 Quality Systems and Standards for a Competitive Edge Figure A6 Mexican ISO 9000 Certificates and Share of World Total 4,000 0.6 3,500 Mexico certificates as a share 0.5 es of world total (%) 3,000 0.4 tificat 2,500 otalt erc 2,000 0.3 orld of w 1,500 of 0.2 % 1,000 number 0.1 500 Mexico ISO 9000 certificates 0 0 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 Sources: ISO 2001, 2003a, 2005. Note: The apparent drop in the number of ISO 9000 certi cates in Mexico in 2003 is due to the transition from the old ISO 9000:1994 standard to the new ISO 9000:2000 standard. In the 2003 survey, the ISO 9000:1994 stan- dard was o cially replaced by the ISO 9000:2000 standard and Mexico did not report ISO 9000:1994 certi cates in the survey, only ISO 9000:2000 certi cates. However, in 2003 both ISO 9000:1994 and ISO 9000:2000 certi - cates were reported in many other countries in the survey, thus giving the impression that there was a drop in the relative number of certi cates in Mexico in 2003. The conformity evaluation culture may be less widespread in Mexico than elsewhere due to high market concentration and informality. Rela- tively high concentration in many goods and services markets means that there are few firms in each sector that have an incentive to get certified-- either to compete in international markets or to deter the entrance of new competitors. The country's large informal sector means that there is no potential demand for certification or testing from a large portion of the economy. A P P E N D I X B An Evaluation of Turkey's National Quality System This appendix describes and evaluates the state of Turkey's national qual- ity system and its implications for the diffusion of standards and quality in the productive sector. It discusses the concept of a national quality system and its structure in Turkey; the adoption and diffusion of quality standards in Turkey, from both the supply and demand perspectives in the certifi- cation market; the accreditation activities and infrastructure, along with their impact on the diffusion of quality in the private sector and European Union (EU) integration; the state of Turkish metrology; and finally, the standards development system in Turkey, again in the context of EU inte- gration and international trade. InTurkey,firm-leveldatashowstatisticallysignificantrelationshipsbetween quality certification and both exports and employment. There is not much information about quality variables in the Turkey Investment Climate Survey. Manufacturing firms are asked in general terms whether they have a quality certification. The econometric results reveal that quality certification has a positive effect on employment and on the probability of being an exporter (Escribano et al. 2006). More specifically, it is estimated that surveyed firms with quality certification employ on average 44.8 percent more permanent workers than firms without certification.1 Because the estimates control for general size categories (large, medium, and small firms) and for age, this result implies that among Turkish firms of similar size and age, those that are con- cerned enough with quality issues to invest in quality certification end up employing more personnel, probably as a consequence of higher demand for and appreciation of their products. Figure B1 shows the percentage of 233 234 Quality Systems and Standards for a Competitive Edge Figure B1 Contribution of Investment Climate Variables to Average Employment by Turkish Firms other control real wages variables 2% productivity 10% 13% labor skills 11% innovation and information and communication technology quality 5% 3% finance and red tape, corruption, corporate crime, informalities, governance and legal issues 12% 44% Source: Turkey Investment Climate Survey. contribution to average log-employment of each set of investment climate variables.2 Most categories incorporate a set of several variables, whereas the quality category only contains information on quality certification. Still, this single variable contributes 3 percent to average employment. Quality certifications also help increase the probability that nonex- porting manufacturing firms will become exporters. More specifically, firms with quality certifications have a 6 percent greater probability of being exporters than other firms. Certified suppliers provide buyers with a signal that they are able to produce goods and services of consistent qual- ity, and thus they increase their chances of being selected by the market as exporters. The percentage contribution of quality certification among all other investment climate variables to the probability of becoming an exporter is, however, modest--only 1 percent, as figure B2 shows. A country's national quality system is supported by a multidimensional infrastructure of organizations whose activities help to evaluate whether a product, process, or service fulfills specified technical requirements.3 The evaluation activities include the actual conformity assessment proce- dures, which consist of any combination of testing, inspection, calibration, and certification and differ according to the product or process. Certifica- tion bodies, inspection bodies, testing laboratories, and calibration labora- tories evaluate the characteristics, quality, and performance of products and processes in enterprises and their conformance to specific standards. Accreditation bodies are used to assess the competence of these confor- mity assessment bodies and to increase the public's and trade partners' confidence in their services. Metrology institutes ensure the accuracy and An Evaluation of Turkey's National Quality System 235 Figure B2 Contribution of Investment Climate Variables to a Turkish Firm's Probability of Becoming an Exporter productivity other control 14% variables 23% labor skills infrastructures 5% 15% innovation and information and communication technology quality 4% 1% finance and red tape, corruption, corporate crime, informalities, governance and legal issues 2% 36% Source: Turkey Investment Climate Survey. precision of the measurements transmitted by the calibration laborato- ries to other conformity assessment bodies and to enterprises. Finally, standards bodies develop the technical specifications used to define the requirements against which conformity assessment is performed. Turkey has, in principle, all of the necessary institutions for a complete national quality system: · A metrology institute. The National Metrology Institute (UME) is responsible for scientific metrology. It was created in 1986 as part of TÜBITAK (Scientific and Technical Research Council of Turkey), a public agency. · A standardization system. The Turkish Standards Institution (TSE) is the national standards body and is responsible for the development of all voluntary standards. It is an independent public organization gov- erned by private law and operates according to an Establishment Law (Law 132) created in 1960. Government ministries are responsible for the development of mandatory standards and sometimes use TSE standards as a basis for these mandatory standards. · A national accreditation system. The Turkish Accreditation Agency (TÜRKAK) is the national accreditation body. It was established by law in 1999 and started its operations in Ankara in 2001. TÜRKAK is a public autonomous agency affiliated with the Prime Ministry and governed according to private law. It is responsible for accrediting certification bodies, inspection bodies, and laboratories according to 236 Quality Systems and Standards for a Competitive Edge Box B1 Conformity Assessment and Access to the EU Market The quality infrastructure plays an important role in European integration. To participate in the EU free market and implement the acquis communautaire, a country needs a conformity assessment infrastructure adapted to harmonized EU legislation. The Old Approach and the New Approach are the two strategies used for technical harmonization in the EU. Member states are obliged to transpose the directives related to the Old and New Approaches into their legislation. The Old Approachdirectivescontaindetailedtechnicalspeci cationsforindividualprod- ucts. Under the New Approach, legislative harmonization is limited to essential requirements, which are stated in general terms and mostly concern the areas of safety, public health, consumer protection, and environmental protection. The development of technical speci cations necessary for the implementation of New Approach directives is requested from the di erent European standards bodies (CEN, CENELEC, and ETSI, the European Telecommunications Standard- ization Institute). These "harmonized standards" retain their status as voluntary European standards but national authorities are obliged to recognize products manufactured according to these standards as conforming to the essential legal requirements of the directives. Since 1985, the New Approach has been the har- national and international standards. Moreover, since 2003 it has signed protocols with a number of ministries to assess the qualifications of conformity assessment bodies applying to become notified bodies (see box B1 for the role of conformity assessment in EU integration). · Certification, testing, calibration, and inspection bodies. There are a number of certification and inspection bodies, as well as testing and calibration laboratories operating in both the private and public sec- tors. TSE also hosts all of these services and holds most of the market in certification and testing. Certification, Testing, and Inspection Adoption of quality system standards in Turkish rms Relative to the number of firms and organizations with ISO 9000 certifi- cation, Turkey lags behind other Organisation for Economic Co-operation and Development (OECD) countries. In Turkey, 80 percent of ISO 9000 An Evaluation of Turkey's National Quality System 237 monization method used for most industrial products. In 2001, 36 percent of manufactured goods (by value) exported from Turkey to the EU were covered by either the Old Approach or the New Approach (Hoekman and Togan 2005). The Global Approach to certi cation and testing establishes European Com- munity policy on conformity assessment. The CE marking on a product symbol- izes conformity to all the obligations required by the applicable New Approach directives. In Turkey, most products that are covered by New Approach direc- tives can be CE-marked through self-declaration of the manufacturer. However, for some products, third-party conformity assessment and delivery of the CE marking must be conducted by noti ed bodies. These bodies must be nomi- nated by member states and accepted by the European Commission. Along with international standards, regional standards represent a growing share of the standards stock of many countries. In Europe, CEN, CENELEC, and ETSI develop regional voluntary standards. The standardization process in these organizations is based on consensus and is similar to that followed by their international counterparts. In Europe, the standards bodies produce European standards, or ENs. Mem- ber organizations are obliged to adopt European standards and withdraw con icting national standards. CEN has an agreement with ISO through which many of the standards are developed in common. As a result, about 30 percent of European standards are identical to ISO standards (CEN 2004). certificates belong to firms in the manufacturing,wholesale and retail trade, real estate, and rental sectors. Comparing the number of certificates to the number of enterprises in these sectors shows that Turkey lags behind all other OECD countries except Poland (figure B3). There are only three ISO 9000 certificates for every 1,000 establishments in Turkey, six times fewer certificates than in Spain on a per-firm basis. Figure B4 shows that Turkey performed below the level predicted by OECD averages. By contrast, if only manufacturing is considered, Turkey performs much better than predicted by OECD averages in ISO 9000 adoption. ISO 9000 diffusion in Turkey is much more prevalent in manufacturing than in other sectors. Only three other OECD economies surpass Turkey with respect to ISO 9000 adoption on a per-firm basis in manufacturing (figure B5). Turkey also performs much above its predicted level of ISO 9000 adoption based on OECD averages (figure B6). In manufacturing, certification is mostly fueled by exports to the EU, where ISO 9000 is often a buyer or regulatory requirement. 238 Quality Systems and Standards for a Competitive Edge Figure B3 Number of ISO 9000 Certificates as a Percentage of the Number of Firms in Manufacturing, Wholesale and Retail Trade, Business Services, and Real Estate in Selected OECD Countries 3.5 3.0 firms 2.5 of % 2.0 as es 1.5 tificat 1.0 erc 0.5 0 . . y y g RepelandSpain Italy Rep es vak Ir ech Hungar GermanFinland KingdomAustria FranBelg ce iumtugal . of. Por Sweden StatTurkeyPoland Denmark ed Slo Cz ea,Repembour ed Netherlands Kor Lux Unit Unit Source: ISO 2005; OECD 2005. Note: Number of establishments used for Korea and Turkey; number of enterprises for others. Certi cate data are December 2004; enterprise data are 2002 (except United States, 2001). Certi cate data include nancial intermediation; enterprise data do not. Figure B4 Number of ISO 9000 Certificates vs. Number of Enterprises in Manufacturing, Wholesale and Retail Trade, Business Services, and Real Estate in Selected OECD Countries 12 Italy log Czech Spain 10 es, Switzerland Rep. Hungary United Austria Korea, 8 Rep. of States tificat Slovak Ireland Finland Poland erc Rep. BelgiumTurkey 6 Denmark ISO of 4 Luxembourg 2 number 0 2 3 4 5 6 7 8 9 thousands of enterprises, log Sources: ISO 2005; OECD 2005. Note: Number of establishments used for Korea and Turkey; number of enterprises for others. Certi cate data are December 2004; enterprise data are 2002 (except United States, 2001). Certi cate data include nancial interme- diation; enterprise data do not. Australia, Greece, Mexico, and New Zealand omitted. An Evaluation of Turkey's National Quality System 239 Figure B5 Number of ISO 9000 Certificates as a Percentage of the Number of Manufacturing Firms in OECD Countries 30 25 firms 20 of % as 15 es 10 tificat erc 5 0 e . . of y g y . Italy CanadaRepFranSwedenHungar ortugal ce y es ium GreecerlandelandurkeyRep Ir T Rep SpainJapan wa FinlandedStat vakea, Nor Belgingdom Polandeland P Ic Switz embour K GermanAustriaech Denmark Slo Cz Kor Netherlands Lux ed Unit Unit Sources: ISO 2005; OECD 2005. Note: Number of establishments used for Canada, Japan, Korea, and Turkey; number of enterprises for others. Certi cate data are December 2004; enterprise data are 2002 (except United States, 2001). Figure B6 Number of ISO 9000 Certificates vs. Number of Manufacturing Firms in OECD Countries 12 United log, Italy Kingdom Japan 10 es Rep. of Korea Spain Switzerland United 8 Czech Rep.States Portugal tificat GreeceNorway Turkey erc Ireland Austria Belgium Poland 6 Slovak Finland ISO Rep. Denmark of 4 2 number Iceland 0 0 1 2 3 4 5 6 7 thousands of enterprises, log Sources: ISO 2005; OECD 2005. Note: Number of establishments used for Canada, Japan, Korea, and Turkey; number of enterprises for others. Certi cate data are December 2004; enterprise data are 2002 (except United States, 2001). Australia, Mexico, and New Zealand are not included. 