79892 DIREC TIONS IN DE VELOPMENT Private Sector Development Clusters of Competitiveness Raj Nallari and Breda Griffith Clusters of Competitiveness Direc tions in De velopment Private Sector Development Clusters of Competitiveness Raj Nallari and Breda Griffith iv  © 2013 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW, Washington DC 20433 Telephone: 202-473-1000; Internet: www.worldbank.org Some rights reserved 1 2 3 4    16 15 14 13 This work is a product of the staff of The World Bank with external contributions. Note that The World Bank does not necessarily own each component of the content included in the work. The World Bank therefore does not warrant that the use of the content contained in the work will not infringe on the rights of third parties. The risk of claims resulting from such infringement rests solely with you. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, 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. Nothing herein shall constitute or be considered to be a limitation upon or waiver of the privileges and immunities of The World Bank, all of which are specifically reserved. Rights and Permissions This work is available under the Creative Commons Attribution 3.0 Unported license (CC BY 3.0) http:// creativecommons.org/licenses/by/3.0. Under the Creative Commons Attribution license, you are free to copy, distribute, transmit, and adapt this work, including for commercial purposes, under the following conditions: Attribution—Please cite the work as follows: Nallari, Raj, and Breda Griffith. Clusters of Competitiveness. Washington, DC: World Bank. doi:10.1596/978-1-4648-0049-8 License: Creative Commons Attribution CC BY 3.0 Translations—If you create a translation of this work, please add the following disclaimer along with the attribution: This translation was not created by The World Bank and should not be considered an official World Bank translation. The World Bank shall not be liable for any content or error in this translation. All queries on rights and licenses should be addressed to the Office of the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e-mail: pubrights@worldbank.org. ISBN (paper): 978-1-4648-0049-8 ISBN (electronic): 978-1-4648-0050-4 DOI: 10.1596/978-1-4648-0049-8 Cover photo: © Corbis. Used with permission of Corbis. Further permission required for reuse. Cover design: Debra Naylor, Washington, DC Library of Congress Cataloging-in-Publication Data Nallari, Raj, 1955– Clusters of competitiveness / Raj Nallari, Breda Griffith.     pages cm. — (Directions in development)   Includes bibliographical references. ISBN 978-1-4648-0049-8 (alk. paper) — ISBN 978-1-4648-0050-4 (ebk.)   1. Competition. 2. Economic development. I. Griffith, Breda. II. World Bank. III. Title.   HB238.N35 2013  338.6’048—dc23 2013022140 Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Contents Preface xi About the Authors xiii Abbreviations xv Chapter 1 Competition, Competition Policy, and Growth 1 Competition and Growth 1 Product Market Regulation and Economic Performance 3 Competition Policy 18 Conclusion 25 Notes 25 References 27 Chapter 2 Competitiveness and Its Indicators 31 Elements of Competitiveness 31 Defining Competitiveness 34 Price Indicators of Competitiveness 39 Nonprice Indicators of Competitiveness 45 Doing Business: Measuring Business Regulations 50 Conclusion 52 Notes 52 References 54 Chapter 3 National Competitiveness 57 Defining National Competitiveness 57 Competitiveness Rankings 59 Conclusion 74 Notes 74 References 74 Chapter 4 Innovation Policy for Competitiveness 77 Innovation: Definition and Measurements 78 Innovation, Growth, and Competitiveness 81 Policies for Innovation 86 How Can Government Help? 91 Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8   v   vi Contents Conclusion 102 Notes 102 References 104 Chapter 5 Competitiveness and Clusters 107 Background to Clusters 108 Cluster Initiatives 115 Policy Implications 124 Conclusion 126 Notes 127 References 128 Box 2.1 Price and Nonprice Indicators of Competitiveness: The Case of Armenia 40 Figures 1.1 Markups in Manufacturing and Nonmanufacturing 3 1.2 The Tree Structure of the Economy-wide PMR Indicator 11 1.3 PMR in Accession and OECD Countries, Aggregate Level, 2008 13 1.4 Decomposition of PMR in Accession Countries, 2008 14 2.1 Business Environment Quality: The Diamond 32 2.2 The Three Pillars of Trade Competitiveness 34 2.3 Outcomes and Determinants of Competitiveness 38 B2.1.1 Armenia’s Share of World Exports 40 B2.1.2 Nominal and Real Effective Exchange Rates for Armenia 40 B2.1.3 GDP Dynamics and Global Competitiveness Rankings for Armenia 41 2.4 The 12 Pillars of Competitiveness 47 2.5 Correlation between Doing Business Rankings and OECD Rankings of Product Market Regulation 51 2.6 Correlation between Doing Business Rankings and World Economic Forum Rankings on Global Competitiveness 52 3.1 REER Competitiveness Gains and Losses, Selected Countries, 2000–11 59 3.2 World Competitiveness Scoreboard 2011: Top 20 Economies 60 3.3 Competitiveness Trends, 2005–11 61 3.4 Trends in the GCI Innovation Pillar Score, 2005–11 67 3.5 Variation in Individual Economies’ Regulatory Environment 71 3.6 Regions Ranked by Strength of Legal Institutions and Complexity and Cost of Regulatory Processes 72 3.7 Impact of Regulatory Reform on Registration of New Firms 73 4.1 New-to-Market Product Innovators as a Percentage of Innovative Firms by R&D Status, 2004–06 79 Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Contents vii 4.2 Contribution of Innovation to Growth in Ghana and the Republic of Korea, 1960–2005 83 4.3 Major Technical Systems from the Middle Ages through the Present 85 4.4 Cross-Border Capital Inflows, 1985–2009 87 4.5 Global Growth in Real GDP, 2000–12 88 4.6 Innovation Accounts for a Large Share of Labor Productivity Growth, Percentage Contributions, 1995–2006 89 4.7 Determinants of Technology Upgrading in Developing Countries 92 4.8 Growing Innovation: The Government as Gardener 93 4.9 Patenting Activity of Young Firms, 2005–07 96 5.1 An Agribusiness Cluster 110 5.2 Determinants of Innovative Capacity 112 5.3 Productivity and the Business Environment 113 5.4 Dimensions of Clusters and Economic Policy 114 5.5 One Approach to Developing a Cluster Initiative 116 5.6 Policy Setting in which CIs Are Conducted 119 5.7 Level of Trust between Firms and between the Private and Public Sector 120 5.8 Objectives Considered Most Important for the CI 120 5.9 Target Industries Selected by Donors or Government for the Purposes of CIs 121 5.10 Cluster Strength 122 5.11 Cluster Strength by Initiator—Developing and Transition Countries 122 5.12 Entity Responsible for Initiating CI by Economy’s Underlying Level of Development 123 5.13 Influence in First Stage of Cluster Initiatives’ Operation, by Sector and Economy Type 124 Tables 1.1 Markups Estimates by Industry 4 1.2 Review of Studies Examining the Effect of PMR on Macroeconomic Outcomes 7 1.3 Regulation and Growth in GDP per Capita, Aggregate PMR 15 1.4 Regulation and Growth in GDP per Capita, Regulatory Domains 16 1.5 Regulation and Growth in GDP per Capita, Threshold Results 17 1.6 Administrative Barriers to Starting a Business, by Region 20 1.7 Structure, Conduct, and Performance of Selected Industries in Five Developing Economies 21 1.8 Persistence of Profitability in Emerging Markets and Advanced Markets 23 1.9 Competition Assessment Framework 24 Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 viii Contents 2.1 The Who, What, and How of Firms’ Competitiveness 36 2.2 Definition of Competitiveness and Its Underlying Elements 37 2.3 Price and Cost Indicators of Competitiveness 42 2.4 Price Measures of Competitiveness 44 2.5 Factors and Subfactors Comprising the National Environment (World Competitiveness Yearbook) 45 2.6 Subindex Weights and Income Thresholds for Stages of Development (Global Competitiveness Index) 46 2.7 Eleven Areas of Business Regulation Measured by Doing Business 50 3.1 Global Competitive Index for Top 10 Countries for 2011–12 versus 2005–06 and 2008–09 Rankings 62 3.2 Twelve Pillar Rankings for GCI Top 10 Countries, 2011–12 63 3.3 GCI and Pillar Rankings for North America and Europe Region, Selected Countries, 2011–12 64 3.4 GCI and Pillar Rankings for Asia and the Pacific Region, Selected Economies, 2011–12 65 3.5 GCI and Pillar Rankings for Latin America and the Caribbean Region, Selected Countries, 2011–12 66 3.6 GCI and Pillar Rankings for Middle East and North Africa Region, Selected Countries, 2011–12 68 3.7 GCI and Pillar Scores—Sub-Saharan Africa, Selected Countries, 2011 68 3.8 Top 10 Economies for Ease of Doing Business in 2013, versus 2012 and 2011 Rankings 70 3.9 Ten Economies Showing Most Improvement in Ease of Doing Business, 2011–13 72 4.1 Simple Innovation Indicators 79 4.2 Extent of Economic Growth beyond Growth Predicted by Rates of Capital Accumulation, Selected Economies, 1960–89 82 4.3 Conventional Breakdown of Sources of Growth, 1970–2000 83 4.4 Level of Productivity in Countries of Various Incomes, 1970–2000 84 4.5 Decomposition of the Predicted Growth in National Market Shares from an Estimated Empirical Model of Cross-Country Competitiveness, 1961–73 84 4.6 Percentage of Rural and Urban Population with Access to Clean Water, 1990 and 2004 89 4.7 Pace of Dissemination of Major Technologies, 1748–2000 90 4.8 OECD and World Bank Policy Principles for Innovation 94 4.9 Summary of Innovation Policy Elements 99 5.1 Four Types of Economic Agglomerations 108 5.2 Social Science Contribution to Understanding of Clusters 109 5.3 Regional Specialization—Clusters in the United States, 2008, Selected Geographic Areas 111 Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Contents ix 5.4 Composition of Regional Economies, United States, 2004 114 5.5 Comparison of Cluster Initiatives by Level of Economic Development 118 5.6 Possible Policy and Strategic Recommendations from a Cluster Initiative 126 Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Preface Competition, competitiveness, innovation, and growth are inherently linked and thus provide a compelling basis for policy analysis and recommendations. A favorable macroeconomy featuring sound policies and stable institutions is necessary for prosperity. Prosperity also depends upon competitiveness, which arises from the microeconomic foundations of a society. How companies operate and the quality of the business environment in which they compete are some of the microeconomic issues underlying competitiveness. The modern, knowledge-driven globalized economy is a product of innovation and ­ ­ competitiveness. Understanding the factors affecting the innovation decisions of firms and industries is critical for the design and ultimate success of policy. Competitiveness is a broad subject with applications at the level of the firm, industry, region, nation, and global economy. Each one of these aspects has a rich literature drawn on by academics and policy makers over a long period. This book seeks to present a broad overview of the main ideas underlying ­ competitiveness and its applications, highlighting and discussing in greater depth the topics that are of relevance currently. Specifically, the book draws out the experiences of and lessons for developing economies and examines in detail the role for policy. Chapter 1 addresses competition and competition policy. Competition is good for growth and is the hallmark of the market economy. For example, competition in product markets is an important determinant of economic growth. Competition can lead to gains in productivity, or more technically to multifactor productivity, that is, combined productivity gains in labor and capital. The extent of regulation in product markets is an indicator of how supportive an economy is of ­ competition. In general, tighter regulation is negatively associated with economic growth, while improved governance lessens the negative effects of regulation. As developing economies proceed with market-oriented reform, competition policy is critical in ensuring favorable efficiency and welfare benefits to society at large. Chapter 2 examines competitiveness by analyzing its many different indicators. Classifying competitiveness is difficult, although some broad ­ classifications provide a framework in which to discuss this aspect of the ­ ­ economy. Competitiveness may be examined across narrow and broad measures, macro and micro indicators, short-term and long-term indicators, and price and nonprice measures. A competitive economy is a successful economy, and Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8   xi   xii Preface economists have long grappled with the reasons why some economies grow or are more successful than others. The chapter begins by looking at what it means to be competitive in a national and international sense and then examines some narrow and broad measures of competitiveness before considering the ­ macroeconomic and microeconomic factors affecting the indicators of competitiveness. The chapter concludes with an examination of the international ­ surveys of competitiveness and their complementarity in presenting indexes of national competitiveness. Chapter 3 looks further at indexes of national competitiveness that describe international competitive performance. From these indexes, top-ranking ­ countries for competitiveness in 2012 are identified. The chapter addresses price and nonprice measures of competitiveness. It examines three surveys related to ­ ­ nonprice measures of competitiveness: those of the Institute for Management Development, the World Economic Forum, and the World Bank. The respective publications are the World Competitiveness Scoreboard, the Global Competitiveness Index, and the Ease of Doing Business Index. The real effective exchange rate (REER) based on unit labor costs and on inflation (as indicated by the consumer price index) is charted for a set of wealthy, highly productive ­ countries for competitiveness based on nonprice measures. Chapter 4 addresses innovation, an increasingly important aspect of competi- tiveness. Innovation is a leading characteristic of the modern knowledge-driven economy. It also plays a crucial role in developing economies wishing to catch up in economic growth and development to developed countries. It is the basis of sustainable economic growth and is critical for addressing the global challenges confronting the world today. Innovation is a major objective of national policy. The chapter highlights the role played by innovation in economic growth and competitiveness before moving to an examination of the elements for an innova- tion policy that contributes to the pursuit of competitiveness. Chapter 5 discusses competitiveness and clusters. Cluster development has been embraced by policy makers as a way of stimulating an area’s economic development and growth. Clusters are systems of interconnectedness between private and public sector entities. They are usually made up of a group of companies, suppliers, service providers, and associated institutions in a particular ­ field or industry. Policy has sought to encourage cluster development through government involvement in cluster-based competitiveness projects. Other policies in science and technology, regional policy, and industrial policy also have ­ ­ implications for cluster development. The chapter looks at the background of cluster development and competitiveness and then addresses cluster initiatives and cluster-based competitiveness projects. It also examines public policy in this area. Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 About the Authors Raj Nallari is the practice manager for the growth and competitiveness practice at the World Bank Institute (WBI). He has worked at the World Bank for more than 20 years in various departments. Previously he worked at the International Monetary Fund. Raj has published on various topics, including growth adjust- ment systems, the labor market and gender, and macroeconomics. He has also edited several volumes of Development Outreach. He holds a PhD in economics from the University of Texas at Austin. Breda Griffith has worked as a consultant with the WBI since 2005 in the areas of growth, poverty, gender, development, and labor markets. She has publications in refereed journals on development and language maintenance, entrepreneur- ship, and small business. Breda has also coauthored books on economic growth, poverty, gender and macroeconomic policy, new directions in development, labor markets in developing countries, and geography of growth. She has developed and facilitated e-learning courses in these areas. She holds a PhD in economics from Trinity College Dublin, Ireland, and an MA in economics from the National University of Ireland. Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8   xiii   Abbreviations ASEAN Association of Southeast Asian Nations BMA Bayesian model averaging CAF Competition Assessment Framework CDM Clean Development Mechanism CEEP Centre for Public Enterprises with Public Participation CI cluster initiative CPI consumer price index ECB European Central Bank FDI foreign direct investment GCI Global Competitiveness Index GDP gross domestic product GNI gross national income HCIs harmonized competitiveness indicators ICT information and communication technology IFS International Financial Statistics IMD International Institute for Management Development IMF International Monetary Fund ISIC International Standard Industrial Classification IT information technology M&A mergers and acquisitions MFP multifactor productivity NIEs newly industrialized economies OECD Organisation for Economic Co-operation and Development PMR product market regulation R&D research and development REER real effective exchange rate SAR special administrative region SIC standard industrial classification SIPIs social infrastructure and political institutions Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8   xv   xvi Abbreviations SME small and medium enterprise SWOT strengths, weaknesses, opportunities, and threats TCD trade competitiveness diagnostic TFP total factor productivity WCY World Competitiveness Yearbook WEF World Economic Forum WTO World Trade Organization Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Chapter 1 Competition, Competition Policy, and Growth Competition is good for growth and is a prime characteristic of the market economy. For example, competition in product markets is an important determi- nant of economic growth. Competition can lead to one-time (static) and ongoing (dynamic) gains in productivity or more technically to multifactor productivity, that is, combined productivity gains in labor and capital. Examples of one-off gains are better resource allocation and less slack in the use of inputs. Dynamic gains are associated with greater tendencies to innovate and the distribution of innovation. The extent of regulation in product markets is an indicator of how supportive an economy is of competition. In general, tighter regulation is negatively associ- ated with economic growth, while improved governance lessens the negative effects of regulation. Thus, economic growth in higher-income countries is posi- tively related to deregulation, but at lower-income levels, we need to look more closely at the competition/regulation trade-off. The degree of competition is important. Competition stimulates economic growth within the overall macroeconomy. As developing economies proceed with market-oriented reform, competition policy is critical in ensuring favorable efficiency and welfare benefits to society at large. Competition policy has an important role to play in promoting growth. Furthermore, competition policy affects competitiveness, domestically and internationally. The following sections examine the relationship between competition and growth and focus on the policy aspects of this relationship. Competition and Growth market. Competition in economics is defined as free entry and exit of firms in any ­ The theoretical literature on competition and economic growth is prolific, with many facets, yet devoid of any firm conclusion (Yun 2004). Hence, the empirical Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8   1   2 Competition, Competition Policy, and Growth studies have been useful in shedding light on the relationship between c­ ompetition and growth. One area of study has been the impact of product market competition on productivity gains, with key studies in this area confirming a positive relationship between competition and productivity growth.1 Product market competition is but one factor among many that affect aggregate performance indicators, such as employment and productivity. Nevertheless, work by the Organisation for Economic Co-operation and Development (OECD 2002, 155) “has identified an empirical connection between strong competition in markets for goods and ser- vices and better productivity and employment outcomes.� Moreover, differences in competitive pressures play an important part in explaining the differences in productivity across countries. Competition affects per capita growth through its effect on productivity. As noted, increases in productivity arise from both static and dynamic e ­ fficiency gains. Regulatory reform that stimulates managerial effort is an ­ example of a static efficiency gain.2 Medium- and long-term gains in p ­ roductivity—dynamic efficiency gains—arise from investments in research and development (R&D), product and process innovation, and the associated buildup of human capital (Høj et al. 2007). There is an established empirical relationship between innova- tion and growth, with some dispute on the effects of competition on innovation.3 The long-run effects of competition are likely to be positive on aggregate labor productivity growth. Lower price-cost margins arising from increased competi- tion lead to job creation and upward pressure on average real wages. Further, lower product-market rents imply lower wage premium in some sectors, thus reducing labor costs and encouraging job creation. An increase of 1.5–2.5 percent in the employment rate may be observed where in-depth reforms have been adopted (Høj et al. 2007). On this point, see Høj et al. (2007), who refer to the work of Alesina et al. (2005), Nicoletti and Scarpetta (2003), and Conway et al. (2006).4 Høj et al. (2007) examine the data on price-cost margins (markups) for 17 countries in the OECD as a measure of competition. Figure 1.1 shows the results for the four different groups of manufacturing industries identified in table 1.1. These industries were identified along two dimensions5—the level of exogenous sunk costs (which identifies whether industries are fragmented or segmented) and the level of endogenous sunk costs (low or high R&D and advertising expenditures) and the markups for nonmanufacturing industries. The cross-country mean of markups for the four manufacturing industries, shown in figure 1.1, are not statistically different from one another. Greater variation is evident in the nonmanufacturing industry markups—average markups are estimated to be below 20 percent in the United Kingdom, Sweden, ­ and the United States; higher for most European countries; and highest for the Republic of Korea (32 percent) and Italy (38 percent) (Høj 2007). Table 1.1 sheds light on the underlying data. Høj et al. (2007) attribute the greater variability in nonmanufacturing indus- tries to a relative absence of competitive pressures in services compared to Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Competition, Competition Policy, and Growth 3 Figure 1.1 Markups in Manufacturing and Nonmanufacturing a. Manufacturing 0.40 0.35 0.30 0.25 Markup 0.20 0.15 0.10 0.05 0 LUX JPN BEL DNK SWE GBR KOR USA FRA DEU NLD NOR ESP CAN AUT ITA b. Nonmanufacturing 0.40 0.35 0.30 0.25 Markup 0.20 0.15 0.10 0.05 0 GBR SWE USA CAN BEL LUX NLD DEU DNK FRA NOR JPN FIN AUT KOR ITA Source: Høj et al. 2007, 53. Note: GBR = Great Britain; SWE = Sweden; USA = United States; CAN = Canada; BEL = Belgium; LUX = Luxembourg; NLD = Netherlands; DEU = Germany; DNK = Denmark; FRA = France; NOR = Norway; JPN = Japan; FIN = Finland; AUT = Australia; KOR = Korea, Rep.; ITA = Italy; ESP = Spain. Markups are calculated for individual 2-digit International Standard Industrial Classification (ISIC) sectors and aggregated over all sectors using country-specific final sales as weights. manufacturing and the diversity of competition policies being pursued in the sample countries. Product Market Regulation and Economic Performance There is a fairly extensive empirical literature on the macroeconomic effects of regulatory reform in the labor and financial markets, while “the area that has been comparatively under-researched is the effect of product market regulation (PMR) on macroeconomic outcomes, with the exception of the effect of barriers to trade� (Schiantarelli 2008). Nicoletti and Scarpetta (2005) concur and note Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 4 Table 1.1 Markups Estimates by Industry United Korea, United Austria Belgium Canada Germany Denmark Spain Finland France Kingdom Italy Japan Rep. Luxembourg Netherlands Norway Sweden States All manufacturing 0.15 0.10 0.15 0.13 0.11 0.14 0.18 0.12 0.11 0.15 0.09 0.12 0.07 0.13 0.13 0.11 0.12 Fragmented low-R&D industries: 0.16 0.11 0.19 0.14 0.13 0.16 0.21 0.12 0.12 0.18 0.08 0.12 0.04 0.16 0.13 0.10 0.13   Textiles, wearing apparel, leather 0.13 0.05 0.13 0.14 0.12 0.13 0.14 0.09 0.10 0.16 0.06 0.12 — 0.14 0.12 — 0.09   Wood and wood products 0.10 0.12 0.25 0.14 0.18 0.16 0.19 0.05 0.16 0.22 0.05 0.13 −0.03 0.08 0.11 — 0.19   Pulp, paper, printing, and publishing 0.19 0.14 0.20 0.19 0.10 0.18 0.23 0.13 0.12 0.18 0.10 0.11 0.10 0.20 0.14 — 0.13   Other nonmetallic mineral products 0.24 0.17 0.22 0.19 0.17 0.19 0.24 0.16 0.16 0.22 0.15 0.18 0.03 0.20 0.14 0.05 0.17   Fabricated metal products 0.13 — 0.13 0.07 — — 0.16 — — 0.19 0.01 0.10 0.03 0.11 0.13 0.12 0.12 Segmented low-R&D industries: 0.17 0.09 0.14 0.13 0.08 0.13 0.13 0.14 0.11 0.13 0.08 0.09 0.12 0.13 0.14 0.11 0.09   Food, beverages, and tobacco 0.13 0.09 0.13 0.12 0.08 0.13 0.10 0.14 0.11 0.14 0.07 0.05 0.13 0.12 0.08 0.08 0.09   Basic metals 0.25 — 0.17 0.18 — — 0.18 — — 0.10 0.10 0.14 0.11 0.24 0.27 0.16 0.08 Fragmented high-R&D industries: 0.16 0.12 0.13 0.13 0.11 0.15 0.17 0.17 0.12 0.16 0.09 0.11 0.06 0.13 0.09 0.13 0.10   Machinery and equipment 0.19 0.20 0.16 0.13 0.09 — 0.17 0.19 0.12 0.15 0.08 0.11 0.08 0.15 0.09 0.13 —   Other manufacturing and recycling 0.09 0.06 0.08 0.16 0.15 0.15 0.17 0.13 0.13 0.17 0.11 0.11 −0.05 0.10 0.09 — 0.10 Segmented high-R&D industries: 0.13 0.09 0.12 0.13 0.11 0.14 0.18 0.12 0.11 0.14 0.10 0.13 0.03 0.12 0.14 0.12 0.13   Chemical, plastics, rubber, and fuel products 0.11 0.09 0.12 0.16 0.11 0.17 0.15 0.11 0.12 0.13 0.10 0.14 0.03 0.13 0.18 0.15 0.15   Electrical and optical equipment 0.15 — 0.14 0.13 0.12 — 0.22 0.15 0.13 0.17 0.13 0.12 — 0.09 0.12 0.12 —   Transport equipment 0.14 0.09 0.13 0.09 0.08 0.11 0.17 0.11 0.07 0.14 0.08 0.11 0.02 0.09 0.11 0.08 0.10 Nonmanufacturinga 