240 Quality Systems and Standards for a Competitive Edge ISO 9000 adoption rates are increasing faster in Turkey than in the rest of the world. There are now about 5,000 ISO 9000 certificates in Tur- key (figure B7). While Turkey accounted for 0.18 percent of ISO 9000 certificates in the current group of EU member, accession, and candidate countries in 1993,4 its share increased to 1.63 percent in 2004. During the 1993­2004 period the number of certificates grew, on average, by 62 percent in Turkey, versus 22 percent in the EU-25 group of member countries. Turkey's four largest manufacturing export sectors have very high ISO 9000 adoption rates. In fact, these adoption rates are higher than in similar sectors in Spain. Figure B8 displays the number of certificates per establishment in manufacturing sectors that account for a large share of exports in Turkey. Turkey's performance is better or comparable to that of Spain in sectors that are important for its exports. Romania only outperforms Turkey in two sectors where Romanian exports play a more important role. ISO 9000 adoption rates among smaller firms still lag far behind adop- tion rates among larger firms in Turkey. The 2005 Turkey Investment Cli- mate Survey found that large manufacturing firms were more than twice as likely to be certified as small firms. While 24.2 percent of small firms and 45.4 percent of medium firms were certified, 64.1 percent of large firms were certified.5 This pattern can be found in most other countries as well. Turkey's KOSGEB (Small and Medium Industry Development Organization) currently offers technical assistance programs for quality certification in small and medium enterprises (SMEs). Figure B7 Growth of the Number of ISO 9000 Certificates in Turkey 6,000 1.8 1.6 ISO 9000 certificates in urkeyT 5,000 Turkey 1.4 ountriesc in EU ISO 9000 certificates in es 4,000 1.2 in Turkey as % of total 1.0 es certificates in EU member, tificat 3,000 accession, and candidate erc 0.8 tificat countries of 2,000 0.6 erc 0.4 1,000 0.2 otalt number of 0 0.0 % 1992 1994 1996 1998 2000 2002 2004 2006 Sources: ISO 2001, 2003b, 2005. An Evaluation of Turkey's National Quality System 241 Figure B8 Number of ISO 9001 Certificates per Number of Establishments in Turkey, Spain, and Romania in Different Manufacturing Sectors 70 3.5 % 60 3.0 as GDP es 50 2.5 of % tificat 40 2.0 as erc establishments 30 1.5 of of valuet 20 1.0 number number 10 0.5 expor 0 0.0 ts ts ts and ed ts and ts, co y etc. and tiles produc oduc plasticoduc fibersement,er, pr Tex M and , chemical tile & fabricatprElectricalequipmentand productobacachiner equipment ts,andete, c, plast, paper, produc tex metal Food and produc lime Pulp paper metal optical erage, Rubber Concr Chemicals Basic bev Spain ­ ISO 9001 certificates per number of establishments Turkey ­ ISO 9001 certificates per number of establishments Romania ­ ISO 9001 certificates per number of establishments Spain ­ value of exports (% GDP) Turkey ­ value of exports (% GDP) Romania ­ value of exports (% GDP) Sources: ISO 2005; UNCTAD 2006; UNIDO INDSTAT 4 (Industrial Statistics Database), Rev. 3, 2005. Note: Turkey export data are 2003; Spain export data, 2002; ISO 9001 data, 2004; establishment data 2000. The supply of certi cation services Although the market for certification services in Turkey shows a reason- able number of suppliers, the competence of some has been questioned and their practices may be anticompetitive. Of the estimated 82 certi- fication bodies in Turkey, only 19 have been accredited by the national accreditation body, TÜRKAK. Although some other Turkish certification bodies are accredited in Europe, many have no accreditation, and this affects the quality of some of the services in the market. As of March 2002, up to 700, or 18 percent, of certificates in Turkey were delivered by nonaccredited bodies.6 Moreover, some local stakeholders, both national and foreign, question the business ethics of some of the smaller certifica- tion bodies active in the Turkish market, some of them subsidiaries of European bodies accredited in their home countries. They appear to be performing poor audits and granting certificates quickly and cheaply, and they may not be adequately monitored by foreign accreditation bodies.7 Many of the clients of these certification bodies appear to be services 242 Quality Systems and Standards for a Competitive Edge companies seeking to satisfy certification requirements for government procurement purposes. It is difficult for certification bodies accredited by TÜRKAK or by other reliable foreign accreditation bodies to compete under these circumstances. Turkey still does not have any notified bodies for the certification of products subject to the New Approach directives of the EU. To partici- pate in the EU free market and implement the body of EU law known as the acquis communautaire, Turkey and other EU countries are obliged to recognize products manufactured according to certain essential legal requirements set out in the directives. In particular, they need to harmo- nize their legislation by adopting the technical specifications contained in the New Approach directives. Products covered by the directives must be "CE-marked" by a designated entity called a notified body to prove that they satisfy such requirements. These bodies must be nominated by member states and accepted by the European Commission. The Turkish government has selected three notified bodies on the basis of an assessment by TÜRKAK, the national accreditation body, but the European Commission has not accepted the proposal and has been con- ducting its own assessment of the candidate notified bodies since 2004.8 A final decision has not yet been made. As is evident from the number of notified bodies in other EU countries, Turkey will need to notify more bodies to serve the needs of its economy (figure B9). Until now, Turkish firms have had to use the services of the subsidiaries of organizations noti- fied in other European countries.Two European subsidiaries share most of this market for CE marking. The presence of domestic notified bodies in Turkey would increase supply for CE marking services and thus increase competition and expand the market to other parts of the country. Figure B9 Number of National Notified Bodies Recognized by the EU in Selected Countries, February 2006 120 112 100 80 notified of 60 52 47 bodies 40 20 20 20 9 number 2 0 0 0 0 y y . Spain oland Rep P German Hungar vak LithuaniaAustraliaTurkeyBulgariaRomania Slo Source: European Commission, Nando Information System Web site. An Evaluation of Turkey's National Quality System 243 Policy recommendations The Turkish government should continue supporting quality manage- ment certification programs. SMEs have lower certification rates than large firms. Continued government support articulated through matching grants and administered by entities closely related to the private sector will ensure that certification becomes more diffused among SMEs. The overall support framework should be designed so as to replicate schemes with the highest impact, avoid overlap among providers, promote col- laboration between the implementing institutions and the funding minis- tries, streamline administrative hurdles throughout the financial process, and neutralize the risk-averse fund allocation approach usually taken by implementing institutions. Special attention should be dedicated to sup- porting quality management certification in the services sector. Service sector enterprises and organizations have adopted far fewer certificates on a per-firm basis in Turkey than in other OECD countries. The services sector provides valuable inputs to Turkey's manufacturing export sector and should not be neglected. The Turkish government should support training and technical con- sultancy services for conformity assessment bodies wishing to become notified bodies. Notified bodies are essential to the integration of Turkey in the EU because they enable Turkish products to comply with EU leg- islation and provide access to the European market. The Turkish government should promote the creation of an organi- zation for quality management consultants. It is difficult for firms and organizations to identify skilled consultants to prepare them for certifica- tion. A business association of consultants could promote the develop- ment of this market and improve quality in firms. Government procurement rules for ISO 9000 certification should ensure that certificates from bodies with questionable practices are not recognized. The government is an important driver of certification and should refuse certificates issued by bodies that do not conform to ethical practices. Accreditation Accreditation activities in Turkey Accreditation activities in Turkey remain rather limited, especially in product certification. TÜRKAK accredits organizations according to the most widely used and recognized European and international standards.9 Currently, 89 organizations have been accredited (figure B10) and 80 244 Quality Systems and Standards for a Competitive Edge Figure B10 Number of Organizations Accredited by the National Accreditation Body in Selected Countries, March 2006 500 ed Turkey Hungary 450 445421 Spain Bulgaria edit 394 cr body 400 Poland ac 350 300 editation cr 250 ac ganizations 200 163 or 150 36 132 of 99 60 107 national 100 70 yb 50 32 4323 39 21 25 34 19 9 2 2 14 2 5 133 2 13 153 number 0 testing calibration system product personnel inspection laboratories laboratories certification certification certification bodies bodies bodies bodies Source: Authors'research. applications are in process. Despite its large economy, Turkey has accred- ited far fewer conformity assessment bodies than many other European countries, except in the case of system certification bodies (for process standards). Most worryingly, TÜRKAK has only accredited two product certification bodies (for product standards) since the start of its opera- tions; these are in the areas of cement and organic foods. In relative terms, when value added in manufacturing and services is used to standardize the comparisons of accredited institutions, Turkey ranks in the last posi- tion among the selected comparator countries in all areas except system certification and inspection (figure B11). The number of accreditations has increased in Turkey since 2001, but the rate of increase has stagnated in the past two years. The num- ber of annual accreditations has risen from four to 32 since the start of TÜRKAK's operations in 2001, but there was no growth in accreditation activities between 2004 and 2005 (figure B12). TÜRKAK estimates that it will have accredited 400 laboratories by 2010, about the same number as currently are accredited in Spain or Poland. However, at the current rate, only 150 laboratories will be accredited by 2010, falling short of the infrastructure necessary to support high-quality products for exports. The demand for TÜRKAK's accreditation services has been low. Few of the current potential accreditation candidates have sought to be accredited by TÜRKAK. Of the estimated 82 system certification bod- ies in Turkey, only 19 are accredited. Of the estimated 50 calibration An Evaluation of Turkey's National Quality System 245 Figure B11 Number of Organizations Accredited by the National Accreditation Body in Selected Countries, Standardized by Value Added in Manufacturing and Services, March 2006 Testing Calibration System certification laboratories laboratories bodies 1,200 25 7 1,000 20 6 800 5 15 4 600 10 3 400 2 200 5 1 0 0 0 y y y SpainTurkey SpainTurkey Spain Bulgaria Poland Poland Hungar Bulgaria Hungar HungarBulgariaPolandTurkey Product certification Personnel certification Inspection bodies bodies bodies 8 2.0 140 7 120 6 1.5 100 5 80 4 1.0 3 60 2 0.5 40 1 20 0 0 0 y y y BulgariaPoland SpainTurkey SpainTurkey Hungar BulgariaPoland Hungar BulgariaSpainPolandTurkeyHungar Source: Authors'research. Figure B12 Number of Accreditations Granted Each Year by TÜRKAK 35 30 25 testing laboratories calibration laboratories editations 20 cr personnel certification bodies ac inspection bodies of 15 product certification bodies 10 system certification bodies number 5 0 2002 2003 2004 2005 Source: TÜRKAK Web site. 246 Quality Systems and Standards for a Competitive Edge laboratories, only 21 are accredited. Many certification bodies do not seek accreditation from TÜRKAK because they are accredited by foreign signatories to the multilateral recognition arrangement of the European Cooperation for Accreditation (EA MLA), accreditation that is recog- nized both in Turkey and abroad. Calibration laboratories, which need calibration certificates from the National Metrology Institute in order to become accredited, are discouraged from applying by the high prices that UME charges to provide the required traceability of measurements. The institutional structure of accreditation The state plays a large role in the governance of accreditation in Tur- key. The public sector is represented to a much greater extent than the private sector in the national accreditation body, holding more than half the seats in the general assembly and on the board of directors. This is not the norm in other countries with more mature accreditation systems, where the private sector plays a more important role in the governance of the accreditation body (table B1). TÜRKAK's Establishment Law (Law 4457) is too constraining and does not provide TÜRKAK with sufficient administrative autonomy. This law creates a rigid organizational structure by establishing the dif- ferent service units and administrative departments within TÜRKAK. The law also prescribes the staff titles and number of employees, includ- ing a maximum limit on the number of staff. The law does not allow TÜRKAK to adequately reorganize itself according to market demand or to improve the efficiency of its services to the private sector. TÜRKAK Table B1 Organization of Accreditation in Selected Countries Share of executive Executive council members director Legal General Consultative appointed by appointed by Country status assembly council government (%) government Bulgaria public yes yes 20 yes Chile private no no 100 no Hungary public yes yes 33 no Mexico private yes no 28 no Poland public yes yes 100 yes Spain private yes yes 0 yes Turkey public yes yes 71 yes Source: Authors'research. An Evaluation of Turkey's National Quality System 247 also has a large number of nontechnical staff relative to the size of its operations. Turkey has one of the largest accreditation bodies in Europe in absolute terms and relative to the number of accredited bodies (figure B13). It appears that the administrative and personnel structure imposed on TÜRKAK by its Establishment Law requires more permanent staff than are necessary for Turkey's current accreditation market. TÜRKAK depends on high service fees to remain financially sustainable. In 2005, TÜRKAK's budget was TRY 3,800,000, equivalent to about $2.5 million and comparable to the budgets of the Polish and Chilean national accreditation bodies (figure B14). The budget was financed almost entirely (95 percent) by accreditation fees. TÜRKAK's Establishment Law states that it should receive assistance from the general budget, but it has not received this funding from the government since 2004.Young accreditation bodies such as TÜRKAK typically need government support to develop their customer base.To remain financially sustainable,TÜRKAK is charging service fees that are much higher than the norm in EU accession and can- didate countries (figure B15). This can act as a disincentive for conformity assessment bodies to seek accreditation from TÜRKAK. Until recently, Turkish accreditations had not achieved a high degree of international recognition. Unlike all other EU member states and OECD countries, Turkey is not a member of the International Accredi- tation Forum (IAF). However, Turkey is a full member of the EA and was accepted to the EA MLA for quality management systems, testing, calibration, and inspection in April 2006. Up to that date, TÜRKAK's lack of international recognition placed important constraints on Turkish Figure B13 Number of Accreditations per Permanent Staff Person in the Accreditation Body, Selected Countries 40 36.7 35 30 editations 25 22.0 cr person ac 20 15.5 of staff 14.1 14.0 15 per10 number 5 2.6 0 o Chile SpainMexic PolandBulgariaTurkey Source: Authors'research. 248 Quality Systems and Standards for a Competitive Edge Figure B14 Budget of the National Accreditation Body, Selected Countries 3.5 3.0 editation cr 2.5 ac millions) 2.0 (US$ 1.5 national of 1.0 body 0.5 budget 0 y Poland Turkey Chile Hungar Bulgaria Source: National accreditation bodies. Note: 2005 data except Chile, 2002. Figure B15 Average Cost of an Accreditation, Selected Countries 14,000 12,000 (US$) 10,000 8,000 editation cr 6,000 ac of 4,000 ostc 2,000 0 Turkey Hungary Bulgaria Poland system certification body testing laboratories Source: National accreditation bodies. exporters seeking services from domestically accredited organizations because their certificates and test results were not recognized abroad. Policy recommendations TÜRKAK's general assembly and board of directors should include a greater share of representatives from the private sector.TÜRKAK serves mainly the private sector and should include more industry representatives, particu- larly of associations of conformity assessment bodies such as TÜRKLAB. TÜRKAK's Establishment Law should be less prescriptive and should allow the organization more autonomy. TÜRKAK's structure is too An Evaluation of Turkey's National Quality System 249 rigid and it may have too many administrative personnel for its current needs. As a result, it cannot adapt itself to changing market requirements. The Establishment Law should not outline the organizational structure of TÜRKAK by defining the different service units and administrative departments, nor should it prescribe staff titles and number of employees. These issues are addressed in a draft of the revised Establishment Law. The government should provide financial and technical support for accreditation activities. This will help strengthen the infrastructure and human capacities of the conformity assessment entities operating locally and foster competition in the market for accredited conformity assess- ment services. Local capacity will allow Turkey to become less dependent on foreign assessors, reducing costs, and will help develop TÜRKAK's customer base, an important requisite for international recognition. Turkey should seek membership in the IAF. Most EA members are also full members of the IAF, allowing them to gain more international rec- ognition and to become involved in technical cooperation with countries outside of Europe. Given that Turkey has already satisfied all the condi- tions for membership in the EA, gaining admission to the IAF should be relatively straightforward once membership fees are paid. The government should support TÜRKAK's application for product certification and environmental management systems in the EA MLA. Unless TÜRKAK becomes a signatory of the MLA in these areas, the mar- ket for product certification and environmental management certification will remain underdeveloped in Turkey. Without access to accredited cer- tification bodies in these areas, local and international recognition of the quality and environmental impact of locally produced goods will suffer. Metrology and Calibration Metrological activities in Turkey Compared to NMIs in other European countries, the National Metrol- ogy Institute of Turkey offered few calibration services in 2005, and most were provided to end users. The national metrology institute of Hungary, for example, provided 10 times as many calibration services and served 10 times as many clients as Turkey's UME (figure B16). Moreover, only 9 percent of all of UME's calibrations were performed for commercial calibration laboratories (primary calibrations); the rest were performed for final industrial users (secondary calibrations). As a result, UME domi- nates the secondary calibration market, providing more calibrations than all other laboratories combined. This is in stark contrast to what is seen 250 Quality Systems and Standards for a Competitive Edge Figure B16 Number of Calibration Activities Relative to Manufacturing Value Added, and Number of Calibration Customers Relative to Number of Manufacturing Firms, in NMIs of Selected Countries, 2005 2.0 0.10 of in 0.08 1.5 es 2005 vic 0.06 calibration in ser 1.0 ecipientsr of es 2005 0.04 of vic ser0.5 0.02 number number calibration 0 0 Hungary Poland Turkey number of calibration number of recipients of services/manufacturing calibration services/number value added of manufacturing firms Sources: National metrology institutes; World Bank 2006b; OECD 2005. Note: 2002 data for number of rms; 2003 data for manufacturing value added. in most other industrial countries, where commercial calibration labora- tories account for the vast majority of NMI clients. The number of calibration laboratories using UME's services is insuf- ficient to ensure an adequate diffusion of traceable measurement among Turkish firms. UME's high share of secondary calibrations is partly caused by the lack of high-quality calibration laboratories in Turkey. Furthermore, the market for calibration services is developing very slowly because industrial users prefer UME's more reputable services over those of private laborato- ries, even for simple calibrations that are well below UME's technical capa- bilities. Without access to a network of commercial calibration laboratories traceable to UME, industrial firms will not be able to obtain the levels of production accuracy and precision needed to produce high-quality goods. The institutional framework of metrology UME has a fairly large and technically competent staff, but these personnel are underutilized. UME employs 200 staff members, 113 of whom work in the laboratories. UME offers competitive salaries and is able to attract highly qualified staff. All of the scientific metrology personnel have at least a university degree in science or engineering, and a large proportion have postgraduate education. UME is not making the most of its techni- cal personnel to provide calibration services to the private sector. Relative to the size of its metrology staff, UME provides very few calibrations. For example, in 2005 the annual number of calibrations per metrologist in the An Evaluation of Turkey's National Quality System 251 national metrology institute was 36 in Turkey, compared to 173 and 698 in Poland and Hungary respectively. Furthermore, UME's laboratories are underutilized because of higher staff turnover than expected and difficul- ties in recruiting technical staff due to government ceilings. Turkey participates in few international inter-laboratory comparisons. The MRA of the International Committee of Weights and Measures (CIPM) is the most comprehensive multilateral agreement for the recog- nition of measurements between NMIs. The measurement comparisons required by the CIPM MRA have become the principal reference for infor- mation on the technical capabilities of NMIs. Turkey has signed the CIPM MRA, but UME needs to participate in international comparisons of mea- surements so that these measurements will be recognized abroad. Turkey registers more "key comparisons"10 than the two other EU candidate coun- tries, Romania and Bulgaria, but still lags behind EU member states such as Poland, Hungary, and Spain, and far behind the Republic of Korea (figure B17). TÜRKAK, like all other signatories of the EA MLA, requires mea- surement traceability to accredit calibration laboratories.Without access to internationally recognized measurements, Turkish calibration laboratories cannot gain accreditation and cannot disseminate the accurate and precise measurements needed in Turkey's export industries. Policy recommendations Turkey's National Metrology Institute should formulate a strategy for its transition from the secondary calibration market to the primary calibration market.To ensure that it is not stifling competition in the market for second- ary calibrations, UME should gradually disengage itself from the secondary calibration market as private metrology institutions appear in this market. Figure B17 Number of Key and Supplementary Measurement Comparisons by Signatories of the CIPM MRA 250 199 200 150 139 omparisonsc 115 90 87 of 100 77 54 50 31 19 2 number 0 . of. y o TurkeyRomaniaMala ysia Spain oland Chile Rep P Hungar Mexic Bulgaria ea, Kor Sources: CIPM online database, appendix B, 2006. 252 Quality Systems and Standards for a Competitive Edge Turkey should participate in more key comparisons to extend the scope of measurements covered by the CIPM MRA. International and regional metrology organizations, such as the BIPM and EUROMET (European Collaboration in Measurement Standards), provide the institutional frame- work to organize international and regional comparisons of national labo- ratories. UME needs to participate in more international inter-laboratory comparisons to provide Turkish candidates for accreditation with traceable measurements that are accepted nationally and internationally. Turkey should use its advanced technical infrastructure to cooperate with neighboring countries that wish to develop a recognized national metrology institute. In addition to increasing scientific ties and exchanges with other countries, this could provide additional sources of income to UME as it leaves the secondary calibration market to the private sector. Standardization Standards development in Turkey Turkey has a large and rapidly growing standards stock. As of 2006, the standards catalogue of the Turkish Standards Institution lists 28,848 stan- dards, slightly fewer than in Romania but more than in most other coun- tries, including the United Kingdom and Korea (figure B18).The standards stock has significantly grown in the past few years (up from 14,616 stan- Figure B18 Number of National Standards, Selected Countries 30,000 number of national standards based on ds 25,000 international standards 20,000 standar number of national of 15,000 standards based on regional standards 10,000 number number of purely 5,000 national (idiosyncratic) 0 standards TurkeyBulgariaKingdomea, Rep . of ysia Chile Romania Mala ed Kor Unit Sources: Web sites of the national standards bodies (ASRO, TSE, BDS, BSI, KATS, DSM, INN). Note: January 2006 data for Romania, Turkey, and Bulgaria; December 2004 data for Malaysia; December 2003 data for the United Kingdom; December 2002 data for Chile and Korea. Regional standards correspond to ENs in the case of Turkey. An Evaluation of Turkey's National Quality System 253 dards in 2000), and 1,122 standards were adopted in 2005 alone.11 As in other countries, most standards in Turkey concern particular products, but there are also a number of process standards, such as ISO 9000 standards. European standards (EN) constitute a growing share of Turkey's stan- dards stock, more than 50 percent as of January 2006 (figure B18). Tur- key has now adopted more than 90 percent of the European standards. However, approximately 5,000 ENs, or one-third of total ENs, have been adopted in English only, without a Turkish translation. These standards will only be translated if they are included in standards preparation work programs at the request of stakeholders. Given the limited level of English proficiency in Turkey, this may constitute a major constraint to stakeholders' access and their ability to export to the EU. Turkey continues to implement EU directives and to eliminate existing technical regulations. The Old Approach and New Approach directives incorporate the technical product and process specifications that a country must adopt in national legislation to participate in the EU free market.The government ministries are transposing and integrating the European legis- lation related to the Old Approach and the New Approach into the Turkish legal system and removing conflicting technical regulations.Approximately 80 to 90 percent of EU legislation has been adopted and transposed into Turkish law, including 25 of 29 New Approach directives. Besides European standards, Turkey continues to actively pursue a policy of adopting international standards. Turkey has adopted more than 9,000 ISO and IEC standards. These represent approximately 31 percent of its standards stock, a comparable or higher share than in most other countries (figure B18). Turkey is adopting standards in sectors that are important for trade, but a few of these sectors have an unusually high number of idiosyncratic standards. Figure B19 shows a roughly linear relationship between the standards stock and the value of trade in each sector. Because most Turk- ish standards are international or regional, this pattern of standardization should decrease barriers to trade in the most relevant sectors. Nonetheless, figure B19 shows that some sectors have many more stan- dards than predicted by their relevance to trade. First, as expected, sectors that are more technology-intensive are generally also standards-intensive. Second, some sectors, such as construction, have a much larger standards stock than predicted by Turkish trade levels. This is because Turkey has adopted almost 15,000 European standards to support its EU candidacy, and many, such as in the construction sector, are not relevant to the Turkish economy. Third, three sectors that play an important role in both imports 254 Quality Systems and Standards for a Competitive Edge Figure B19 Standards Stock vs. Value of Trade in Turkey 2,000 construction food 1,800 electrical telecom audio & video 2006) 1,600 agriculture chemicals ch ar 