0.28 0.20 0.20 0.25 0.25 — 0.27 0.26 0.16 0.38 0.26 0.32 0.02 0.24 0.26 0.17 0.19   Electricity, gas, and water supply 0.34 0.23 0.35 0.37 0.41 — 0.37 0.27 0.15 0.30 0.46 0.32 — 0.19 0.48 — 0.20   Wholesale and retail trade, repairs 0.28 — 0.16 0.12 0.28 — 0.25 0.25 0.16 0.45 — — 0.24 0.30 0.24 — 0.14   Transport and storage 0.14 — 0.26 0.13 0.18 — 0.33 0.22 0.10 — 0.17 — — 0.21 0.27 0.18 0.16   Post and telecommunications 0.20 — 0.35 0.38 0.24 — 0.36 0.40 0.21 — 0.32 — — 0.26 0.29 — 0.28   Financial intermediation 0.37 — 0.14 0.18 0.35 — 0.34 0.20 0.21 0.32 0.27 — 0.21 0.33 0.34 — 0.25   Business services 0.27 — — 0.44 0.20 — 0.19 0.28 — — 0.16 — 0.19 0.12 0.16 0.14 0.20 Source: Høj et al. 2007, 43. Note: — = not available; R&D = research and development. Figures are averages using sectoral production as weights. Weights are country specific. a. Nonmanufacturing excludes construction, real estate activities, and personal services. Competition, Competition Policy, and Growth 5 that cross-country policy differences in PMR explain cross-country differences in economic performance, while the macroeconomic effects are likely to be signifi- cant because PMR extends to more and more industries and to changes in gen- eral purpose regulation. Competition in product markets is obviously stymied by overly strict regula- tion in product markets. We would therefore expect to see higher markups associated with less product regulation. Schiantarelli (2008) provides a review of cross-country experience that examines the effect of PMR and reform on markups, firm dynamics, investment, employment, innovation, productivity, and output growth. Specifically, PMR affects (1) the allocation of resources between sectors producing different goods; (2) the allocation of resources between firms with different productivity in each sector; (3) the productivity of existing firms; and (4) the pace of productivity growth by altering the incentives to innovate and by determining the speed with which new products and processes replace old ones (Schiantarelli 2008). Schiantarelli (2008) reviews the theoretical literature on each of these four facets, highlighting the ambiguity and caveats that arise, particularly in the studies of innovation and PMR. For example, the introduction of PMR would be expected to reduce monopoly profits. As these provide one of the main incentives for innovative activity, particularly in the creative destruction models championed by Schumpeter (1942), PMR may adversely affect the desire to innovate. Empirical research in this area is critical to understanding the impact of PMR. The empirical results on the relationship between innovation and competition suggest an inverted U-shaped pattern whereby ­ innovation is affected adversely by a very competitive or very monopolistic environment.6 Further lessons from the microeconometric studies suggest that greater com- petition has a positive effect on the level and growth of productivity;7 but in cases where greater firm productivity results in the firm increasing its market share and the market becoming more concentrated, then productivity is biased downward.8 However, where privatization is accompanied by regulatory reform, in particular for service sector firms, there are gains in productivity from increased competition. Other empirical studies have examined the dynamics of productivity by decomposing aggregate productivity growth in different compo- nents, such as a “within� component arising from productivity improvements in continuing firms; a “between� component due to the reallocation of resources between continuing firms; and a component due to entry and exit (Schiantarelli 2008). The results from these studies are mixed, depending on the level of decomposition used and the time frame. The following results are cited by Schiantarelli (2008): 1. Bartelsman, Haltiwanger, and Scarpetta (2004) find that the “within� compo- nent of labor productivity is the most important component for the developed and nontransition emerging countries. Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 6 Competition, Competition Policy, and Growth 2. Foster, Haltiwanger, and Krizan (2001) find that net entry, the third component, becomes important (and positive) only at a horizon of 5 to 10 ­ years. 3. Bartelsman, Haltiwanger, and Scarpetta (2004) find that entry is more impor- tant in transition countries, while being negative in most OECD countries and in the nontransition emerging economies. 4. The “between� component empirical studies have had mixed results (Schiantarelli 2008). Turning to the macroeconomic effects, Schiantarelli (2008) reviews the empiri- cal studies examining the effects of PMR on macroeconomic performance, that is, investment, employment, innovation, productivity, and output growth. Most of the empirical studies reviewed employ a reduced-form approach whereby PMR is an explanatory variable—either directly or indirectly through intermediate variables such as markup or firms’ entry, exit, or turnover rates—in equations for factor demand, productivity, or innovation (see table 1.2). An Example Using the OECD PMR Indicators The OECD indicators of PMR allow us to examine the relationship between competition as measured by markups and the rules and regulations that have the potential to reduce the strength of competition (Høj et al. 2007).9 The ­indicators, which measure the degree to which policies promote or inhibit competition in product markets, were constructed initially in 1998; they were updated in 2003 (the indicators were extended to include employment protection legislation), and again in 2008 (the indicators were substantially revised and the economy- wide PMR was extended to include greater sectoral information than hereto- fore). The most recent revision, in 2011, witnessed a move from “most-favored nation tariffs� to “effectively applied tariffs� and the inclusion of the Foreign Direct Investment (FDI) Regulatory Restrictiveness Index.10 The PMR indicators measure the economywide regulatory and market environments in 20 OECD countries in or around 1998, 2003, and 2008, and in 4 other OECD countries (Chile, Estonia, Israel, and Slovenia) as well as in Brazil, China, India, Indonesia, the Russian Federation, and South Africa in or around 2008. They are consistent across time and countries. The structure of the PMR indicator system is shown in figure 1.2. It takes a bottom-up approach, where the indicators can be related to specific under- lying policies (Conway et al. 2010). It summarizes a large number of formal rules and regulations that have a bearing on competition. Hence, it is an objective measure of the regulatory stance. The information is organized into 18 low-level indicators that are progressively aggregated into three broad regulatory areas: • State control of business enterprises • Legal and administrative barriers to entrepreneurship • Barriers to international trade and investment Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Table 1.2 Review of Studies Examining the Effect of PMR on Macroeconomic Outcomes Effect of PMR on Study Focus Data source Results Markup Griffith and Harrison Two-step approach: Fraser Institute (2002) index of ease of starting a Many of the indicators measuring (2004) Effect of product market reforms on new business, of price controls, of time spent with tightness of regulation have a the level of rentsa government bureaucracy, of average tariff rates, significant positive effect on Effect of variations in the markup and of regulatory trade barriers; European Center of markups. on factor accumulation R&D and Enterprises with Public Participation; and the Eurostat productivityb Structural indicators on state aid, public procurement, and on percentage that is publicly advertised Turnover Cincera and Two-step approach Dun and Bradstreet database on the number of entries Deregulation tends to be significantly Galgau (2005) Effect of regulation on entry and exit and exits for 352 digit sectors for 9 OECD countries associated with more entry and of new firms exit. Effect of entry and exit on factor demand and productivity Turnover Loayza, Oviedo, and Effect of firm turnover rates on Harmonized data set on firm dynamics constructed by PMR slows down the reallocation of Serven (2005) productivity growth and its Bartelsman, Haltiwanger, and Scarpetta (2004) for 6 resources following a shock. components Latin American countries and 9 industrial economies Entry/exit Scarpetta et al. (2002) Effect of regulation on entry OECD firm-level database constructed from business For firms employing between 20 and registers or social security databases 99 workers, PMR has a negative and significant effect on entry. For the 100 to 499 class, the effect is positive and significant Entry/exit Brandt (2004) Effect of regulation on entry OECD firm-level database constructed from business Barriers to entry coefficient are not registers or social security databases significant; some evidence of regulatory and administrative opacity affects entry rates. table continues next page 7 8 Table 1.2  Review of Studies Examining the Effect of PMR on Macroeconomic Outcomes (continued) Effect of PMR on Study Focus Data source Results Entry/exit Klapper, Laeven, and Effect of regulation on entry Cross-country firm-level Amadeus data set to construct Regulation reduces entry relative to Rajan (2004) entry and exit rates for Western and Eastern European the “normal� industry-specific rate countries one observes in the United States with low barriers to entry. Investment Alesina et al. (2005)c Effect of regulation on investment Time-varying sector country-specific measures of A reduction in regulation, particularly (by focusing on investment in regulation if it affects barriers to entry, has nonmanufacturing industries that a significant and sizable positive have experienced changes in their effect on the investment rate. regulatory framework) Employment Fiori et al. (2007)c Effect of regulation on service Employment rate equation for the business sector in Gains from reducing barriers to entry employment OECD countries, including country-specific constants in product markets are higher and controlling for endogeneity of policies when labor market policies are tight; domestic product market deregulation generates a decline in bargaining power of workers (by promoting deregulation of labor market/affecting union density and coverage). table continues next page Table 1.2  Review of Studies Examining the Effect of PMR on Macroeconomic Outcomes (continued) Effect of PMR on Study Focus Data source Results Factor demand Griffith and Harrison Effect of regulation on factor demand Not specified Decrease in regulation leads to (2004)c (employment and investment) via decrease in markup; markup is markup negatively and significantly related to employment and investment. Cincera and Effect of regulation on factor demand Not specified; instrumental variables used for entry/exit Entry is not a significant determinant Galgau (2005) (employment and investment) via because of endogeneity of the growth in investment, while entry exit is associated with a significant decrease in the pace of capital accumulation; for employment growth, the effect is not significant. Innovationd Bassanini and Effect of product and labor market 18 manufacturing industries in 18 OECD countries Nontariff barriers have a negative Ernst (2002) regulation (domestic economic effect on R&D intensity positive regulation, administrative differential effect for employment regulation, and tariffs and nontariff protection in high-tech industries barriers) on R&D intensity (relative relative to low-tech in centralized to output) systems of industrial relations. Griffith and Effect of PMR on R&D through Not specified Markup has a positive and significant Harrison (2004) changes in the markup effect on R&D. table continues next page 9 10 Table 1.2  Review of Studies Examining the Effect of PMR on Macroeconomic Outcomes (continued) Effect of PMR on Study Focus Data source Results Productivity Nicoletti and Effect of regulation on TFP growth 17 manufacturing and 6 service industries for 18 OECD Some evidence of a positive effect and output Scarpetta (2003) countries of privatization and entry growthe liberalization on TFP growth exists. Also evidence shows that entry barriers in manufacturing may affect the pace of technology absorption, especially for countries far from the world frontier. IMF (2004) Effect of regulation on per capita GDP 15 developed countries with a maximum of five Both product market reform and growth observations on growth rates calculated over 3-year trade reform have a positive and averages significant effect on growth, although it may take time for the full effects to be realized. Loayza, Oviedo, and Effect of product and labor market Both developing and developed countries A negative and significant direct Serven (2004) regulation on growth effect of product and labor market regulation on growth; better governance reduces the negative effect of regulation; overall effect of regulation is sizable and negative for most developing countries; zero or mildly positive for most developed. Source: Schiantarelli 2008. Note: GDP = gross domestic product; IMF = International Monetary Fund; OECD = Organisation for Economic Co-operation and Development; PMR = product market regulation; R&D = research and development; TFP = total factor productivity. a. As measured by the ratio between value added and the sum of labor and capital costs. b. Product market reform is an instrument for markup. c. Authors find a positive effect of deregulation on investment and employment in the service sector, but no evidence of a positive (or negative) effect for manufacturing. d. Schiantarelli (2008) concludes that the macroeconomic studies are not supportive of a strong positive effect of lower regulation on direct input measures of firms’ innovative activities. The evidence for manufacturing is sensitive to country selection in the sample. e. Most of the evidence points toward a positive effect of less stringent regulation on productivity growth; the effect of deregulation is larger in more developed countries, while better governance appears to mitigate the negative effects of regulation (Schiantarelli 2008). Figure 1.2 The Tree Structure of the Economy-wide PMR Indicator Product market regulation State control Barriers to entrepreneurship Barriers to trade and (0.33) (0.33) investment (0.33) Involvement Regulatory and Administrative Explicit barriers Public Barriers to Other in business administrative burdens on to trade and ownership competition barriers operations opacity start-ups investment (0.50) (0.33) (0.50) (0.50) (0.33) (0.33) (0.50) Legal barriers Scope of public Licenses and Admin. burdens Barriers Price (0.25) enterprise permits system for corporations to FDI (0.33) controls (0.33) Antitrust (0.33) (0.50) (0.50) Admin. burdens exemptions Government involvement for sole (0.25) Tariffs Regulatory in network sectors Communication proprietor firms (0.33) barriers Use of Barriers in (0.33) and (0.33) (1.0) command network Direct control simplification and control Sector. specific of rules and sectors (0.25) Discriminatory over business regulation administrative enterprises procedures procedures (0.50) burdens Barriers in (0.33) (0.50) (0.33) (0.33) services (0.25) Source: Wölfl et al. 2010. © OECD. Used with permission; further permission required for reuse. Note: FDI = foreign direct investment; PMR = product market regulation. Percentages in boxes refer to weights assigned. 11 12 Competition, Competition Policy, and Growth The overall PMR indicator is a summary statistic of the general stance of PMR (Conway et al. 2010). Wölfl et al. (2010) analyze the link between regulation (using the PMR) and growth for the OECD countries and non-OECD countries between 1998 and 2008. Their findings, based on growth regressions, suggest that less restric- tive PMR is conducive to growth. For example, an improvement of 0.5 index points of barriers to entrepreneurship11 would translate into approximately a percent higher average annual rate of gross domestic product (GDP) per 0.4 ­ capita growth. However, the authors suggest that for less advanced countries, benefits arising from product market competition may be offset by other structural weaknesses. Thus, at early stages of industrial development, greater PMR (through, for example, some restrictions on foreign trade and invest- ment) may be positive for growth.12 The following results arise from the study by Wölfl et al. (2010): 1. Regulation is more restrictive of competition in accession countries,13 enhanced engagement countries,14 and non-OECD countries,15 compared to OECD countries. a. Furthermore, regulation is most restrictive in enhanced engagement coun- tries compared to most accession countries (see figure 1.3). b. Regulatory settings among Estonia, Slovenia, and Romania (non-OECD countries) are closer to those of the OECD average, compared to China, Russia, Israel, and Ukraine, where regulation is more restrictive. 2. Breaking down PMR into regulatory domains—state control, barriers to entre- preneurship, and barriers to trade and investment—suggests that the accession countries also face more restrictive regulation across all domains. a. Israel and Ukraine are characterized by restrictive regulation across the areas of state control, barriers to trade and investment, and barriers to entrepreneurship. b. Russia and Croatia also experience relatively high state control and barriers to trade and investment, and state control is relatively high in Bulgaria and Romania as well (see figure 1.4). c. Barriers to entrepreneurship and barriers to trade and investment in Chile, Estonia, Slovenia, Bulgaria, and Romania are at a level close to the OECD average. 3. The higher level of state control in enhanced engagement countries is due mainly to more widespread public control of business enterprises and a stronger use of coercive instead of incentive-based regulations. The higher overall barriers to entrepreneurship in India and Brazil are attributable largely to substantial red tape when setting up enterprises; in Indonesia, they are due to restricted entry in a large number of sectors. An onerous licensing and permits system characterizes the substantial barriers to entrepreneurship in South Africa. Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Figure 1.3 PMR in Accession and OECD Countries, Aggregate Level, 2008 5.0 5.0 Integrated PMR 4.5 4.5 4.0 4.0 Accession 3.5 3.5 countries average 3.0 3.0 Index Index 2.5 2.5 2.0 2.0 OECD average 1.5 1.5 1.0 1.0 0.5 0.5 0 0 ce ile Ch a ra l n d om Ice tes Ne Ca nd er a nm s Sp d No ain Ne J ay Ze an Fi nd rm d n y Sw st y er a Be and S ium re en Fr ep. ce Po Italy us l bo tria ec ep rg pu ic M lic Tu ico Po key Gr nd ov a Ire ark de rae Lu A uga De land i Hu an Au gar th nad itz rali Sl oni tio n Ge lan en Re l ee Cz k R u h ub b an rw w ap Ko wed la la a la a ex R ite gd r Fe Is al l lg St t rt n a, Es Un Kin Sl xem d a ite n ov ia Un ss Ru Source: Wölfl et al. 2010. Note: OECD = Organisation for Economic Co-operation and Development; PMR = product market regulation. Figures are based on the “integrated� PMR indicator. Indicator values refer to one particular year and may no longer reflect the current regulatory stance in some (fast-reforming) countries. Confidence intervals are 90 percent and based on the random-weights approach. Index points are from 0 to 5 (least to most restrictive). 13 14 Competition, Competition Policy, and Growth Figure 1.4  Decomposition of PMR in Accession Countries, 2008 3.5 3.0 Index points 2.5 2.0 1.5 1.0 0.5 0 a e a il e n ni ni ag l tio ae Ch e to er ra ov Isr Es av de Sl ED Fe OC n ssia Ru Barriers to trade and investment Barriers to entrepreneurship State control Source: Wölfl et al. 2010. Note: Based on the “integrated� product market regulation (PMR) indicator. Indicator values refer to one particular year and may no longer reflect the current regulatory stance in some (fast-reforming) countries. Index points are from 0 to 3.5 (least to most restrictive). The Organisation for Economic Co-operation and Development (OECD) average is a simple average. Wölfl et al. (2010) investigate the relationship between PMR and growth in GDP per capita for 1998–2007 and for two subperiods, 1998–2003 and 2003–2007. The hypothesis is that PMRs impact economic growth, as summa- rized by the following equation: y = α + β ∗ PMR + δ ∗ X + ε, where y is the average annual GDP growth rate per person aged 16–64 over a particular time period; PMR is the PMR indicator at different levels of disaggre- gation; and X is a matrix of control variables. Table 1.3 shows the results from the growth regression for the cross-section of countries over the entire time period and the cross-section over the two nonoverlapping subperiods. The results are in line with cross-sectional growth regressions featuring conditional convergence in GDP per capita at an implied rate of 1.6 percent and the expected coefficients for investment (Wölfl et al. 2010). However, population growth and human capital are not significant in any of the three regressions—a fact that Wölfl et al. (2010, 20) attribute to the “smaller heterogeneity in terms of population growth and human capital among the ­ countries in the sample.� Aggregate product market regulation, the PMR variable, affects growth in the cross-section for the entire period only, ­ suggesting that those countries with relatively liberal regulation in 1998 grew faster in average GDP per capita over the subsequent decade. The coef- ficient estimate suggests that a reduction in the overall PMR by 0.5 index points would translate into a 0.3 percent higher average annual rate of growth of per capita GDP. Disaggregating the PMR indicator into its component parts suggests that the indicator for barriers to entrepreneurship appears to be driving most of the Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Competition, Competition Policy, and Growth 15 Table 1.3 Regulation and Growth in GDP per Capita, Aggregate PMR Cross-section analysis Dependent Panel analysis Dependent variable: variable: Average growth in GDP per Average growth in GDP per person aged person aged 16–64, Period 1998–2007 16–64, Period 1998–2003 and 2003–2007 Post-inclusion Post-inclusion Coefficient probability Coefficient probability Ln(GDP per capita) −0.016*** 99.4 −0.011* 77.1 Ln(population growth) −0.003 21.5 −0.005 21.9 Ln(investment/GDP ratio) 0.032** 95.9 0.035*** 99.4 Secondary enrollment ratio 0.000 19.1 0.000 10.6 Ethnic fragmentation −0.002 21.1 −0.001 8.8 Government consumption 0.000 26.8 0.000 8.5 Inflation 0.000 21.4 −0.001* 81.8 % of land area in tropics and subtropics −0.033** 98.4 −0.031* 94.3 Rule of law 0.000 14.1 −0.005+ 55.0 Domestic credit to private sector 0.000 26.6 0.000 46.5 Crisis dummy −0.018 48.3 PMR −0.006* 79.7 −0.003 45.5 Observations 43 86 Source: Wölfl et al. 2010. Note: GDP = gross domestic product; PMR = product market regulation. Bayesian model averaging (BMA) techniques have been applied to the Barro type growth regression. BMA accommodates both a relatively large number of controls and a small number of observations and accounts for the so-called model uncertainty associated with the process of selecting the control variables. The variables chosen in the BMA here are standard variables used in growth regressions based on country samples that cover countries of different levels of development (Wölfl et al. 2010, 20). Constant always included but not reported. +Posterior inclusion probability ≥ 50 and < 75; *posterior inclusion probability ≥ 75 and ≤ 95; **posterior inclusion ≥ 95 and ≤ 99; ***posterior inclusion ≥ 99. The posterior model probability can be viewed as a measure of the relative data fit. In summing over all models that contain a particular regressor, the posterior inclusion probability of that regressor can be obtained. This statistic provides a probability measure of how important a regressor is to explain the dependent variable. correlation between PMR and growth.16 Furthermore, barriers to entrepreneur- ship are also significant in the panel regression (see table 1.4). The estimated coefficient on the entrepreneurship indicator for the entire time period sug- gests that an improvement of 0.5 index points would translate into a higher average annual rate of GDP per capita—between 0.35 percent and 0.4 ­percent— over the subsequent decade (Wölfl et al. 2010). Disaggregating the barriers to entrepreneurship variable further—into regulatory and administrative capacity, administrative burdens on start-ups, ­ and barriers to competition—“indicates that the link between barriers to entrepreneurship and growth is due mainly to the sub-domain ‘barriers to ­ competition’—which captures legal barriers to entry and antitrust exemp- tions� (Wölfl et al. 2010, 23).17 Wölfl et al. (2010) also examine the relationship between PMR and the level of economic development. Their main line of inquiry is whether the role of PMR differs for countries with different levels of development. They cite Acemoglu, Aghion, and Zilibotti (2006) and Aghion and Howitt (2005), who argued that countries that are less developed may benefit from policies that both foster capital deepening and introduce product market rigidities, especially Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 16 Competition, Competition Policy, and Growth Table 1.4 Regulation and Growth in GDP per Capita, Regulatory Domains Cross-section analysis Dependent Panel analysis Dependent variable: variable: Average growth in GDP per Average growth in GDP per person aged person aged 16–64, Period 1998–2007 16–64, Period 1998–2003 and 2003–2007 Post-inclusion Post-inclusion Coefficient probability Coefficient probability Ln(GDP per capita) −0.016*** 99.8 −0.014* 89.7 Ln(population growth) −0.002 14.5 −0.003 14.0 Ln(investment/GDP ratio) 0.031** 96.7 0.033** 98.7 Secondary enrollment ratio 0.000 18.7 0.000 9.8 Ethnic fragmentation −0.004 34.4 −0.001 8.3 Government consumption 0.000 30.1 0.000 8.0 Inflation 0.000 15.1 −0.001* 81.1 % of land area in tropics and subtropics −0.028** 98.3 −0.031** 96.9 Rule of law 0.000 9.6 −0.003 38.1 Domestic credit to private sector 0.000 14.5 0.000 40.6 Crisis dummy −0.018+ 50.0 State control 0.000 17.6 0.000 9.2 Barriers to entrepreneurship −0.008* 94.4 −0.007* 81.5 Barriers to trade and investment 0.000 17.2 0.000 9.6 Observations 43 86 Source: Wölfl et al. 2010, 22. Note: GDP = gross domestic product; PMR = product market regulation. Bayesian model averaging (BMA) techniques have been applied to the Barro type growth regression. BMA accommodates both a relatively large number of controls and a small number of observations and accounts for the so-called model uncertainty associated with the process of selecting the control variables. The variables chosen in the BMA here are standard variables used in growth regressions based on country samples that cover countries of different levels of development (Wölfl et al. 2010, 20). Constant always included but not reported. +Posterior inclusion probability ≥ 50 and < 75; *posterior inclusion probability ≥ 75 and ≤ 95; **posterior inclusion ≥ 95 and ≤ 99; ***posterior inclusion ≥ 99. The posterior model probability can be viewed as a measure of the relative data fit. In summing over all models that contain a particular regressor, the posterior inclusion probability of that regressor can be obtained. This statistic provides a probability measure of how important a regressor is to explain the dependent variable. in relation to foreign competitors. The suggestion is that for low levels of GDP per capita, PMR may not have any effect on growth or may indeed have positive effects. Two different methods are used by Wölfl et al. (2010) to test for differ- ences in the effects of regulation on growth. The results from the second method, using a threshold approach based on Hansen (1999), are shown in table 1.5. Three separate regimes are identified based on initial GDP per capita, and the aggregate PMR measure is significantly negatively correlated with GDP per capita growth for middle- and high-income groups, especially the latter18 (see table 1.5). The significant, positive correlation for the low- income regime suggests that some regulation may be advantageous for the low-income countries. Looking at the disaggregated PMR measure, barriers to entrepreneurship account for the large negative relationship between regulation and growth for the high-income countries, while barriers to trade and investment appear important for the low-income sample. The significant negative coefficient on trade barriers for the middle-income sample of coun- tries suggests that trade barriers curb growth as development proceeds. Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Competition, Competition Policy, and Growth 17 Table 1.5 Regulation and Growth in GDP per Capita, Threshold Results Panel analysis Dependent variable: Average growth in GDP per person aged 16–64, Period 1998–2003 and 2003–2007 Coefficient with aggregate PMR Coefficient with regulatory domains Ln(GDP per capita) 0.006 −0.006 Ln(investment/GDP ratio) 0.055*** 0.036*** Inflation −0.001*** −0.001*** % of land area in tropics and subtropics −0.040*** −0.034*** Domestic credit to private sector 0.000*** Crisis dummy −0.028 PMR Low regime (lngdp per capita ≤ 9.6) 0.009*** Middle regime (10 ≥ lngdp per capita > 9.6) −0.006** High regime (lngdp per capita > 10) −0.012*** State control Low regime (lngdp per capita ≤ 9.6) 0.004 Middle regime (10.25 ≥ lngdp per capita > 9.6) 0.001 High regime (lngdp per capita ≤ 10.25) −0.002 Barriers to entrepreneurship Low regime (lngdp per capita ≤ 9.6) −0.016 Middle regime (10.25 ≥ lngdp per capita > 9.6) 0.007 High regime (lngdp per capita > 10.25) −0.008** Barriers to trade and investment Low regime (lngdp per capita ≤ 9.6) 0.014** Middle regime (10.25 ≥ lngdp per capita > 9.6) −0.013*** High regime (lngdp per capita > 10.25) 0.002 Memorandum: LR stat. p-value LR stat. p-value H0: linear vs H1: 2. regime model 16.342 0 7.661 0 H0: 1 regime vs H1: 3. regime model 3.532 0 9.492 0 R2 adj. 0.65 0.61 Observations 86 86 Source: Wölfl et al. 2010. Note: GDP = gross domestic product; PMR = product market regulation. Constant always included but not reported. Standard errors are in parenthesis. P-values are bootstrapped. For details, see Hansen (1999). Significance level: * = 10 percent, ** = 5 percent, *** = 1 percent. State control does not appear to have any effect on growth, independent of development. These results agree with the findings from the first method (Wölfl et al. 2010). Summary There are a number of data sets that provide information on PMR (see note 8). The conclusions that emerge from a review of the data sets are these: • Regulatory burdens vary widely across the world, with regulation in poorer countries more stringent than in richer countries, and greater in countries with a French legal origin or with a socialist legal origin than in others. Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 18 Competition, Competition Policy, and Growth • The dispersion of regulatory regimes is greater in developing countries relative to developed countries. • There has been a generalized tendency toward the relaxation of regulation concerning entry, accompanied by a decrease in tariff and nontariff barriers to trade in manufacturing. • Many developing countries, including India and China, have been moving toward less restrictive regulation. • OECD countries have experienced substantial deregulation in services and in sectors such as telecommunications, utilities, and transport. • Regulatory reform began first in the United States in the early 1980s. The United Kingdom, Canada, New Zealand, and the Nordic European countries started to reform in the mid-1980s, whereas Australia and other European countries began market reform in the mid-1990s. • Regulatory reform has often been accompanied by privatization, so that there has been a tendency for the share of output produced by public enterprises to decrease (Schiantarelli 2008). Competition Policy Effective competition does not happen automatically. Competition may be harmed by vested interests, inappropriate government policies, and anticom- petitive behavior of incumbent firms. Ellis and Singh (2010a, 4) write that “appropriate policies are crucial to create the conditions within which compe- tition can thrive, and competition authorities can help to build a culture of competition, and increase awareness of competition issues amongst policy makers and the public.� Competition policy extends to all policies that gener- ate an environment in which competition can flourish. Other government policies with a bearing on competition are trade policy, regulation, privatiza- tion, industrial policy, and competition law. Implementing and maintaining an effective competition framework is critical for attracting investment and developing the private sector (Ellis 2008; Godfrey 2008; Broadman 2007). Competition policies must be cognizant of a country’s developmental stand- ing and its governance capacities. This approach, while promoting the concept of competition, prevents the adoption of a o ­ ne-size-fits-all model and the pursuit of “maximum competition� that may be harmful to socioeconomic development. Competition policy is a prime part of an economy’s develop- ment strategy. It is of particular importance to developing economies in this era of globalization and liberalization. Research on the state of competition in most developing economies is ­ stymied by a lack of sufficient data and difficulty in attributing outcomes to competition policy per se (especially where competition reform was part of a package of economic reform). Nevertheless, globalization and the liberalization of developing country economies provide compelling arguments for competi- tion policies. The enormous structural and regulatory changes that have Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Competition, Competition Policy, and Growth 19 occurred in developing economies over the last few decades as a result of privatization and deregulation require an appropriate competition policy to ­ ensure improved economic performance. For example, replacing a privatized public natural monopoly with a privatized private sector monopoly is not good for social welfare. Singh (2002) suggests that ownership is not the issue, but rather the extent of competition, which depends on the external environment. The unprecedented activity in mergers and acquisitions that took place in the 1990s has reshaped the world economy and provides another “important reason for developing countries to have competition laws� (Singh 2002, 9). Singh (2002) outlines the concerns for developing countries. One direct effect is the increased market power of large multinationals and their potential abuse of dominance, such as by acquiring domestic firms. The developing country may be affected indirectly as a result of the reduced contestability of the market owing to the few large players. Competition law would help restrain cartels and other uncompetitive conduct by large multinationals. However, the success of competition law relies upon an adequately developed institutional and legal framework. Developing economies present many challenges and opportunities to the implementation of effective competition policy. For many firms, in particular large firms, competitive success is tied to the government. The existence of an economic elite with close ties to the government prevents the development of competition policy (Ellis 2008). Breaking down the barriers that protect this elite is the focus of competition policy. Competition authorities have a role to play in mobilizing interest groups to lobby for reform. Interest groups are those that stand to gain from reforms, such as households, industrial consumers, and poten- tial new entrants. On the other hand, competition policy can facilitate a level playing field for small and medium enterprises. Many of the poor in developing economies are small entrepreneurs, including farmers. They stand to benefit from an improved competitive environment in which entry and exit barriers are low, inputs are priced fairly, and opportunities exist for selling output at fair and com- petitive prices (Godfrey 2008). Godfrey (2008) refers to a database on media allegations of anticompetitive behavior in Sub-Saharan Africa for the 10 years ending December 2004 that revealed a number of competition concerns in the region. Concerns arose from anticompetitive practices in the sugar and flour industries, in the prices of manu- facturing inputs, and in the output markets for cotton, tea, coffee, and tobacco. Aghion, Braun, and Fedderke (2006) found that markups are significantly higher in manufacturing industries in South Africa compared to manufacturing indus- tries worldwide. Broadman (2007) compared the administrative barriers to start- ing a business across a number of developing regions and found that these were significantly more onerous in Sub-Saharan Africa (table 1.6). More recently, Ellis and Singh (2010b) examined the extent of competition for four markets in five developing economies from Africa and Asia. Table 1.7 shows the details. Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 20 Competition, Competition Policy, and Growth Table 1.6 Administrative Barriers to Starting a Business, by Region Starting a business Sub-Saharan Africa East Asia South Asia Number of procedures 11.0 8.2 7.9 Time in days 63.8 52.6 35.3 Cost (% per capita income) 215.3 42.9 40.5 Minimum capital (% per capita income) 297.2 109.2 0.8 Source: Broadman 2007, cited in Godfrey 2008. Ellis and Singh (2010b) conclude the following from their case studies: 1. Markets with more competition, more players, more dynamic entry and exit, and more intense rivalry for customers tend to deliver better market outcomes. 2. Competition is often constrained—some industries by their nature are highly concentrated (cement and beer, for example)—but competition authorities have an important role to play in monitoring, publicizing, and tackling anti- competitive behavior. 3. The state has a large role in determining competition and market outcomes, through regulation, state ownership and privatization, price controls or subsi- dization, import protection, industrial policy, corrupt business deals, and ownership by individual politicians or their families. ­ 4. In some countries and markets, there is a close relationship between govern- ment and business that creates a powerful economic elite with vested interests in opposing procompetition, progrowth reforms. Singh (2002) examines the evidence for competition and competition policy in emerging markets after the Asian financial crisis. He finds, contrary to opinion, that competition in the more advanced emerging markets is just as intense as in advanced economies. An example of this finding is shown in table 1.8, which compares the persistence of profitability as measured by the time series estimates of persistence coefficients. The coefficients are lower for developing economies compared to advanced economies, suggesting that the developing economies are “subject to no less, if not greater competition, than advanced countries� (Singh 2002, 4). Moreover, Singh suggests that competition among small enterprises in the emerging markets is more intense than in advanced economies. Tybout (2000) reviewed the empirical evidence on competition in emerging markets and suggested that the evidence did not support the view that manufacturing plants and jobs had lower turnover rates in emerging markets than in OECD countries. Competition policy in developing economies was formulated and imple- mented over the last three decades—just 16 developing economies had a formal competition policy before 1990—with 50 countries completing leg- ­ islation for competition laws in the 1990s and a further 27 in the early 2000s (Singh 2002). The relative absence of competition policy was easily Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Table 1.7 Structure, Conduct, and Performance of Selected Industries in Five Developing Economies Sugar Cement Beer Mobile telephony Ownership Notes Ownership Notes Ownership Notes Ownership Notes Bangladesh State Low productivity; Many different Greater evidence of Highly No further information Relatively Lowest tariffs across poor market price and nonprice concentrated competitive five countries; performance; players competition regulated obsolete price floor is technology; a