1,400 fluid systems R2 = 0.267 (M metallurgy ds 1,200 manufacturing aircraft health care rubber & plastics 1,000 textile & leather mechanical petroleum standar systems & electronics domestic 800 products SET components & commercial road equipment products vehicles of 600 handling equipment IT & office machines shipbuilding mining & minerals 400 wood glasspaints & colorsenergy & heat transfer testing paper number 200 railway & ceramics image clothing military technology clocks & jewelry 0 11 12 13 14 15 16 17 value of trade in 2003 (US$ thousands), log Sources: UNCTAD 2005; TSE Web site. and exports--food technology, chemical technology, and agriculture--are heavily standardized, but most of these standards (65­82 percent) are purely idiosyncratic, that is, they are country-specific standards not shared with any other country.12 Turkey's large share and absolute number of idiosyncratic standards in these sectors could generate technical barriers to trade. These standards can have a negative impact on imports if other countries must bear the additional costs of modifying their products specifically for the Turkish market, and they could even have a negative impact on exports by placing Turkey at a comparative disadvantage in relation to international standards. The institutional standardization framework TSE is de facto a voluntary standards organization, although the legisla- tion also gives it a role in mandatory standards. According to its Estab- lishment Law, TSE recommends to the ministries standards that are to be made mandatory. This role has created confusion and represents an obstacle to its membership in the European Committee for Stan- dardization (CEN) and the European Committee for Electrotechnical Standardization (CENELEC). As a result, TSE's technical board has not recommended that any standards be made mandatory since 2002. TSE's administrative structure is dominated by representatives of the public sector. Contrary to many other standards bodies, TSE is not an open membership organization but operates according to a representa- An Evaluation of Turkey's National Quality System 255 Table B2 Governance of the National Voluntary Standards Bodies, Selected Countries Share of executive council reserved Executive director General Consultative for government appointed by Country assembly council appointees (%) government Bulgaria no yes 100 yes Hungary yes no 33 no Korea, Rep. of no yes 100 yes Mexico yes yes 0 no Romania yes yes 5 no Spain yes no 0 no Turkey yes no 55 no United Kingdom yes yes 0 no Source: National standards bodies. tive system (table B2). Slightly more than half of the seats in the general assembly, the highest authority, are allocated to representatives of rel- evant government ministries and public agencies; the rest are reserved for representatives of universities and the productive sector. Remunerated experts participate in the preparation of standards in TSE, an uncommon practice in standards bodies. National standards organizations typically implement a work program and delegate the technical aspects of standards development to technical committees made up of representatives from government agencies, the productive sector, consumer associations, academia, and research institutions. Each technical committee is responsible for developing standards in a specific area and may be formed and dissolved by the standards body as the need arises. In other countries, members of technical committees are usually volunteers who do not receive remuneration from the standards body for their work, but they may be sponsored by their organizations. In Turkey, standards preparatory groups (SPG) develop draft standards in particular sectors through technical experts appointed by TSE, who receive com- pensation for their work.13 To conform to international best-practice and consensus-driven standardization, TSE is changing this practice in favor of open voluntary committees, and SPGs will be eliminated by 2007. TSE has been granted institutional autonomy from the government in a number of areas but faces restrictions on its workforce. Since January 2006, a new budgetary implementation law for public entities imposes restrictions and delays on contracting outside experts and purchasing equipment. As a result, TSE has become less autonomous than other 256 Quality Systems and Standards for a Competitive Edge national standards bodies. This could make TSE less effective at deliver- ing the services needed to finance the standards development process, particularly in areas that require very specialized technical skills. Turkey is a full member of the major international standards organiza- tions but does not have full membership in the European standards bod- ies. Participation in the international and regional standardization process ensures that the needs of the private sector are reflected in standards that are shared by Turkey and its trade partners. Turkey is a full member of ISO, IEC, and ITU, but is only an affiliate member of the regional stan- dards bodies CEN and CENELEC. As an affiliate member, Turkey has limited input in the standards development process, can only participate as an observer in the technical committees, and has no voting rights on standards in the general assemblies. Most of the current barriers to full CEN membership are being addressed in new Turkish legislation under preparation. This legislation will abolish TSE's role in mandatory standards, will eliminate the SPGs so that stan- dards are developed through consensus, and will clearly separate the stan- dardization and commercial functions of TSE to avoid conflicts of interest. At the international level, Turkey does not participate very actively in the development of standards.There are important benefits to participating in the international standardization process. Active participation increases the chance that national concerns will be reflected in international stan- dards. It also enables the compatibility of national standards with inter- national standards. Participation in international standards development can also provide a forum for exchange of technical information with rep- resentatives from international industrial and scientific organizations. As of December 2004, Turkey was a participant member in only 107 ISO technical committees out of a total of 734. This is far fewer than Romania, which has a much lower GDP than Turkey (figure B20). Turkey's limited participation in international standards development is partly due to a lack of interest in the private sector. Moreover, travel costs make it expensive to participate in international technical committees and working groups. Policy recommendations TSE should ensure that its standardization and commercial activities are administratively,physically,and financially separated.There are many coun- tries, such as the United Kingdom, Spain, and Korea, where the national standards body also offers a number of commercial services, such as certi- fication and testing, but the separation of functional areas in the standards body is typically made explicit by law or in government agreements. In the An Evaluation of Turkey's National Quality System 257 Figure B20 Participant and Observer Memberships in ISO Technical Committees and Subcommittees 700 participant 600 observer 500 400 memberships of 300 200 number100 0 . of y ysia o Chile Rep Spain Poland Turkey Mexic ea, Romania Hungar Mala Bulgaria Kor Source: ISO Web site. Note: January 2006 data for Bulgaria, Hungary, Malaysia, Poland, Romania, and Turkey; October 2005 data for Chile, Korea, Mexico, and Spain. United Kingdom, the memorandum of understanding between BSI British Standards, the national standards body, and the UK government requires that management decisions concerning the standards division should not be influenced by other BSI business activities. The memorandum also subjects any collaboration of the standards division and other commercial divisions to competition law. Such an arrangement, if implemented in Tur- key, would guarantee the absence of conflicts of interest in TSE's functions. This issue is currently being addressed by TSE. If TSE does not clearly separate its standardization and commercial activities, it is recommended that the commercial branch of TSE be privatized and that the government provide full funding for TSE's standardization activities. As an autonomous public body providing commercial services to the private sector, TSE should not be subject to government procurement and hiring legislation. To be competitive in the certification and testing services market, TSE needs to be able to independently set its salaries, select its workforce, hire external contractors, and purchase technical equipment. In most EU countries, the national standards bodies have been granted full autonomy from the government to allow them to remain competitive in their nonstandardization activities.14 TSE's administration should include more private sector representa- tives. In particular, TSE's general assembly should include a majority of private sector representatives and at least one consumer association. 258 Quality Systems and Standards for a Competitive Edge TSE should ensure that its role in the development of mandatory stan- dards is not ambiguous. Currently, TSE's Establishment Law states that it can recommend mandatory standards to the government ministries. This issue is currently being addressed in the draft of the revised TSE Establishment Law. TSE should continue to replace the SPGs with voluntary techni- cal committees for the development of national standards. TSE should ensure that the new technical committees use the consensus principle and include representatives from all stakeholder groups. Turkey should continue removing obsolete standards and standards that conflict with EN standards. Many Turkish standards were not devel- oped through an open voluntary participation process and may not rep- resent the needs of stakeholders. Turkey should continue adopting the remaining European and inter- national standards and should ensure that ENs are properly translated into Turkish. Turkey should ensure that its most important trade sectors do not become overstandardized and should make full use of available international and European standards. Turkey should intensify its efforts to obtain full membership in CEN and CENELEC and should increase its participation in technical commit- tees. Membership in the European standards bodies and active participa- tion in the work of technical committees will allow Turkey to provide its inputs to the regional standards that it adopts. The Turkish government should support participation of the private sector in European and inter- national standardization activities. Many of the European standards are harmonized standards and effectively support Turkish legislation trans- posed from European directives. It can be difficult for firms, especially SMEs, to represent Turkish interests in CEN, CENELEC, ISO, and IEC technical committees due to high travel costs. Turkish participation in these organizations should be supported by government funding. Conclusion Turkey has made significant progress toward establishing a modern, market-based quality standards regime that will facilitate its accession to the EU and entry into other international markets. Turkey has a func- tioning national quality system in place, comprising a national standards institution, an accreditation agency, and a national metrology institute, as well as a fair number of testing and calibration laboratories and certi- fication and inspection entities. Quality awareness among Turkish firms Table B3 Turkey's National Quality System: Deficiencies, Ongoing Actions, and Recommendations Recent and ongoing government activities and initiatives planned Recommended actions Deficiencies, based on for the short term for the medium term Area the diagnostic (6 months) (6 months to 3 years) Certi cation ISO 9000 certi cation Technical assistance Continue support for rates are low among programs for quality quality certi cation among small rms and in certi cation are provided SMEs and in the services nonmanufacturing by KOSGEB. sector through matching sectors. Turkish noti ed bodies grant programs. Poor business practices are being designated Government procurement exist among certain and presented to the rules should ensure that certi cation bodies. European Commission. quality certi cates are There are no national issued by organizations noti ed bodies. with ethical practices. Accreditation There are high A draft of a revised The government should accreditation fees and Establishment Law support building up local low demand for TÜRKAK provides TÜRKAK with capacity for advisory and accreditation. more autonomy over consultancy services, International its administration and mostly through training. recognition of Turkish personnel. TÜRKAK should seek accreditation is low. TÜRKAK has been membership in the IAF. accepted in the EA MLA. Metrology Demand is low for UME's Develop a strategy for primary calibrations, UME to transition from the which mostly cater to secondary to the primary the secondary market. calibration market. Not enough of UME's Participate in more measurements comparisons of calibration are comparable laboratories. internationally. Use advanced metrological infrastructure for regional cooperation. Standardization Turkey has a very large TSE is gradually reviewing Develop a strategy to standards stock. and removing obsolete translate EN standards into EN standards have not standards. Turkish. all been translated into A draft Establishment Increase participation in Turkish. Law eliminates TSE's role international standards TSE's role in mandatory in mandatory standards, development technical standards is ambiguous. creates open voluntary committees. The standards committees, includes development process is consumer representation not open and voluntary. in TSE's governance Turkish participation structure, and separates is limited in the TSE's di erent functions. development Turkey is applying for full of regional and membership in CEN and international standards. CENELEC. Source: Authors'research. 