concern— inefficient putting smaller farming operators and methods new entrants at Struggling to a disadvantage Kenya State compete Very Competition authority Highly Many anticompetitive Relatively Entry of two new with private concentrated is investigating issue concentrated practices identified— concentrated players has producers of joint ownership territorial allocation, until recently coincided with domestically and among three price fixing, exclusive a fall in tariffs by internationally suppliers dealership 50 percent Need substantial Vietnam State Many different Greater evidence of Least Prices are lowest; Mainly state Operators appear to levels of costly suppliers price and nonprice concentrated nonprice competition compete fiercely government competition of five seems strongest, but subsidization countries; there are allegations seven of exclusive dealing, producers abuse of dominance table continues next page 21 22 Table 1.7  Structure, Conduct, and Performance of Selected Industries in Five Developing Economies (continued) Sugar Cement Beer Mobile telephony Ownership Notes Ownership Notes Ownership Notes Ownership Notes Zambia Private Produces highest Very Competition authority Monopoly Highest prices of State Lowest mobile amounts of sugar concentrated investigating supply producer five countries; penetration rate; per hectare of constraints competition authority highest prices; five countries Price of cement has has imposed government- Profitable, fallen by almost conditions controlled internationally 10 percent since Evidence of barriers to gateway that competitive 2008, coinciding with entry, resale price charges high industry, new market player maintenance, and prices to private domestic sugar in 2009 exclusive dealership operators prices are high Monopolistic market structure Ghana Two State-led sugar Two suppliers Allegations of price Two firms Allegations of price Two operators Intense potential industry hikes not investigated leadership; prices are competition; entrants collapsed; due to absence of low, suggesting fierce good mobile country now competition authority competition penetration; imports all sugar. relatively low Two entrants prices; effective looking for regulator government guarantees against imports Source: Ellis and Singh 2010b. Competition, Competition Policy, and Growth 23 Table 1.8 Persistence of Profitability in Emerging Markets and Advanced Markets Emerging market Advanced market country Score Source country Score Source Brazil 0.013 Glen, Lee, and Canada 0.425 Khemani and Shapiro (1990) India 0.229 Singh (2001) France 0.142 Korea, Republic of 0.323 Germany 0.410 Malaysia 0.349 Germany 0.485 Schwalbach, Grasshof, and Mahmood (1989) Mexico 0.222 Germany 0.509 Schohl (1990) Zimbabwe 0.421 Japan 0.465 Odagiri and Yamawaki (1990) United Kingdom 0.482 Cubbin and Geroski (1990) United Kingdom 0.488 Geroski and Jacquemin (1988) United States 0.183 Mueller (1990) United States 0.540 Waring (1996) Source: Singh 2002. understood in the developing economy context, in which there was consid- erable state control over industry. The government intervened directly when it perceived any anticompetitive behavior and fixed prices for other essential products (Singh 2002). A World Bank (2002) survey of competition laws identified intercountry differences along three key dimensions of competi- tion law: the definition of dominance, the treatment of cartels, and enforcement. The Competition Assessment Framework (CAF) was designed by the U.K. Department for International Development (DFID 2008) to identify and assess the nature of anticompetitive practices.19 It was designed for policy makers in developing economies to “provide guidance on how a sector-by-sector approach to the state of competition can be undertaken� (Godfrey 2008, 8). It is appli- cable to all country situations, from a functioning competition authority, sector regulators, and competition law to the absence of all of these. It can help in formulating policy advice on the effects of anticompetitive practices in key markets, can inform the design of programs or projects to catalyze private sector development, and can serve as part of a holistic “growth diagnostic� ­ (Godfrey 2008). A summary of the CAF as described by Godfrey (2008) is shown in table 1.9. Singh (2002) suggests that a competition policy for a developing economy must be able to (1) restrain anticompetitive behavior by domestic privatized firms; (2) limit abuses of monopoly power by megacorporations created by the international merger movement; and (3) promote development. Singh (2002) is quite pessimistic about the ability of some developing countries to implement competition policy, in particular because competition policy requires a strong state that many developing countries lack. For developing economies with strong governments, even if not always democratic (for example, China, India, Brazil, and Mexico), he suggests a broad-based competition policy that would “in some Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 24 Competition, Competition Policy, and Growth Table 1.9 Competition Assessment Framework Themes Questions Notes Selecting sectors Applying objective measures to select a sector Sector should be important to the economy; and markets for for competition assessment, including its role should be suggestive of competition assessment in the economy, importance to consumers, problems concern about prices, past performance, entry barriers, and market concentration Identifying the relevant Questions to identify the relevant market or Need to identify existing suppliers and markets and the markets in the sector buyers and their importance in the competitors market Examining the market How to assess the level of concentration in the High concentration is often, although not structure market exclusively a significant factor in market behavior Looking for barriers to Whether there are any significant barriers to entry Questions examine natural barriers, strategic entry barriers, regulatory and policy barriers, and gender barriers Ascertaining if Reviews the legislation, policies, and institutions This may include licensing restrictions, FDI government policies of governments at all levels (national, state, restrictions, and trade barriers or institutions limit local); questions seek to ascertain if state- competition owned enterprises receive any preferences that might restrict competition by the private sector Considering vested Questions seek to identify the objectives, power, Sensitive area; vested interest may be interests and influence of stakeholders opposed to personal, corporate, or institutional competition in a market Looking for signs of Questions are aimed at identifying practices Dominance is possible where a firm has anticompetitive of firms that can impede competition; for strong market power arising from high conduct by firms example, abuse of dominance, collusion market share and barriers to entry; cartels among competitors, and impact of M&A may be solely domestic firms or a mix of domestic and international; M&A may benefit or harm competition Drawing conclusions Review conclusions to each of the preceding and Annexes to the CAF provide additional form a view on overall state of competition in information on the definition of markets each sector and calculation of market concentration Source: Godfrey 2008. Note: CAF = Competition Assessment Framework; FDI = foreign direct investment; M&A = mergers and acquisitions. instances involve restriction of competition and in others its vigorous promotion� (2002, 16). Optimal competition rather than maximum competition should be the goal, with a blend of competition and cooperation designed to achieve sus- tained economic growth. As noted earlier, competition policy should not assume that one size fits all, but rather depend on the development stage of the underly- ing economy, as well as the effectiveness of its government and its institutional framework. In addition, he suggests that the private sector’s propensity to invest be maintained at high levels and notes that the need for a steady growth of prof- its may necessitate governmental involvement in investment decisions (that is, to prevent overcapacity and falling profits). He highlights the crucial importance of industrial policy in achieving the structural changes needed for economic devel- opment; this role for industrial policy necessitates its coherence with competi- tion policies (Singh 2008). Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Competition, Competition Policy, and Growth 25 Conclusion The chapter addressed two concepts—competition and competition policy. The degree and nature of competition in an economy have implications for growth. The chapter examined competition in product markets. Regulation in product markets suggests how supportive an economy is of competition. In principle, more stringent regulation is negatively associated with ­ economic growth, while improved governance lessens the negative effects of regulation. The first part of the chapter looked at PMR and economic performance, identifying the ambiguities and caveats that arise at the microeconomic, firm, and macroeconomic levels. In general, less restrictive PMR is conducive to growth. However, this relationship is dependent upon the level of develop- ment in the economy, and at early stages of development, greater PMR through restrictions on foreign trade and investment may be positive for growth. Regulation tends to be more stringent in developing economies, although many, including China and India, are moving toward less restrictive regulation. The chapter then addressed the issue of competition policy, noting that effec- tive competition does not happen automatically and highlighting the role of competition policy and other policies in establishing the appropriate environ- ment for competition to thrive. Trade policy, regulation, privatization, industrial policy, and competition law all have a bearing on competition. Furthermore, competition policy needs to take into account an economy’s developmental stage and its governance capabilities. The chapter cited a number of reasons why competition policy is particu- larly important for developing economies. One has to do with the current era of globalization and liberalization. Furthermore, the success of the struc- tural and regulatory changes that took place in developing economies following mass privatization and deregulation requires appropriate competi- ­ tion policies. The large-scale activity in mergers and acquisitions during the 1990s reshaped the world economy with implications for developing economies. The chapter also presented some evidence for the extent of competition and competition policy in developing and emerging market economies. We ­ concluded with a synopsis of the CAF that is supporting competition policy reform in India. It was designed to identify and assess the nature of anticompetitive practices and is helpful in formulating policy advice. Notes 1. See Ahn (2002); Nickell (1996); Disney, Haskel, and Heden (2000); and Klette (1999), referenced in Yun (2004). 2. Increasing managerial effort may be in response to the risk of losing market share or to greater opportunities for comparing performance across firms; see Nickell (1996). Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 26 Competition, Competition Policy, and Growth 3. See Ahn (2002). Furthermore, Nicoletti et al. (2001), Bassanini and Ernst (2002), and Jaumotte and Pain (2005) suggest that “too strict and too high nontrade barriers are associated with low research and development (R&D) intensity in the business sector� (Høj et al. 2007, 6). 4. Alesina et al. (2005) find that procompetitive reforms increase capital deepening in nonmanufacturing industries. Nicoletti and Scarpetta (2003) find the same reforms considered by Alesina et al. (2005) improve multifactor productivity and facilitate a faster catch-up to the technological leader. Conway et al. (2006) find similar effects of competition on investment in information and communications technology and labor productivity growth (Høj et al. 2007). 5. The two dimensions and four groups of manufacturing industries classification derive from Oliveira Martins, Price, and Mulder (2002). The authors mapped manufacturing sectors into this twofold classification using an estimate of the minimum efficient scale to determine which markets/industries were fragmented or segmented and R&D intensity to classify industries according to the level of endogenous sunk costs. 6. See Blundell, Griffith, and Van Reenen (1999) and Aghion et al. (2005). 7. The paper by Nickell (1996) is seminal in this area. Its findings were confirmed by Disney, Haskel, and Heden (2000). 8. This is found in Nickell (1996) for U.K. firm data, in Klette (1999) for Norwegian plant data, and in Bottasso and Sembenelli (2001) for Italian firm data. 9. The product market regulation (PMR) is just one data series. Other data series that facilitate an examination of PMR are the Eurostat data on sectoral and ad hoc state aid, public procurement, and openly advertised public procurement; data from the European Centre for Public Enterprises with Public Participation (CEEP); the World Bank Doing Business Indicators database, available at http://www.doingbusiness.org; the World Bank Investment Climate Assessment survey; and data on the effect of regulation collected by the Fraser Institute (Schiantarelli 2008). 