259 260 Quality Systems and Standards for a Competitive Edge seems to be improving, as shown by the increasing rate of growth of quality certificates, although in absolute terms Turkey is still far from rel- evant comparator countries. The country has already replaced almost all national technical standards with EU and international standards and has significantly reduced the number of mandatory standards. The recently created accreditation body, TÜRKAK, was accepted into the EA MLA for quality management systems, testing, calibration, and inspection in April 2006. UME, the metrology institution, is well equipped and has highly qualified staff. It already offers reliable measurement traceabil- ity, which is needed for the proper functioning of the Turkish quality infrastructure. Nonetheless, policy and institutional changes are still needed to fur- ther improve Turkey's national quality system and foster adoption of quality standards at the firm level. Furthering international recogni- tion of the main institutions of the national quality system is crucial to remove constraints faced by Turkish producers and exporters. UME needs to participate in more international comparisons of measurements, and TÜRKAK has to broaden its participation in the recently signed EA MLA by applying to also become a signatory in the areas of product cer- tification and environmental management systems certification. Legisla- tion regulating the functioning of these institutions needs to be revisited to clarify responsibilities, increase flexibility, and facilitate the adoption of more efficient and transparent practices with increased participation of the private sector. Finally, it is necessary to increase Turkish labs' and firms' requests for accreditation and certification. Matching grants have proven successful in encouraging adoption of quality standards in several countries. See table B3 for a summary. Notes Chapter 1 1. Manufacturing refers to the industries belonging to ISIC Rev. 3.1, divisions 15 through 37. Chapter 2 1. The height-to-width ratio of the A4 format is such that if two pages are placed next to each other, the resulting double page will have the same height-to-width ratio as a single page. This makes scaling multiple sheets onto a single sheet through photocopying extremely efficient, because two or more sheets can be scaled down by a mul- tiple of two to fit exactly on a single sheet, without loss of space or onerous scaling factor calculations. 2. From the history section of the BSI Group Web site, http://www .bsi-global.com/News/History/index.xalter. 3. United States of America v. Microsoft Corporation. Findings of Fact. United States District Court for the District of Columbia, Novem- ber 5, 1999. 4. In U.S. dollars, $180,390, $1,288, and $3,092,400, respectively. 5. The terms "mandatory standards,""obligatory standards," and "techni- cal regulations" all refer to government-enforced standards. 6. See chapter 5 for discussion of mutual recognition agreements. 7. The best examples of buyer-driven chains are in the United States apparel industry. In 1995, the top five largest national apparel retail- ers accounted for 68 percent of all apparel sales in publicly held retail outlets (Gereffi 1999). 261 262 Notes Chapter 3 1. There is still some debate about the merits of the QWERTY key- board in the literature. 2. The terms "standards body," "standardization body," "standardizing body,""standards development organization,"and"standardization orga- nization" are all commonly used to refer to the same type of entity. 3. The consensus principle aims to protect minority interests. Often a pure form of unanimity is not required to adopt a standard, but a dem- onstrated attempt is made to reconcile negative votes for a draft stan- dard and a significant majority is required to approve the standard. Chapter 4 1. "Certification" refers to the issuance of a written assurance and "regis- tration" refers to the recording of the certification in the certification body's client register. The terms are used interchangeably in many countries. 2. The quality management system certification bodies with the larg- est global market shares are BSI British Standards (UK), Perry John- son Registrars (U.S.), URS (UK), QMI (Canada), ABS (U.S.), SGS (Switzerland), DNV (Norway), Underwriters Limited (U.S.), LRQA (UK), and INTERTEK (U.S.). See the QSU Publishing Company Web site, http://www.qsuonline.com/BodyPages/BigTen.html. 3. These guidelines are documented in ISO/IEC Guides 62, 65, and 66. 4. Relying exclusively on reputation may have anticompetitive effects, because it favors incumbents at the expense of new entrants. 5. Guidelines for testing and calibration laboratories are documented in the ISO/IEC 17025 standard; guidelines for inspection bodies are documented in the IOS/IEC 17020 standard. 6. Examples include ISO/IEC Guide 58, the ISO/IEC 17011 standards, and the IAF Code of Conduct. 7. Standards used to assess laboratories include the ISO/IEC 17025;those used to assess quality systems include the ISO/IEC 9001:2000. Chapter 5 1. For example, ISO 9000 and ISO 14000 registration. 2. Legitimate objectives include those concerned with national security requirements;prevention of deceptive practices;and protection of human health or safety, animal or plant life or health, or the environment. Notes 263 3. The Code of Good Practice is a section of the WTO TBT agree- ment that outlines principles for the development and application of standards and is open to acceptance by the standardizing bodies of member states. It includes specific recommendations regarding trans- parency and the involvement and consultation of interested parties throughout the standards development process. The WTO is to be notified of work programs concerning standards on a periodic basis. The code prohibits the use of standards as barriers to trade. It com- mits standardization bodies to using international standards as a basis for national standards, unless these are ineffective or inappropriate. The Code of Good Practice has been accepted by 152 standardiza- tion bodies in 111 countries. 4. The ITU uses the term "recommendations" instead of "standards." 5. In free trade areas, countries agree to eliminate tariffs and quotas but each country maintains an independent external trade policy, making this a weaker form of economic integration than other RTAs such as customs unions. 6. Such as ISO/IEC 17025 for calibration and testing laboratories. 7. For example, ISO/IEC 17011:2004 specifies general requirements for accreditation bodies tasked with assessing and accrediting confor- mity assessment bodies. It is also used as a requirements document for the peer evaluation process for mutual recognition arrangements between accreditation bodies. 8. When MRAs cover accreditation in the areas of product certification, testing, or calibration, they usually cover the entire range of possible subareas--that is, accreditation for all types of products, all types of tests, or all types of calibrations, respectively--rather than only one specific subarea. 9. OIML members are required to comply with the OIML interna- tional recommendations regarding metrological performance and testing procedures. These recommendations are used as a basis for EU directives, and EU member states are legally obliged to adopt them in their regulations. Hence, adhering to the recommendations can facilitate trade with the EU. 10. OIML certification provides the manufacturer with the possibility of obtaining an OIML certificate and test report confirming that the instrument type complies with relevant international recommenda- tions. The certificates and test reports are delivered by issuing author- ities in OIML member states and are accepted in other countries on a voluntary basis, thereby eliminating the costly duplication of 264 Notes application and test procedures for measurement instruments. The OIML issued 112 certificates in the first 10 months of 2005. The OIML is currently establishing a mutual acceptance arrangement for test reports of type evaluation. Chapter 6 1. Referred to as ISO 9001:1994, ISO 9002:1994, and ISO 9003:1994. 2. Rao, Ragu-Nathan, and Solis's survey data measured a number of fac- tors that underlie quality practices and results, including leadership, information and analysis, strategic quality planning, human resources development, quality assurance, supplier relationships, and customer orientation. 3. The South African survey was confined to agribusiness firms, which may account for cost differences. 4. Grajek (2004) uses a gravity equation for international trade, which predicts that the volume of trade between two countries will be proportional to their economic mass and inversely proportional to the distance between them and other trade barriers. This equation is combined with an international diffusion equation in which ISO 9000 adoption in one country is influenced by ISO 9000 adoption in close trade partners. Trade barriers are defined to be a decreasing function of bilateral adoptions of ISO 9000, that is, ISO 9000 certi- fication in both trade partners should establish a common language that decreases trade barriers. 5. Sectoral proclivities for certification also occur in highly globalized sec- tors where a large portion of the output is destined for exports or where transnational production networks are dominant (Schuurman 1997). 6. Guler, Guillén, and Muir Macpherson ruled out the possibility of reverse causation by measuring all independent variables with a one-year lag. Chapter 7 1. In Peru, a single entity, INDECOPI, is responsible for an exception- ally large number of tasks. These include not only national standards development, metrology, and accreditation, but also intellectual property registration, consumer protection, and enforcement of poli- cies on competition. 2. A notable exception is INDECOPI in Peru, where the government is responsible for more than half of the standardization budget. Notes 265 3. ISO has 190 technical committees, 544 technical subcommittees, and 2,188 working groups. 4. The four priority objectives are (a) align domestic standards in APEC economies with international standards; (b) achieve recognition of conformity assessment, including mutual recognition arrangements in regulated and voluntary sectors; (c) promote cooperation for tech- nical infrastructure development to facilitate broad participation in mutual recognition arrangements in both regulated and voluntary sectors; and (d) ensure the transparency of the standards and confor- mity assessment of APEC economies. 5. Resolution 502 of the Andean Community. 6. Resolution 2/92 of Mercosur. Chapter 8 1. QS 9000 is an older standard established by the top three U.S. auto- makers and used worldwide, along with other regional standards. ISO/TS 16949 is a newer uniform standard developed by all of the world's major automakers. Most automakers set deadlines for their suppliers to upgrade to ISO/TS 16949 between 2004 and 2006. 2. This could be due to a high certification rate in Mexico, but it may be also due to the fact that some manufacturers in Germany and Japan are registered to standards other than QS 9000. 3. These statistics only consider the formal sectors. Chapter 9 1. The autonomy index (min=0, max=1) is calculated as the fraction of the following statements that are true: The national accredita- tion body has the autonomy to (a) set its fees; (b) set the salaries of its workforce; (c) hire its personnel; (d) grant and revoke accredi- tations; (e) join international organizations and sign international agreements; (f) set its own budget; (g) create new administrative divisions; (h) offer new services. Chapter 11 1. The Inter-American Development Bank has proclaimed CYGA to be one of the best enterprise development programs of its type in Latin America (COPANT 2005). 266 Notes 2. Workshops are offered in the areas of process re-engineering, contin- ual improvement, customer service, basic process integration, logis- tics, and model change optimization. Firms wishing to participate are first screened through a free prediagnostic process that establishes the feasibility of a workshop. Appendix A 1. Guidelines for testing and calibration laboratories are documented in the ISO/IEC 17025 standard; guidelines for inspection bodies are documented in the IOS/IEC 17020 standard. Appendix B 1. The semi-elasticity of employment with respect to quality certifica- tion is 0.448. 2. The percentage contributions are constructed by dividing the prod- uct of the average value of the investment climate variables and the corresponding elasticity or semi-elasticity by the average value of log- employment, all multiplied by 100. Using log-employment instead of employment allows us to take advantage of the additive properties of logarithms. 3. These include the requirements described in voluntary and manda- tory standards. 4. In addition to Turkey, accession and candidate countries include Bulgaria, Croatia, the Former Yugoslav Republic of Macedonia, and Romania. 5. Firm size is defined by number of employees: small, 1­49; medium, 50­249; large, 250+. 6. This number reflects the difference between the number of cer- tificates registered through The ISO Survey of Certifications 2004 (ISO 2005), which only includes accredited certification bodies, and the total number of certificates estimated in a separate local survey reported in Support to the Quality Infrastructure in Turkey (CEN 2003). 7. These bodies are said to deliver certificates in as little as a week, while it usually takes a few months to conduct a proper audit and register a company. Notes 267 8. This may be due to the lack of international recognition of the Turk- ish accreditation system. TÜRKAK has just signed the European Co- operation for Accreditation Multilateral Recognition Arrangement (see the next section). 9. ISO/IEC 17025, ISO/IEC 17020, EN 45012, EN 45011, ISO/IEC 17024, Guide 66. 10. The recognition of national measurements by signatories of the CIPM MRA relies on a database of "key comparisons" of national measure- ment standards, based on participation in international and regional inter-laboratory comparisons. 11. Turkey has also been engaged in a strategy of removing all obsolete standards. In 2005, more than 9,000 standards prepared before 2000 were reviewed in consultation with stakeholder groups. It is evident that many of the standards are not useful. For example, TSE, the national standards body, found 1,800 standards whose documenta- tion had not been sold to a single user during a three-year period. 12. Turkey is using more of its own agricultural and food standards than France, Belgium, and Italy, even though these sectors represent a relatively similar share of trade in all these countries. In the case of chemicals, idiosyncratic standards represent twice the proportion of the standards stock as they do in Italy, where they account for a larger share of trade. 13. SPGs also examine international documents to make proposals for ISO and IEC votes. 14. Moreover, in some countries the national standards body is explicitly allowed to cross-subsidize the standardization activities with the pro- ceeds of other commercial activities. For example, the memorandum of understanding between BSI British Standards and the UK govern- ment states that BSI must apply profits from its nonstandardization activities to its standardization activities. References Publications Albuquerque, P., B. Bronnenberg, and C. J. Corbett. 2004. "A Spatio-Temporal Analysis of the Global Diffusion of ISO 9000 and ISO 14000 Certification." Working Paper, Anderson School of Management, University of California, Los Angeles. Aldaz-Carroll, E. 2006. "Regional Approaches to Better Standards Systems." Policy Research Working Paper 3948, World Bank, Washington, DC. Anderson, S. W., J. D. Daly, and M. F. Johnson. 1999. "Why Firms Seek ISO 9000 Certification: Regulatory Compliance or Competitive Advantage?" Produc- tion and Operations Management 8 (1): 28­43. APEC (Asia-Pacific Economic Cooperation). 2004a. Standards and Conformance 2004 Collective Action Plan. http://www.apec-iap.org/document/APEC_ 2004_Standards_and_Conformance.htm. ------. 2004b. Report of 2004 VAP Results. Subcommittee on Standards and Conformance, APEC, Santiago, Chile. Ayyagari, M., T. Beck, and A. Demirgüç-Kunt. 2003. "Small and Medium Enter- prises across the Globe: A New Database." Policy Research Working Paper 3127, World Bank, Washington, DC. Bair, J., and G. Gereffi. 2001. "Local Clusters in Global Chains: The Causes and Consequences of Export Dynamism in Torreon's Blue Jean Industry." World Development 29 (11): 1885­1903. Bazan, L., and L. Navas-Alemán. 2001. "Comparing Chain Governance and Up- grading Patterns in the Sinos Valley, Brazil." Paper presented at workshop on "Local Upgrading in Global Chains," Institute of Development Studies, University of Sussex, Brighton, UK, February 14­17. ------. 2003. "Upgrading in Global and National Value Chains: Recent Chal- lenges and Opportunities for the Sinos Valley Footwear Cluster, Brazil." Paper 269 270 Quality Systems and Standards for a Competitive Edge presented at workshop of the EADI Working Group on Industrialisation Strategies in Developing Countries, "Clusters and Global Value Chains in the North and the Third World," Novara, Italy, October 30­31. Beck, T., A. Demirgüç-Kunt, and R. Levine. 2003. "Small and Medium Enter- prises, Growth, and Poverty: Cross-Country Evidence." Policy Research Working Paper 3178, World Bank, Washington, DC. Bhatnagar, S. 2006. "India's Software Industry." In Technology, Adaptation, and Exports: How Some Developing Countries Got It Right, ed. V. Chandra, 49­82. Washington, DC: World Bank. BIPM (International Bureau of Weights and Measures). 2003. Evolving Needs for Metrology in Trade, Industry and Society and the Role of the BIPM. Paris: BIPM. Birch, J. 2003. Benefits of Legal Metrology for the Economy and Society. Paris: Inter- national Committee of Legal Metrology. Blind, K. 2004. The Economics of Standards: Theory, Evidence, Policy. Northamp- ton, MA: Edward Elgar. Blind, K., and A. Jungmittag. 2005."Trade and the Impact of Innovations and Stan- dards: The Case of Germany and the UK." Applied Economics 37: 1385­98. Brown, A., T. van der Wiele, and K. Loughton. 1998. "Smaller Enterprises' Expe- riences with ISO 9000." International Journal of Quality & Reliability Manage- ment 15 (3): 273­85. Casadesús, M., and G. Giménez. 2000. "The Benefits of the Implementation of the ISO 9000 Standard: Empirical Research in 288 Spanish Companies." TQM Magazine 12 (6): 432­41. CEN (European Committee for Standardization). 2003. Support to the Quality Infrastructure in Turkey: Country Report 2002­2003. Brussels: CEN. ------. 2004. Compass: European Standardization in a Nutshell. 2nd ed. Brussels: CEN. ------. 2005. Annual Report 2004/2005. Brussels: CEN. Chandra, V., ed. 2006. Technology, Adaptation, and Exports: How Some Developing Countries Got It Right. Washington, DC: World Bank. Chaturvedi, S., and G. Nagpal. 2003. "WTO and Product-Related Environmen- tal Standards: Emerging Issues and Policy Options before India." Discussion Paper 36/2002. Research and Information Systems for the Non-Aligned and Other Developing Countries, New Delhi. Chen, M. X., and A. Mattoo. 2004. "Regionalism in Standards: Good or Bad for Trade?" Policy Research Working Paper 3458, World Bank, Washington, DC. Chen, M. X., T. Otsuki, and J. S. Wilson. 2006. "Do Standards Matter for Export Success?" Policy Research Working Paper 3809, World Bank, Washington, DC. References 271 Chow-Chua, C., M. Goh, and T. B. Wan. 2003. "Does ISO 9000 Certification Improve Business Performance?" International Journal of Quality & Reliability Management 20 (8): 936­53. Cohen, W. M., and D. A. Levinthal. 1989. "Innovation and Learning: The Two Faces of R&D." Economic Journal 99 (397): 569­96. Commission of the European Communities. 2001. "Statistical and Technical An- nex to the Report on the Functioning of Community Product and Capital Markets." 7.12.2001 SEC (2001) 1993. Brussels. COPANT (Pan American Standards Commission). 2005. COPANT News 4 (1). Corbett, C. J. 2003. "Global Diffusion of ISO 9000 Certification through Sup- ply Chains." Working Paper, Anderson School of Management, University of California, Los Angeles. CORFO (Chilean Economic Development Agency). 2005. "Cuenta pública del vicepresidente ejecutivo de CORFO Oscar Landerretche Gacitúa." Santiago. Delaney, H., and R. van de Zande, eds. 2000. A Guide to EU Standards and Con- formity Assessment. NIST Special Publication 951. Gaithersburg, MD: Na- tional Institute of Standards and Technology. DeVaux, C. R. 2001. A Guide to Documentary Standards. NISTIR 6802. Gaith- ersburg, MD: National Institute of Standards and Technology. De Ferranti, D., G. E. Perry, I. Gill, J. L. Guasch, N. Schady, W. F. Maloney, and C. Sánchez Páramo. 2003. Closing the Gap in Education and Technology. World Bank Latin America and Caribbean Studies. Washington, DC: World Bank. Dick, G. P. M. 2000. "ISO 9000 Certification Benefits: Reality or Myth?" TQM Magazine 12 (6): 365­75. DIN (German Institute for Standardization). 2000. Economic Benefits of Stan- dardization: Summary of Results. Berlin: Beuth Verlag GmbH. Dolan, C., J. Humphrey, and C. Harris-Pascal. 1999. "Horticultural Commodity Chains: The Impact of the UK Market on the African Fresh Vegetable Indus- try." IDS Working Paper 96, Institute of Development Studies, University of Sussex, Brighton, UK. Dolan, C., and J. Humphrey. 2004. "Changing Governance Patterns in the Trade in Fresh Vegetables between Africa and the United Kingdom." Environment and Planning A 36 (3): 491­509. DTI (Department of Trade and Industry). 2005. "The Empirical Economics of Standards." DTI Economics Paper 12, UK Department of Trade and Industry, London. Escanciano, C., E. Fernández, and C. Vázquez. 2002. "Linking the Firm's Techno- logical Status and ISO 9000 Certification: Results of an Empirical Research." Technovation 22: 509­15. 272 Quality Systems and Standards for a Competitive Edge Escribano, A., and J. L. Guasch. 2005a. "Assessing the Impact of the Investment Climate on Productivity Using Firm-Level Data: Methodology and the Cases of Guatemala, Honduras, and Nicaragua." Policy Research Working Paper 3621, World Bank, Washington, DC. ------. 2005b. "Investment Climate Assessment on Productivity and Wages: Analysis Based on Firm Level Data from Selected South East Asian Coun- tries." Finance, Private Sector, and Infrastructure Group, Latin American and the Caribbean Region, World Bank, Washington, DC. Escribano, A., J. L. Guasch, J. Pena, and M. de Orte. 2006. "Turkey: Investment Climate Assessment on Productivity, and Allocative Efficiency: Effects on Exports, FDI, Wages, Employment, Location, Holdings, and Privatization of Firms." Background paper, World Bank, Washington, DC. Gereffi, G. 1994. "The Organization of Buyer-Driven Global Commodity Chains: How U.S. Retailers Shape Overseas Production Networks." In Commodity Chains and Global Capitalism, ed. G. Gereffi and M. Korzeniewicz, 95­122. Westport, CT: Praeger. ------. 1999. "International Trade and Industrial Upgrading in the Apparel Com- modity Chain." Journal of International Economics 48: 37­70. Giulani, E., and M. Bell. 2004. "The Micro-determinants of Meso-level Learning and Innovation: Evidence from a Chilean Wine Cluster." Research Policy 34 (1): 47­68. Goldin, I., and K. Reinert. 2006. Globalization for Development: Trade, Finance, Aid, Migration, and Policy. Washington, DC: World Bank. Grajek, M. 2004. "Diffusion of ISO 9000 Standards and International Trade." WZB Discussion Paper SP II 2004-16, Wissenschaftszentrum, Berlin. GSMAssociation. 2005."GSM Subscriber Statistics." http://www.gsmworld.com/ news/statistics/pdf/gsma_stats_q3_05.pdf. Guler, I., M. F. Guillén, and J. Muir Macpherson. 2002. "Global Competition, Institutions, and the Diffusion of Organizational Practices: The International Spread of ISO 9000 Quality Certificates." Administrative Science Quarterly 47: 207­32. Henten, A., and D. Saugstrup. 2004. "3G Standards: The Battle between WCD- MA and CDMA-2000." CTI Working Paper 90, Center for Tele-Information, Technical University of Denmark, Lyngby. Heras, I., G. P. M. Dick, and M. Casadesús. 2002. "ISO 9000 Registration's Impact on Sales and Profitability: A Longitudinal Analysis of Performance Before and After Accreditation." International Journal of Quality & Reliability Manage- ment 19 (6): 774­91. References 273 Hoekman, B., and S. Togan. 2005. Turkey: Economic Reform and Accession to the European Union. World Bank Trade and Development Series. Washington, DC: World Bank. Hummels, D., and P. J. Klenow. 2005. "The Variety and Quality of a Nations Ex- ports." American Economic Review 95 (3): 704­23. Humphrey, J., and H. Schmitz. 2000. "Governance and Upgrading: Linking In- dustrial Cluster and Global Value Chain Research." IDS Working Paper 120, Institute of Development Studies, University of Sussex, Brighton, UK. IAAC (InterAmerican Accredation Cooperation). 2004. "Evaluatión de la En- cuesta de Capacitación y Entrenamiento y Propuesta de Plan General de Capacitación y Entrenamiento." Subcommittee on Capacity Building and Training, IAAC, Mexico City. IAF/ILAC (International Accreditation Forum/International Laboratory Accredi- tation Cooperation). 2005. IAF/ILAC Multi-Lateral Mutual Recognition Ar- rangements: Key Performance Indicators: A Tool for the Evaluation Process. IAF/ ILAC A3:206. Silverwater, NSW, Australia: IAF/ILAC. ILAC (International Laboratory Accreditation Cooperation). 1994. Guidelines on Scopes of Accreditation. ILAC-G4:1994. Silverwater, NSW, Australia: ILAC. IMF (International Monetary Fund). 2005. World Economic Outlook. September. Washington, DC: IMF. ISO (International Organization for Standardization). 1994. "Code of Good Practice for Standardization." ISO/IEC Guide 59:1994. Geneva. ------. 2001. The ISO Survey of ISO 9001:2000 and ISO 14001 Certificates: Tenth Cycle. Geneva: ISO. ------. 2003a. ISO Members. 10th ed. Geneva: ISO. ------. 2003b. The ISO Survey of ISO 9001:2000 and ISO 14001 Certificates: 2003. Geneva: ISO. ------. 2004. The ISO Survey of Certifications 2003. Geneva: ISO. ------. 2005. The ISO Survey of Certifications 2004. Geneva: ISO. ITC (International Trade Centre). 2004. Gestión de la Calidad de Exportación: Libro de Respuestas para Pequeños y Medianos Exportadores. Buenos Aires: Instituto Nacional de Tecnología Industrial (INTI), Cámara de Exportadores de la República Argentina (CERA), and Centro de Comercio Internacional (CCI). ITU (International Telecommunication Union). 2005. World Telecommunication Indicators Database. 8th ed. Geneva: ITU. ------. 2006. "About Mobile Technology and IMT-2000." ITU, Geneva. http:// www.itu.int/osg/spu/imt-2000/technology.html. 274 Quality Systems and Standards for a Competitive Edge Jalote, P. 2001. "The Success of the SPI Efforts in India." Working Paper 208016, Department of Computer Science and Engineering, Indian Institute of Tech- nology, Kanpur. Jones, R., G. Arndt, and R. Kustin. 1997. "ISO 9000 among Australian Compa- nies: Impact of Times and Reasons for Seeking Certification on Perceptions of Benefits Received." International Journal of Quality & Reliability Management 14 (7): 650­60. Jooste, A., E. Kruger, and F. Kotzé. 2003. "Standards and Trade in South Africa." In Standards and Global Trade, ed. J. S. Wilson and V. O. Abiola, 235­70. Washington, DC: World Bank. Jungmittag,A., K. Blind, and H. Grupp. 1999."Innovation, Standardization and the Long-term Production Function:A Cointegration Analysis for Germany 1960­ 1996." Zeitschrift für Wirtschaftsund. Sozialwissenschaften 119 (2): 205­22. Kaplinsky, R., and J. Readman. 2001. "Integrating SMEs in Global Value Chains: Towards Partnership for Development." Vienna: United Nations Industrial Development Organization. Karapetrovic, S., D. Rajamani, and W. Willborn. 1997. "ISO 9000 for Small Busi- ness: Do It Yourself." Industrial Management 39 (3): 24­31. Krechmer, K. 2000. "The Fundamental Nature of Standards: Economic Perspec- tive." Paper presented at the International J. A. Schumpeter Society Econom- ics Conference, Manchester, UK, June 28­July 1. Lima, M. A. M., M. Resende, and L. Hasenclever. 2000. "Quality Certification and Performance of Brazilian Firms: An Empirical Study." International Journal of Production Economics 66: 143­47. KPMG Consulting. 2001. Potential Economic Impact of the CIPM Mutual Recogni- tion Arrangement. Sèvres, France: BIPM. Magd, H., N. Kadasah, and A. Curry. 2003. "ISO 9000 Implementation: A Study of Manufacturing Companies in Saudi Arabia." Management Auditing Journal 18 (4): 313­22. Marban, R., and J. Pellecer. 2003. Legal Metrology. Washington, DC: Organization of American States. Maskus, K. E., and J. S. Wilson. 2000. "Quantifying the Impact of Technical Barri- ers to Trade: A Review of Past Attempts and the New Policy Context." Paper presented at the World Bank workshop on "Quantifying the Trade Effect of Standards and Technical Barriers: Is It Possible?" Washington, DC, April 27. Mendel, P. J. 2000. "International Standardization and Global Governance: The Spread of Quality and Environmental Management Standards." In Organiza- tions, Policy and the Material Environment, ed. A. J. Hoffman and M. T. Ventr- esca, 407­24. Stanford, CA: Stanford University Press. References 275 Moenius, J. 2004. "Information versus Product Adaptation: The Role of Standards in Trade." Kellogg International Business & Markets Research Center Working Paper 1, Kellogg School of Management, Northwestern University. Monge-González, R., J. Rosales-Tijerino, and G.Arce-Alpízar. 2005."Cost-Benefit Analysis of the Free Trade Zone System: The Impact of Foreign Direct Invest- ment in Costa Rica." Office for Trade, Growth and Competitiveness, Organi- zation of American States, Washington, DC. Naik, G. 2006. "Bridging the Knowledge Gap in Competitive Agriculture: Grapes in India." In Technology, Adaptation, and Exports: How Some Developing Coun- tries Got It Right, ed. V. Chandra, 243­74. Washington, DC: World Bank. Neumayer, E., and R. Perkins. 2005. "Uneven Geographies of Organizational Practice: Explaining the Cross-National Transfer and Diffusion of ISO 9000." Economic Geography 81 (3): 237­59. OECD (Organisation for Economic Co-operation and Development). 2005. OECD in Figures: 2005 Edition. Paris: OECD. Otsuki, T., J. S. Wilson, and M. Sewadeh. 2001. "Saving Two in a Billion: Quan- tifying the Trade Effects of European Food Safety Standards on African Ex- ports." Food Policy 26 (5): 495­514. Quazi, H. A., and S. R. Padibjo. 1998. "A Journey toward Total Quality Manage- ment through ISO 9000 Certification: A Study on Small- and Medium-sized Enterprises in Singapore." International Journal of Quality & Reliability Man- agement 15 (5): 489­508. Quazi, H. A., C. Wing Hong, and C. Tuck Meng. 2002. "Impact of ISO 9000 Certification on Quality Management Practices: A Comparative Study." Total Quality Management 13 (1): 53­67. Quinn, T. J. 2003. "Open Letter Concerning the Growing Importance of Metrol- ogy and the Benefits of Participation in the Metre Convention, Notably the CIPM MRA." Bureau International des Poids et Mesures, Paris. Quinn,T. J., and J. Kovalesky. 