10. The Foreign Direct Investment (FDI) restrictiveness index, originally developed in 2003, measures the restrictiveness of a given country’s policy toward FDI on a scale of 0 (no restrictions) to 1 (no FDI). Four types of measures are covered: (1) foreign equity restrictions; (2) screening and prior approval requirements; (3) rules for key personnel; and (4) other restrictions on the operation of foreign enterprises. The index covers 22 sectors, the scores for which are averaged to obtain a country score: the FDI index for the country concerned. The index is available for all Organisation for Economic Co-operation and Development (OECD) members, adherents to the Declaration on International Investment and Multinational Enterprises, enhanced engagement countries, and other G-20 countries (Kalinova, Palerm, and Thomsen 2010). ­ 11. Empirical studies suggest that the correlation between procompetitive policies and growth is driven largely by factors that promote entrepreneurship and competition. Thus, the “barriers to entrepreneurship� variable is the one considered. 12. See also Acemoglu, Aghion, and Zilibotti (2006) and Aghion and Howitt (2005). 13. The accession countries in the study are Chile, Estonia, Israel, Russia, and Slovenia. 14. Enhanced engagement countries are Brazil, China, India, Indonesia, and South Africa. 15. Nonmember OECD countries in this study are Bulgaria, Croatia, Romania, and Ukraine. 16. The insignificance of the state control variable agrees with results from the empirical literature that suggest that privatization can bear fruit only if it is combined with Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Competition, Competition Policy, and Growth 27 liberalization. An insignificant “barriers to trade and investment� variable may reflect underlying country differences caused by countries’ different stages of economic development (Wölfl et al. 2010). 17. Wölfl et al. (2010) urge caution when “interpreting these results as the sub-domains are highly correlated in [this] small sample.� 18. Countries in the low regime include Brazil, Bulgaria, China, India, Indonesia, Romania, Russia, South Africa, Turkey, and Ukraine. Countries in the middle regime include Chile, Croatia, Estonia, Hungary, Republic of Korea, Mexico, Poland, and, the Slovak Republic. Countries in the high regime are Australia, Austria, Belgium, Canada, the Czech Republic, Denmark, Finland, France, Germany, Greece, Iceland, Ireland, Italy, Japan, Luxembourg, New Zealand, Netherlands, Norway, Portugal, Slovenia, Spain, Sweden, Switzerland, the United Kingdom, and the United States. 19. “The CAF was designed in response to demand from the current DFID/FIAS (UK Department of International Development/Foreign Investment Advisory Service at the World Bank) program that is supporting competition policy reform in India� (Godfrey 2008, 8). References Acemoglu, D., P. Aghion, and F. Zilibotti. 2006. “Distance to Frontier, Selection, and Economic Growth.� Journal of the European Economic Association 4 (1): 37–74. Aghion, P., N. Bloom, R. Blundell, R. Griffith, and P. Howitt. 2005. “Competition and Innovation: An Inverted-U Relationship.� Quarterly Journal of Economics 20 (2): 701–28. Aghion, P., M. Braun, and J. Fedderke. 2006. “Competition and Productivity Growth in South Africa.� Working Paper No. 132, Center for International Development at Harvard University, Cambridge, MA. Aghion, P., and P. Howitt. 2005. “Appropriate Growth Policy: A Unifying Framework.� Joseph Schumpeter Lecture. Journal of the European Economic Association 4 (4): 269–314. Ahn, S. 2002. “Competition, Innovation and Productivity Growth: A Review of Theory and Evidence.� Economics Department Working Paper 317, Organisation for Economic Co-operation and Development, Paris, France. Alesina, A., S. Ardagna, G. Nicoletti, and F. Schiantarelli. 2005. “Regulation and Investment.� Journal of the European Economic Association 3 (4): 791–825. Bartelsman, E., J. Haltiwanger, and S. Scarpetta. 2004. “Microeconomic Evidence of Creative Destruction in Industrial and Developing Countries.� World Development Report background Paper, World Bank, Washington, DC. Bassanini, A., and E. Ernst. 2002. “Labour Market Institutions, Product Market Regulation, and Innovation: Cross-country Evidence.� Economics Department Working Paper 316, Organisation for Economic Co-operation and Development, Paris, France. Blundell, R., R. Griffith, and J. Van Reenen. 1999. “Market Share, Market Value and Innovation in a Panel of British Manufacturing Firms.� Review of Economic Studies 66: 529–54. Bottasso, A., and A. Sembenelli. 2001. “Market Power, Productivity, and the EU Single Market Program: Evidence from a Panel of Italian Firms.� European Economic Review 45 (1): 167–86. Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 28 Competition, Competition Policy, and Growth Brandt, N. 2004. “Business Dynamics, Regulation and Performance.� OECD Directorate for Science Technology and Industry Working Paper 2004/3, Organisation for Economic Co-operation and Development, Paris, France. Broadman, H. 2007. Africa’s Silk Road: China and India’s New Economic Frontier. Washington, DC: World Bank. Cincera, M., and O. Galgau. 2005. “Impact of Market Entry and Exit on EU Productivity and Growth Performance.� European Economy Economic Paper 222, Directorate General Economic and Monetary Affairs, European Commission, Brussels, Belgium. Conway, P., D. de Rosa, G. Nicoletti, and F. Steiner. 2006. “Regulation, Competition and Productivity Convergence.� Economics Department Working Paper 509, Organisation for Economic Co-operation and Development, Paris, France. Conway, P., R. Herd, T. Chalaux, P. He, and J. Yu. 2010. “Product Market Regulation and Competition in China.� Economics Department Working Paper 823, Organisation for Economic Co-operation and Development, Paris, France. Cubbin, J. S., and P. A. Geroski. 1990. “The Persistence of Profits in the United Kingdom.� In The Dynamics of Company Profits: An International Comparison, edited by D. C. Mueller, 147–68. Cambridge, UK: Cambridge University Press. DFID (Department for International Development). 2008. Competition Assessment Framework: An Operational Guide for Identifying Barriers to Competition in Developing Countries. London: DFID. Disney, R., J. Haskel, and Y. Heden. 2000. “Restructuring and Productivity Growth in UK Manufacturing.� Discussion Paper 2463, Centre for Economic Policy Research, London. Ellis, K. 2008. “How to Achieve Growth: The Million-Dollar Question.� ODI Opinion, Overseas Development Institute, London. http://www.odi.org.uk/sites/odi.org.uk/ files/odi-assets/publications-opinion-files/820.pdf. Ellis, K., and R. Singh. 2010a. “The Economic Impact of Competition.� Project Briefing No. 42, Overseas Development Institute, London, UK. ———. 2010b. Assessing the Economic Impact of Competition. London: Overseas Development Institute. http://dspace.cigilibrary.org/jspui/bitstream/123456789/​ 29339/1/Assessing%20the%20Economic%20Impact%20of%20Competition.pdf?1. Fiori, G., G. Nicoletti, S. Scarpetta, and F. Schiantarelli. 2007. “Employment Outcomes and the Interaction between Product and Labor Market Deregulation: Are They Substitutes or Complements?� Department of Economics Working Paper 663, Boston College, Boston, MA. Foster, L., J. Haltiwanger, and C. J. Krizan, 2001. “Aggregate Productivity Growth: Lessons from Microeconomic Evidence.� In New Developments in Productivity Analysis, edited by C. Hulten, E. Dean, and M. Harper, 303–72. Chicago: University of Chicago Press. Fraser Institute. 2002. Economic Freedom of the World: 2002 Annual Report. Vancouver: Fraser Institute. Glen, J., K. Lee, and A. Singh. 2001. “Persistence of profitability and competition in emerg- ing markets.� Economics Letters 72: 247–53. Geroski, P. A., and A. Jacquemin. 1988. “The Persistence of Profits: A European Comparison.� Economic Journal 98: 375–89. Godfrey, N. 2008. “Why Is Competition Important for Growth and Poverty Reduction?� Paper prepared for the Global Forum on International Investment VII, Paris, March 27–28. http://www.oecd.org/investment/globalforum/40315399.pdf. Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Competition, Competition Policy, and Growth 29 Griffith, R., and R. Harrison. 2004. “The Link between Product Market Reform and Macro-economic Performance.� European Economy Economic Paper 209, Directorate General Economic and Monetary Affairs, European Commission, Brussels, Belgium. Hansen, B. 1999. “Threshold Effects in Non-dynamic Panels: Estimation, Testing, Interference.� Journal of Econometrics 93 (2): 345–68. Høj, J., M. Jimenez, M. Maher, G. Nicoletti, and M. Wise. 2007. “Product Market Competition in the OECD Countries: Taking Stock and Moving Forward.� Economics Department Working Paper 575, Organisation for Economic Co-operation and Development, Paris, France. Jaumotte, F., and N. Pain. 2005. “Innovation in the Business Sector.� Economics Department Working Paper 459, Organisation for Economic Co-operation and Development, Paris, France. Kalinova, B., A. Palerm, and S. Thomsen. 2010. “OECD’s FDI Restrictiveness Index: 2010 Update.� Working Paper on International Investment, No. 2010/3, Organisation for Economic Co-operation and Development Investment Division, Paris, France. Khemani, R. S., and D. M. Shapiro. 1990. “The Persistence of Profitability in Canada.� In The Dynamics of Company Profits: An International Comparison, edited by D. C. Mueller, 77–104. Cambridge, UK: Cambridge University Press. Klapper, L., L. Laeven, and R. G. Rajan. 2004. “Business Environment and Firm Entry: Evidence from International Data.� Working Paper 10380, National Bureau of Economic Research, Cambridge, MA. Klette, T. J. 1999. “Market Power, Scale Economies and Productivity: Estimates from a Panel of Establishment Data.� Journal of Industrial Economics 47 (4): 451–76. Loayza, N. V., A. M. Oviedo, and L. Serven. 2004. “Regulation and Macroeconomic Performance.� Working Paper, World Bank, Washington, DC. ———. 2005. “Regulation and Microeconomic Dynamics: A Comparative Assessment for Latin America.� Working Paper, World Bank, Washington, DC. Mueller, D. C., ed. 1990. The Dynamics of Company Profits: An International Comparison. Cambridge, UK: Cambridge University Press. Nickell, S. 1996. “Competition and Corporate Performance.� Journal of Political Economy 104: 724–46. Nicoletti, G., A. Bassanini, E. Ernst, S. Jean, P. Santiago, and P. Swaim. 2001. “Product and Labor Market Interactions in OECD Countries.� Economics Department Working Paper 226, Organisation for Economic Co-operation and Development, Paris, France. Nicoletti, G., and S. Scarpetta. 2003. “Regulation, Productivity and Growth: OECD Evidence.� Economic Policy 18 (36): 11–72. ———. 2005. “Product Market Reforms and Employment in OECD Countries.� Economics Department Working Paper 472, Organisation for Economic Co-operation and Development, Paris, France. Odagiri, H., and H. Yamawaki. 1990. “The Persistence of Profits in Japan.� In The Dynamics of Company Profits: An International Comparison, edited by D. C. Mueller, 169–86. Cambridge, UK: Cambridge University Press. OECD (Organisation for Economic Co-operation and Development). 2002. “Product Market Competition and Economic Performance.� OECD Economic Outlook 2: 155–62. Oliveira Martins, J., T. Price, and N. Mulder. 2002. “A Taxonomy of Market Cluster.� Mimeo, Organisation for Economic Co-operation and Development, Paris. Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 30 Competition, Competition Policy, and Growth Scarpetta, S., P. Hemmings, T. Tressel, and J. Woo. 2002. “The Role of Policy and Institutions for Productivity and Firm Dynamics: Evidence from Micro and Industry Data.� Economics Department Working Paper 329, Organisation for Economic Co-operation and Development, Paris, France. Schiantarelli, F. 2008. “Product Market Regulation and Macroeconomic Performance: A Review of Cross-country Evidence.� Working Paper, Boston College, Boston. http:// fmwww.bc.edu/ec-p/wp623.pdf. Schohl, F. 1990. “Persistence of Profits in the Long Run: A Critical Extension of Some Recent Findings.� International Review of Industrial Organization 8: 385–403. Schumpeter, J. A. 1942. Capitalism, Socialism and Democracy. New York: Harper and Brothers. Schwalbach, J., U. Grasshof, and T. Mahmood. 1989. “The Dynamics of Corporate Profits.� European Economic Review 3: 1625–39. Singh. A. 2002. “Competition and Competition Policy in Emerging Markets: International and Developmental Dimensions.� Research Papers for the Intergovernmental Group of Twenty-Four on International Monetary Affairs, G-24 Discussion Paper 18, United Nations, New York.    . 2008. “The Past, Present, and Future of Industrial Policy in India: Adapting to the Changing Domestic and International Environment.� Working Paper No. 376, ESCRC Centre for Business Research, University of Cambridge, Cambridge, U.K. Tybout, J. 2000. “Manufacturing Firms in Developing Countries: How Well Do They Do, and Why?� Journal of Economic Literature 38: 11–44. Waring, G. 1996. “Industry Differences in the Persistence of Firm-specific Returns.� American Economic Review 86: 1253–265. Wölfl, A., I. Wanner, O. Röhn, and G. Nicoletti. 2010. “Product Market Regulation: Extending the Analysis beyond OECD Countries.� Economics Department Working Paper 799, Organisation for Economic Co-operation and Development, Paris, France. World Bank. 2002. World Development Report: Building Institutions for Markets. Washington, DC: World Bank and Oxford University Press. Yun, M. 2004. “Competition and Productivity Growth: Evidence from Korean Manufacturing Firms.