2005."The Development of Modern Metrology and Its Role Today." Philosophical Transactions of the Royal Society 363: 2307­27. Raballand, G., and E. Aldaz-Carroll. 2005. "How Do Differing Standards Increase Trade Costs? The Case of Pallets." Policy Research Working Paper 3519,World Bank, Washington, DC. Ramos, A. 1995. "Hacia la calidad total: La difusión de las normas ISO de la serie 9000 en la industria argentina." Working Paper 66 (LC/BUE/L.149), Eco- nomic Commission for Latin America and the Caribbean, Buenos Aires. Rao, S. S., T. S. Ragu-Nathan, and L. E. Solis. 1997. "Does ISO 9000 Have an Ef- fect on Quality Management Practices? An International Empirical Study." Total Quality Management 8 (6): 335­46. 276 Quality Systems and Standards for a Competitive Edge Ritchie, C., R. Sutton, C. Taylor, and B. Warneke. 1999. `The Dynamics of Stan- dards Creation in the Global Wireless Telecommunication Markets." IS 2224, Group D, Project 1, University of California, Berkeley. http://www-bsac.eecs .berkeley.edu/archive/users/warneke-brett/is224/project1/paper1.html. Rudaheranwa, N., F. Matovu, and W. Musinguzi. 2003. "Enhancing Uganda's Ac- cess to International Markets." In Standards and Global Trade, J. S. Wilson and V. O. Abiola, 371­426. Washington, DC: World Bank. Santos, S. R. 2002."Análise crítica de um método para implementação das normas ISO 9000 em pequenas e médias empresas brasileiras." Universidade Federal do Rio Grande do Sul, Escola de Engenharia, Porto Alegre, Brazil. Schmitz, H. 1999. "Global Competition and Local Cooperation: Success and Failure in the Sinos Valley, Brazil." World Development 27 (9): 1627­50. Schmitz, H., and P. Knorringa. 2000. "Learning from Global Buyers." IDS Work- ing Paper 100, Institute of Development Studies, University of Sussex, Brigh- ton, UK. Schuurman, H. 1997. Quality Management and Competitiveness: The Diffusion of the ISO 9000 Standards in Latin America and Recommendations for Gov- ernment Strategies. Desarollo Productivo Series, no. 41. Santiago, Chile: Eco- nomic Commission for Latin America and the Caribbean. Sharma, D. S. 2005. "The Association between ISO 9000 Certification and Finan- cial Performance." International Journal of Accounting 40 (2): 151­72. Singels, J., G. Ruël, and H. van de Water. 2001. "ISO 9000 Series Certification and Performance." International Journal of Quality & Reliability Management 18 (1): 62­75. Stephenson, S. 1997. "Standards, Conformity Assessment and Developing Coun- tries." Policy Research Working Paper 1826, World Bank, Washington, DC. Stevenson, T. H., and F. C. Barnes. 2001. "Fourteen Years of ISO 9000: Impact, Criticisms, Costs, and Benefits." Business Horizons 44 (3): 45­51. Swann, G. M. P. 2000. "The Economics of Standardization." Final Report for Standards and Technical Regulations Directorate, UK Department of Trade and Industry. Manchester Business School, Manchester, UK. Swann, G. M. P., P. Temple, and M. Shurmer. 1996. "Standards and Trade Perfor- mance: The UK Experience." Economic Journal 106: 1297­1313. Terlaak, A., and A. A. King. 2006. "The Effect of Certification with the ISO 9000 Quality Management Standard: A Signaling Approach." Journal of Economic Behavior and Organization 60 (4): 579­602. Terziovski, M., D. Samson, and W. Dow. 1997. "The Business Value of Quality Management Systems Certification: Evidence from Australia and New Zea- land." Journal of Operations Management 15 (1): 1­18. References 277 Tschang, T., A. Amsden, and S. Sadagopan. 2001. "Measuring Technological Up- grading in the Indian Software Industry:A Framework of R&D Capabilities and Business Model." Working paper, Asian Development Bank Institute, Tokyo. Turner, C., G. F. Ortmann, and M. C. Lyne. 2000. "Adoption of ISO 9000 Quality Assurance Standards by South African Agribusiness Firms." Agribusiness 16 (3): 295­307. UNCTAD (United Nations Conference on Trade and Development). 2002. UNCTAD Handbook of Statistics 2002. New York: UNCTAD. ------. 2004. UNCTAD Handbook of Statistics 2004. New York: UNCTAD. ------. 2005. Trade and Development Report 2005. New York: UNCTAD. ------. 2006. UNCTAD Handbook of Statistics 2006. New York: UNCTAD. UNIDO (United Nations Industrial Development Organization). 2003. "Labora- tory Accreditation in Developing Economies: Tested Once, Accepted Every- where." Working Paper 2, UNIDO, Vienna. Vargas, M. A., and R. M. Alievi. 2003. "Learning Trajectories and Upgrading Strat- egies in the Footwear Productive System of the Sinos Valley/RS." In Systems of Innovation and Development, ed. J. E. Cassiolatto, H. M. M. Lastres, and M. L. Maciel, 352­75. Northampton, MA: Edward Elgar. Wallace, W. 2004. "African Farmers Dig Deep to Comply with EU Food Rules." Financial Times, April 7, 16. Wayhan,V. B., E.T. Kirche, and B. M. Khumawala. 2002."ISO 9000 Certification:The Financial Performance Implications." Total Quality Management 13 (2): 217­31. Wilson, J. S., and V. O. Abiola. 2003. Standards and Global Trade: A Voice for Africa. Washington, DC: World Bank. Wilson, J. S., and T. Otsuki. 2004."Standards and Technical Regulations and Firms in Developing Countries: New Evidence from a World Bank Technical Barri- ers to Trade Survey." Draft. Poverty Reduction and Economic Management, Development Economics, World Bank, Washington, DC. Wilson, J. P., M. A. T. Walsh, and K. L. Needy. 2003. "An Examination of the Economic Benefits of ISO 9000 and the Baldrige Award to Manufacturing Firms." Engineering Management Journal 15 (4): 3­10. Wood, A. 1997. "Openness and Wage Inequality in Developing Countries: The Latin American Challenge to East Asian Conventional Wisdom." World Bank Economic Review 11 (1): 33­57. World Bank. 2004. "Peru's Second Programmatic Decentralization and Competi- tiveness Structural Adjustment Loan." World Bank, Washington, DC. ------. 2005a. "Bolivia: Country Economic Memorandum." Latin America and the Caribbean Region, World Bank, Washington, DC. 278 Quality Systems and Standards for a Competitive Edge ------. 2005b. Global Economic Prospects 2005: Trade, Regionalism and Develop- ment. Washington, DC: World Bank. ------. 2005c. World Development Indicators 2005. Washington, DC: World Bank. ------. 2006a. "Mexico's Competitiveness: Reaching Its Potential." Report 35388-MX, World Bank, Washington, DC. ------. 2006b. World Development Indicators 2006. Washington, DC: World Bank. WTO (World Trade Organization). 1994. Agreement on Technical Barriers to Trade. LT/UR/A-1A/10. Geneva. ------. 2005. World Trade Report 2005: Exploring the Links between Trade, Stan- dards and the WTO. Geneva: WTO. Web sites APLAC (Asia-Pacific Laboratory Accreditation Cooperation): http://www .aplac.org BIPM (International Bureau of Weights and Measures): http://www.bipm.fr CITAC (Cooperation on International Traceability in Analytical Chemistry): http://www.citac.cc CODEX: http://www.codexalimentarius.net/web/index_en.jsp COMPITE: http://www.compite.org.mx Contactopyme: http://www.contactopyme.gob.mx COPANT (Pan American Standards Commission): http://www.copant.org EA (European Co-operation for Accreditation): http://www.european -accreditation.org European Commission, Nando Information System: http://europa.eu.int/comm/ enterprise/nando-is IAAC (InterAmerican Accreditation Cooperation): http://iaac-accreditation.org IAF (International Accreditation Forum): http://www.iaf.org IEC (International Electrotechnical Commission): http://www.iec.ch ILAC (International Laboratory Accreditation Cooperation): http://www.ilac .org INN (Instituto Nacional de Normalización): http://www3.inn.cl/ IRCA (International Register of Certified Auditors): http://www.irca.org References 279 ISO (International Organization for Standardization): http://www.iso.ch ITU (International Telecommunication Union): http://www.itu.int Macaws SRL: http://www.macaws.net/ Mexican Ministry of the Economy: http://www.economia-iso9000.gob.mx OAS Foreign Trade Integration System (SICE): http://www.sice.oas.org OIML (International Organization of Legal Metrology): http://www.oiml.org PAC (Pacific Accreditation Cooperation): http://www.apec-pac.org. QSU Publishing Co.: http://www.qsuonline.com/ SADC Web site on Standardisation, Quality Assurance, Accreditation, and Metrology: http://www.sadc-sqam.org SADCA (SADC Accreditation): http://www.sadca.org TSE (Turkish Standards Institution): http://www.tse.org.tr TÜRKAK: http://www.turkak.org.tr WTO (World Trade Organization): http://www.wto.org Index Citations to material in the endnotes are adaptation, 43­44 italicized. costs, 33, 261n.4 advocacy, 214, 202 A Africa, horticultural market, 44, 45 American National Standards Institute accreditation, 14, 66, 69­74, 204­207 (ANSI), 57­59 activities, 173, 177­180, 243­246 APEC, 142, 144, 265n.4 budget, 248 application programming interfaces confidence, 82 (APIs), 25­26 costs, 71, 225, 248 Argentina establishing, 70­71 certification costs, 112 function of, 69­70 quality promotion initiatives, 199 governance and autonomy, 172­175, assessment 265n.1 second-party, 64, 65 government support, 205­207 third-party, 64, 65, 68 guidance, 70 assessors, 175 institutions, 14, 69­70, 70­74, auditors, 161, 213 171­177, 204, 219, 246­248 automobile sector, 157­158 international integration, 180­182 certification, 158, 265nn.1,2 international recognition, 92­94, 207 autonomy Latin America, 171­184 accreditation institutions, 172­175, legal status, 172 265n.1 markets, 205­206 standards-setting institutions, 131­132 Mexico, 219, 224­228 organization, 171­172, 204­205, 244, B 245, 246 policy recommendations, 248­249 best practices, accreditation bodies, 181 private sector participation, 173 Bolivia, leather PDA covers, 7 process, 71 Brazil regional cooperation, 72 certification, 109, 112 resources, 175­177 footwear, 5, 43 scope, 70 quality promotion programs, 193­194 staff, 174, 247 British internationalization, 89 time required, 176­177 budget, 60, 202 Turkey, 234­235, 243­249, 259 business environment indicator, 167 acquis communautaire, 242 certification rates and, 168 281 282 Quality Systems and Standards for a Competitive Edge buyer-driven value chains, 41­42, 42­43, coffee, quality Uganda, 20­21 44­45 Colombia, quality promotion programs CYGA, 194­195 C other, 195­196 Colombian Fund for the Modernization calibration, 10, 68­69, 189­190, 262n.5 and Technological Development certificates, 95 of Micro, Small and Medium demand for, 212­215 Enterprises (FOMIPYME), laboratories, 13, 92, 178­179, 180, 195, 196 263n.6 Committee for Standardization (CEN), Mexico, 219, 222 87, 89 number offered, 190 Committee on Standards-Related Turkey, 236 Measures (CSRM), 146 see also measurement; testing communication, 214 Canada, legal metrology, 79­80 compatibility 16­17, 18, 20­21 Center for Quality, Safety and competition, 3­4, 6, 203, 216 Productivity (QSP), 193­194 constraints to, 25­26 certification, 10, 65­68, 82, 262n.1 increased, 21­24 accreditation, 204 ISO 9000 and, 99­105 barriers, internal firm-level, 105­114 mobile phones, first- and second- benefits, 66, 68, 228 generation, 22­23 bodies, 12, 65, 92, 262n.2, 263n.6 quality and, 4 consulting services, external, 109 compliance, 215 costs, 110­113, 264n.3 evaluation, 63­80 demand for, 212­215 Mexico, 226 financial performance and, 102­103 concept modeling, 126 firm performance and, 108 confidentiality, certification bodies, 66 market share, 160 conflicts of interest, 73 Mexico, 219, 228­232 Conformity Assessment Framework, objectivity, 66, 69, 73­74 63­80 organizational performance and, conformity assessment, 10, 81­96, 100­102, 264n.2 236­237 performance, 153­155 incompatible procedures, 81­82 policy recommendations, 243 Mexico, 230, 232 cost, 229 principles, 65­66 consolidation, national quality system, prior, certification and, 120 203­204 quality, 160 Consorcio del Jamón Serrano Espanõl, services, 241­243, 159­161, 266n.6 quality control, 67 time-series studies, 101­120 coordination failures, 214 Turkey, 233, 234, 236­243, 259, COPANT, 144­145, 265n.5 266nn.1,4,5 costs, 19, 80 why firms get certified, 99­100 accreditation, 71, 225, 248 certified pesticide-residue tests, 47 adaptation, 33, 261n.4 CERTIpyme, 196 certification, 68, 110­113, 264n.3 Chile conformity evaluation, 229 quality promotion initiatives, 198­199 consulting, 112 wine market, 5 Latin America, 139 Chilean Economic Development Agency measurement activities, 27 (CORFO), 198­199 metrology, 78, 80 China, certification and export Mexico, 229 performance, 106­107 reductions, 24 Code of Good Practice, 84, 263n.3 Index 283 regional coordination, 91 F registration, 112­113 feedback, 20­21 standards, 53, 90, 139 fees. See costs trade, 34 financial performance, certification and, credibility, metrology, 95 102­103 cultural factors, certification and, 123, financial resources, limited, 109­113 125­126 first-party assessment, 64 flexibility, 83, 262n.2 D food standards, 35­36 decentralized approach, 202 footwear, 5, 43 de facto standards, 51 foreign bodies de jure standards, 52 accreditation, 241, 266n.7 disputation, legal metrology and, 79 certification, 159­160, 181 diversification, 6 foreign direct investment, 163 duplication, certification and testing, 82 foreign markets, standards and, 35 foreign multinationals, 119, 264n.6 E foreign technical regulations, 34 formal standards, 52 economic environment functional upgrading, 42 certification and, 119­126 funding, standardization, Latin America, ISO 9000 diffusion and, 165­170 132­134 economic growth, standards and, 28­32 economies of scale, 17 G education certification and, 120­126 GDP, exports and, 3, 162 ISO 9000 diffusion and, 170 Genera Bureau of Standards (DGN), 129 efficiency, productive and innovative, Germany 17­18 economic growth, 31­32 EMA, 225­227 standards, 36, 38 engineering tolerances, 76 globalization, 2, 216 environment, 79 global production networks, 6 environmental management systems governance, standards-setting institutions, (EMSs), 182 131­132 certificates, 149­151 government, 8 equivalence, 83 accreditation, 205­207 EurepGAP certification, 47, 48 agency or unit, single, 202 European Union certification, 119, 120, 243 food safety 36 fiscal functions, 79 grapes, 46 metrology, 208 market, access to, 236­237 regulations, 119 New Approach directives and role, 213­215, 119, 243 Turkey, 242 standardization, 54­56 standards, 88­89 standards bodies, 60 evaluation, 203­204, 215 Turkey, 243 export performance withdrawal, 202 certification and, 105, 106­107 grapes, market requirements, 46­49 ISO 9000 diffusion, 163­165, 265n.3 exports H certification and, 162 ham, cured, 67 GDP and, 3, 162 harmonization, 36, 37, 40, 83, 88­89, quality and, 40 89­90 284 Quality Systems and Standards for a Competitive Edge Hazard Analysis and Critical Control International Organization for Point (HACCP), 195 Standardization (ISO), 83, 84, 86, health, 79 92, 93, 263n.7 health care, metrology, 77 guides and standards for accreditation horticultural value chain, upgrading, bodies, implementation, 182 44­45 ISO 9000, 97­126 human capital, 202 barriers, internal firm-level, 105­114 accreditation, 205, 247 benefits, 99 metrology, 208 certifications, 161 standards, 135 competitive advantage and, 99­105 diffusion, 114­126, 162­170 I firm size, 166 function of, 97­99 incentives, 213­214 global adoption, 98 INDECOPI, 129, 264nn.1,2 Latin America and the Caribbean, independence, certification bodies, 66 149­150, 151­153 India manufacturing, 156 grapes, market requirements, Mexico, 166, 231, 232 46­49 regional shares, 114 software, certification, 120­126 role, 103­105 indirect network externalities, 20­21 standardized, 155, 156 industrialization, 41, 261n.7 Turkey, 237­241, 243, 266nn.4,5 industry ISO 14000, 155 consortia, 51 Latin America and the Caribbean, standards and, 39 150, 151­153 information, 16, 55 manufacturing and services, 156 asymmetries, 19 number, 157 certification and, 124 ISO/TS 16949:2002, 158, 265n.1 imperfect, 19 standards development, participation program, 202 in, 142 provision, 28 International Accreditation Forum (IAF), infrastructure, metrology, 78 93, 181, 182, 207 innovation membership, 92 constraints, 24­25 International Bureau of Weights and diffusion of, 18 Measures (BIPM), 191 growth and, 31 International Committee of Weights and standardization and, 28 Measures (CIPM), 95­96, 251, inspection, 10, 68­69, 219, 228­232 267n.10 bodies, 13, 92, 179, 263n.6 international cooperation, 84­91, Mexico, 219 182­184 number of accreditation, 179 International Electrotechnical Turkey, 236­243 Commission (IEC), 83, 84, 86, 92, institutional environment 93, 263n.7 certification rates and, 169 international integration, 81­96, 180­182 indicator, 168 accreditation, 207 ISO 9000 diffusion and, 167­168 Latin America, 139­147 InterAmerican Accreditation metrology, 209 Cooperation (IAAC), standardization, 211 182, 207 International Laboratory Accreditation InterAmerican Metrology System Cooperation (ILAC), 93, 181, (SIM), 191 182, 207 interchangeability 16, 261n.1 membership, 92 Index 285 international laboratory comparisons, mandatory standards, 33, 52, 56, 60­61, 191­192 211, 261n.5 International Organization of Legal Latin America, 135­136 Metrology (OIML), 96, Mexico, 220 293­294nn.9,10 manufacturing, 2­4, 261n.1 international organizations calibration, 250 membership, 184, 192 exports, global vs regional, 3, 4 MRAs, 183 ISO 9000 diffusion, 165 international recognition, 202 ISO 9001 certificates, number, 155 international standards, 84 Turkey, 240, 250 accreditation, 180­182 value added, 154­155 Latin America, 139­141 market factors membership, 139 certification, 115­119 national standards and, 140 ISO 9000 diffusion, 162­165 participation, 141­142 markets, 6 process, 87 failures, 55 International Telecommunication Union forces, 53 (ITU), 84, 86 player types, 118­119 international trade, standards and, 32­41 requirements, changing, 46­49 investment climate, Turkey, 234, 235, sponsored standards, 52 266n.2 standards, 51­52 measurement standards, 13, 16, 26­27, 74 J key and supplementary, 251 laboratories, 92, 263n.6 jurisdiction, 51­61 traceability, 64, 75, 208, 262n.7 see also calibration L Measurement Canada, 79­80 labor productivity, UK, 30 Mercosur, 145­146, 265n.6 laboratories metrology, 74­80, 207­209 accreditation process, time required, 177 activities, 189­190 international comparison, 191­192 costs, 78 Latin America government support, 208 accreditation, 171­184 health care, 77 certification, 149­170 infrastructure, 76 metrology, 185­192 institutional framework, 250­251 registration costs, 113 international cooperation and standardization, 127­147 recognition, 190­192 law international integration, 209 rule of, 203 Latin America, 185­192 NMI, 78 legal, 78­80, 96, 185­189 leather personal digital assistants (PDA) manufacturing, 250 covers, 7 Mexico, 219, 222, 224 legal metrology, 78­80 organization, 207­208 benefits, 79­80 policy recommendations, 251­252 institutions, 185­189 scientific, 74­78, 185­189 international cooperation, 96 Turkey, 234­235, 249­250, 259 organization, 80, 185­189 see also national metrology institutes Mexico M accreditation 219, 224­228 certification, testing, calibration, and management system certification bodies, inspection, 219, 228­232 number of accreditations, 178 framework, 219­220 286 Quality Systems and Standards for a Competitive Edge Mexico--continued principles for implementation and ISO 9001 adoption and firm size, 166 reform, 201­203 metrology, 219 schematic, 11 national quality system, evaluation, see also quality 219­232 National Standardization Institute quality promotion initiatives, (INN), 129 197­198 National Training Service (SENA), standardization, 219 195­196 Microsoft, 25­26 network effects, exploitation of, 19­20 Middle East, grapes, 46 nondiscrimination, 83 minimum quality standards, 17 notified bodies, Turkey, 242 minimum safety standards 17, 19 mobile phones, 22­23, 84­85 O multilateral recognition arrangements organizational performance, certification (MLA), 93, 94 and, 100­102,105, 264n.2 mutual recognition agreements (MRAs), time-series studies, 101­102 83, 93, 94, 95­96, 181­184, 207, output growth, UK, 30 263n.8 oversight, 215 membership, 184 metrology, 209 P Mexico, 224, 227 separate and individual, 96 pallets, country-specific standards and cost, 34 N Peru quality promotion initiatives, 196­197 NAFTA, 146 standards-setting institutions, 264nn.1,2 National Committee on Productivity physical realization, 74­75 and Technological Innovation policies, 193­199 (COMPITE), 197, 266n.2 metrology, 208 participation and subsidies, 198 political commitment, 202 National Metrology Center (CENAM), PRECISO, 196, 197 219, 222, 223 prices national metrology institutes (NMIs), 13, developing country vs developed 74­76, 185­192, 207­208 country, 4 accreditation, 187­188 manufactures, 3 costs, 78 quality and, 5 funding, 77­78 primary standard, 74­75 international recognition, 95­96 private sector, 10, 202 metrology, 209 accreditation, 173, 204­205 scope, 76 standards, 131 see also metrology process upgrading, 42 national metrology laboratories producers accreditation, 188 national quality system, 9­10 number of measurement areas, 189 standardization, 55 National Program for Quality Assurance product upgrading, 42 (PNAC), 195 production national quality systems elasticities, 32 components, 12­14 networks, 216 consolidating, 203­204 productivity establishment, 1­2 quality standards and, 32 organization of the book, 11, 14 labor, 30 overview, 9­11 Index 287 property rights, 203 regional organizations, 143 public and nonprofit institutions, 10 MRAs, 183 public safety, 79 regional standards, 84, 87­89, 211­212 PYMEvaluación, 196 regional trade agreements (RTAs), 90­91, 263n.5 Q registration costs, 112­113 registry of firms, 213 QS 9000, 158, 265nn.1,2 regulators, certification and, 68 quality, 2, 8 reproduction, 75 awareness and understanding, 105­107, reputation, 66, 68, 262n.4 109, 258, 260 research and development (R&D), 76, certification and, 104, 122­124 124­125 competition and, 4 resources, standardization, 210 control, 67 risk reduction, 70 defining, 9 regional coordination, 91 economic impact and effect, 15­49 market crises, 4 S producers, 43 SMEs, 8 safety standards, 17, 19 standards, 20­21 school enrollment, certification rates support services, 113­114 and, 169 Turkey, 258, 260 scientific capabilities, 188­189 see also national quality systems scientific metrology, 74­78 quality management, 97 activities, 187 certification and, 105, 181 organization, 185­189 consultants, Turkey, 243 scientists and technologists, standards, 97­126 standardization, 210 quality management systems (QMSs), 98, search costs, 19 264n.1,182 sectors, certification, 116­118, 264n.5 efficiency and effectiveness, 98 services, 6 implementation, 217 supply, limited, 113­114 reform, 216 value added, 154­155 requirements, 98 SGS, Peru, 196­197 standards, 118, 237­241 shoe industry, 5, 43 Turkey, 237­241 small and medium enterprises (SMEs), 6, Quality and Environmental Management 8, 202, 214­215 Program (CYGA), 194­195, certification and, 104 265n,1 cost reductions, 24 QWERTY keyboard, 25, 54, 262n.1 GDP and certification rate, 166 ISO 9000 diffusion, 165 R Peru, 196­197 quality and 8 recommendations, 201­217 standards and quality as an entry point, accreditation, 248­249 41­49 certification, 243 Turkey, 240, 243 standardization, 256­258 software sector, certification and, 120­126 Turkey, 243, 249, 250, 251­252, 257­258 South Africa, timber 88 reference 16, 186, 202 Southern African Development regional accreditation bodies, 205 Community (SADC), regional agreements, 143 accreditation, 72 regional cooperation and coordination, 91, Spain, cured ham, 67 182­184 specialized services model, 126 standards upgrading and, 142­147 288 Quality Systems and Standards for a Competitive Edge SPS, implementation, 147 upgrading, 89­91, 142­147, 211 staff. See human capital voluntary, 52, 54, 55, 60­61, 128, standardization, 209­212 220, 255 development activities, 211 Standards and Quality Accreditation innovation and, 28 Program (ProCal), 199 institutional framework, 254­256 standards-setting institutions, 12, 56­61, international integration, 211 262n.2 Latin America, 127­147, 264n.2 autonomy, 267n.14 Mexico, 219, 220 centralized approach, 56­57 organization, 209­210 decentralized approach, 57­59 participation, 212 development, 56, 136­138, 252­254 policy recommendations, 256­258 funding sources, 132­134 process, 18 government support, 60 resources for, 132­134, 264n.2 Latin America, 128­138 stakeholders, 55, 73 legal status, 59 subsidizing, 267n.14 members, 57, 59 Turkey, 235, 252­259 organization, 128 standards, 2, 6, 8 performance, 136­139 accreditation, 180­182 principles, 59 age, 28 procedures, 59­60, 262n.3 British vs German, 36, 38 regional, 84, 87­89, 211­212 constraints, 28­29 staff, 135 coordinated approach, 90 Turkey, 252­254 cost, 139, 221 standards preparatory groups (SPGs), country-specific, 33, 34 Turkey, 255, 267n.13 defining, 9 subcomponents, 6 diffusion of, 138­139 support programs, 193­199 economic effects, negative, 24­27 economic effects, positive, 17­24 T economic growth and, 28­32 technical barriers to trade (TBT), 34­35 functions of, 15­17 implementation, 147 idiosyncratic, 36, 37 notifications, 146 international, 84, 87, 139­142, 180­182 WTO and, 83­84 international differences, 81 technical capabilities, 188­189 international trade and, 32­41 technical committees and subcommittees, Latin America, 138­141 86­87, 141, 265n.3 mandatory, 33, 52, 56, 60­61, 135­136, membership, 257 211, 220, 261n.5 technical experts, number, 176 membership, 139 technical regulations, 52 Mexico, 220 Latin America, 135 national, 140 technology number of, 40­41, 86, 137, 138, 252, adoption, coordination mechanisms, 267n.11 52­54 obsolete, 25, 211 diffusion, 29 participation, 141­142 telephony, 22­23, 84­85 process, 87 terms-of-trade index, 3­4 regional coordination, 142­147 testing, 10, 68­69, 262n.5 shared, 38­39 costs, 229 SMEs, 8, 41­49 demand for, 212­215 stock, 137, 254, 267n.12 Mexico, 219, 228­232 time-series investigation, 29 Turkey, 236­243 Turkey, Europe vs, 253 Index 289 testing laboratories, 12­13, 229, 266n.1 policy recommendations, 243, 249, accreditation, 178­179, 180 250, 251­252, 257­258 third-party assessment, 64, 65, 68 standardization, 235, 252­258, 259 timber, South Africa, 88 testing, 236­243 timing, 25, 109 TNC, ISO 9000 diffusion, 163 U total factor productivity (TFP), 29 Uganda, coffee industry, 20­21 total quality management (TQM), United Kingdom, 36, 38 103­104, 105 grapes, 46 traceability, 64, 75, 208, 262n.7 horticultural market, 44 trade, 1 labor productivity, 30 agreements, membership, 143 output growth, 30 barriers, 33, 81, 83, 226 United States, accreditation, certification and, 115­116, 264n.4 73, 262n.6 costs, country-specific 34 upgrading, 42 ISO 9000 diffusion, 162­165 functional, 42 standards and, 39­40 process and product, 42 Traditional Specialty Guaranteed standards upgrading, 89­91, (TSG), 67 142­147, 211 training, 195 value chain, 44­45 Mexico, 221 Turkey, 243 V transaction costs, 19, 79 transnational corporations (TNC), 6, 43, value chain 118­119 buyer-driven, 41­42, 42­43 transparency, 83­84 certification and, 121­126 TSE, 254­256 horticultural, 44­45 recommendations, 257­258 variety reduction, 16, 17, 261n.1 TÜRKAK, 242, 244, 246­249, 260, 266n.8 verification, 190 accreditations granted, 245 voluntary standards, 52, 60­61 Turkey bodies, 54, 55, 128, 255 accreditation, 234­235, 243­249, 259 Latin America, 128 calibration, 236 Mexico, 220 certification, 233, 234, 236­243, 259 Turkey, 254­256 inspection, 236­243 metrology, 234­235, 249­252, 259 W national quality system, evaluation, wine, Chile, 5 233­260 266n.3 World Trade Organization (WTO), TBT, national quality system, summary 83­84, 147 table, 259 E C O - A U D I T Environmental Benefits Statement The World Bank is committed to preserving Saved: endangered forests and natural resources. · 8 trees This book, Quality Systems and Standards for a Competitive Edge, is printed on 60 # New · 6 million BTUs of total Life Opaque, a recycled paper with 30% post- energy consumer waste. The Office of the Publisher · 727 pounds (CO2 equivalent) follows the recommended standards for of greenhouse gases paper usage set by the Green Press Initiative, · 3,019 gallons of waste water a nonprofit program supporting publishers in using fiber that is not sourced from · 388 pounds of solid waste endangered forests. For more information, visit www.greenpressinitiative.org. This excellent book clearly lays out why improving quality and standards is critical . . . It is particularly relevant to firms and policy makers in all developing countries and in particular for Latin American ones . . . [Its] virtue is that it not only explains why quality and standards have to be improved, but most importantly, it shows what policy makers and firms have to do, and how to do it. This is a must-read book. Carl J. Dahlman Henry R. Luce Professor of International Relations and Information Technology Georgetown University This very important and must-read book sets out a clear conceptual approach to quality and standards, taking full account of the cons as well as the pros. And it is extremely rich in its examples and insights from the real world. This is policy-oriented microeconomics at its best. Sir Nicholas Stern Head, Government Economic Service, United Kingdom This book provides a comprehensive tool for policy makers confronted with the challenges of building trade competitiveness in the new global economy. The authors are some of the World Bank's leading experts on private sector development in Latin America, and their insights on what works, what doesn't work, and why are indispensable for practitioners in the field. Danny Leipziger Vice-President, Poverty Reduction and Economic Management Network World Bank ISBN 0-8213-6894-X