� In Competition, Competitiveness and Development: Lessons from Developing Countries, UNCTAD/DITC/CLP/2004/1, United Nations Conference on Trade and Development, Geneva. http://unctad.org/en/Docs/ditcclp20041ch4_​ en.pdf. Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Chapter 2 Competitiveness and Its Indicators Competitiveness is a broad concept applied at many different levels and ­ measured by many different indicators. It is, according to Siggel (2007, 5), an “ambiguous concept,� because of a failure to rigorously define competitiveness in the early economic literature. “Competitiveness� is used interchangeably with “comparative advantage,� “favorable business environment,� and “productivity,� for example. Furthermore, underlying structural factors that influence productiv- ity may not show up directly in measures of competitiveness but may account for improved terms of trade. The interest in competitiveness has increased in recent decades at the national and firm levels, as economists and academics producing the business literature once more seek to understand why some countries grow faster than others and why some firms and regions fare better than others.1 An abundance of indicators is used to measure competitiveness at the national, regional, industry, and firm level. Indicators measure the success of countries in facilitating an economic environment that enables firms’ domestic and global competitiveness. Categorizing these indicators into discrete units is difficult, given the difficulty in sometimes differentiating cause from effect; but some broad classifications are possible, for example, narrow versus broad measures, macro versus micro, short term versus long term, price versus nonprice. The following sections identify the various elements of competitiveness and address price and nonprice measures of competitiveness. Elements of Competitiveness Economists from the classical school, most notably Adam Smith, equated ­ competitiveness with the market mechanism arising from the production and distribution of goods and services based on price and quality. Thus, by this view output and wealth are created at the micro level. The nature and productivity of the economic activities taking place is paramount. Purely local industries count for competitiveness, because their productivity not only sets their wages but also has a major influence on the cost of doing business and the cost of living in the country. Competitiveness among enterprises relies on (1) efficiency, that is, being Clusters of Competitiveness   •  http://dx.doi.org/10.1596/978-1-4648-0049-8   31   32 Competitiveness and Its Indicators productive; (2) choice,2 since in order to be competitive, an enterprise needs to choose those domains in which its productivity provides greater value added than that of its competitors3; and (3) resources, since an enterprise makes choices based upon the resources it can mobilize; included here are government, ­ infrastructure, technology, finance, and education (Garelli 2011). The quality of the business environment is critical for a firm’s productivity. Porter’s diamond theory (see for example Porter et al. 2008) explains how the business environment affects competitiveness. The diamond model (figure 2.1) incorporates (1) factor (input) conditions (natural endowments, human resources, capital availability, physical infrastructure, administrative infrastructure, ­information infrastructure, scientific and technological infrastructure); (2) ­context for firm strategy and rivalry (local rules and incentives that encourage investment and productivity, vigorous local competition); (3) related and supporting industries (capable, locally based suppliers and supporting industries, presence of ­ clusters instead of isolated firms); and (4) demand industries (demanding and sophisticated local customers and needs). Figure 2.1  Business Environment Quality: The Diamond Context for firm strategy and rivalry Local rules and incentives that encourage investment and productivity For example, incentives for capital Factor (input) investments, intellectual conditions property protection Vigorous local competition Demand High-quality, e cient, and specialized Industries inputs to business Openness to foreign and Natural endowments local competition Demanding and sophisticated Human resources local customers and needs Capital availability Challenging quality, saftey, Physical infrastructure and environmental standards Administrative infrastructure (for example, registration, permitting) Information infrastructure (for example, economic data, Related and supporting corporate disclosure) Industries Scienti c and technological infrastructure Capable, locally based suppliers and supporting industries Presence of clusters instead of isolated rms Source: Porter et al. 2008. © World Economic Forum. Used with permission; further permission required for reuse. Clusters of Competitiveness   •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Competitiveness and Its Indicators 33 The success of nations is linked to prosperity that derives from economic growth plus “something else� (Garelli 2011, 489). The something else depends upon the level of development of the underlying economy—perhaps access to food and shelter in poorer countries or environmental protection in more ­ developed economies. Unlike enterprises, nations do not generate economic value added by themselves but play a more indirect role by creating an environment that supports the activities of enterprises, including innovation. ­ Advances in communications and globalization work increase the interdepen- dence between nations and enterprises (Garelli 2011). Regions and metropolises play a key role in the study of competitiveness. Globalization has made regions, port cities, and inner cities hubs of economic activity, particularly in manufacturing, commerce, and service industries. Business support services, hi-tech and biotech parks, industrial clusters, and information and communication technology (ICT) centers are all springing up to take full advantage of the agglomeration effects of concentrated economic activity and globalization. Mobile factors of production, in addition to promoting the ­ ever-increasing movement of goods and services across geographical boundaries, are leading to rapid urbanization and bringing in their train the need for green spaces, smart spaces (research and development [R&D] firms and universities and ­colleges for high-skilled workers), support services, essential infrastructure, and affordable housing for the masses. Linking competitiveness to spatial ­ development has implications for urban governance in that it improves location advantages and ensures availability of high-skilled workers, managers, and entrepreneurs. At the global level, a country is said to be competitive if it is able to hold or increase its share of products (exports) in the world economy. Undervaluing or devaluing a nation’s currency relative to other currencies to gain a competitive advantage, or using industrial policy to increase exports (through subsidies, tariffs on substitutable imports, lower wages in export industries, and/or aid-for-trade), can however bring problems of its own. The need to keep wages low or follow a two-track wage structure (one for cheap exports and a higher one for domestic consumption) reveals a lack of true competitiveness and holds down an economy’s average standard of living. Similarly, government subsidies for ­ preferred industries and sectors burden public finances, drain national income, ­ and bias choices away from the most productive use of the nation’s resources. Undervaluation and devaluation of exchange rates imply a collective national pay cut by discounting the products and services sold in world markets, while raising the cost of the goods and services imported from abroad. Therefore, in a dynamic world, the best policy is to have in place measures that continually increase productivity both at the micro level (farms and firms) and at the macro or ­ national level. Many exporters in developing economies face persistent barriers to competing in global markets.4 Guilherme Reis and Farole (2012, 3) identify the barriers that arise from distortionary macroeconomic policies, poor factory conditions, and ­ ineffective public policies that “prevent the exploitation of intra- and Clusters of Competitiveness   •  http://dx.doi.org/10.1596/978-1-4648-0049-8 34 Competitiveness and Its Indicators Figure 2.2 The Three Pillars of Trade Competitiveness Removing economic biases arising from tari and nontari barriers, real exchange Aligning macro rate misalignment, and distortive tax regimes ensuring overall scal health of the incentives economy, e cient labor market operation, product and factor market conditions, property rights protection, e ective regulation, and ease of rm entry and exit. Improving backbone services and inputs such as energy, telecommunications, nance, and other services inputs; improving capacity and coordination of Improving backbone government agencies at the border, international transit arrangements, regional services and reducing and multilateral agreements, and policy reforms that ensure more competitive transactions costs markets for international transport, logistics, and other services that facilitate trade transactions. Proactive policies for Promoting technology creation and adaptation, streamlining product standards overcoming and certi cations, providing trade nance, supporting industry clusters, facilitating government and special economic zones and other spatial developments, and ensuring market failures coordination of economic actors and linkages and spillovers to the local economy. Source: Guilherme Reis and Farole 2012. interindustry spillovers.� The existence of these leads to what Guilherme Reis and Farole (2012, 3) refer to as “the emergence of the ‘behind-the-border’ or ‘competitiveness’ agenda, which targets the supply-side constraints to export performance.� This competitiveness approach can be structured on three pillars (see figure 2.2). Defining Competitiveness Boltho (1996, 3) suggests that the measure of competitiveness is “relative price and/or cost indices expressed in some common currency.� This definition refers to the short term and assumes by implication that structural factors do not feature. For example, price competitiveness will be said to increase in a scenario ­ whereby outward investment rises predicated on lower government borrowing, leading to a decline in the value of the domestic currency as export prices fall and import prices increase in domestic currency terms. However, such price competitiveness is unsustainable, especially if the increase in import prices causes domestic inflation to rise, or if productivity falls in light of lower inward investment (Cantwell 2005). ­ On the other hand, cost-based competitiveness is more substantive in a ­ scenario in which a fall in, for example, unit labor costs leads to lower prices, in turn causing exports to increase and imports to fall, with a resultant increase in the value of the domestic currency. In this scenario, “the perspective rise in the value of the currency is simply the reflection of competitiveness, defined as Clusters of Competitiveness   •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Competitiveness and Its Indicators 35 a relatively rapid growth in productivity and the value of (output and) exports� (Cantwell 2005, 547). An increasing value of the domestic currency by itself is not the achievement of competitiveness. A broader definition of competitiveness examines the medium- and long-run effect of structural factors on economic performance. Economies are competitive when they specialize according to their factor endowments and begin to trade with one another. Krugman (1994, 30, 44) criticized the concept of “national competitiveness� that referred to an economy as more or less competitive when compared to another economy. He cautioned against the “dangerous obsession with competitiveness.� Unlike enterprises, competitiveness among nations is a nonzero sum game—when nations engage in trade based on specialization arising from their factor endowments each participant benefits. In other words, coun- tries need to build dynamic comparative advantage. Ricardo’s theory of compara- tive advantage represents the earliest attempt to understand how nations compete.5 Cantwell (2005, 3) suggests that at the country level, “competitiveness is about the way in which the pattern of international trade evolves over time to reflect changing capabilities and hence comparative advantage.� Thus, it is more about the evolution in the comparative advantage of countries. He cites a num- ber of authors who have contributed to the literature on this position,6 and he examines in detail the relationship between innovation and competitiveness at the firm, industry, national, and regional level. Lall (2001) suggests the ways in which countries can build comparative advantage, depending upon the level of government involvement. Government may provide help for market failure at a functional or a selective level where markets and institutions are deficient. On the other hand, it has no role when factor accumulation is driven solely by free markets and well-functioning institutions. Nabi and Luthria (2002) examine the role of government in facilitating com- petitiveness at the company level by focusing on the demand- and supply-side determinants of competitiveness. They suggest that the role of government is more indirect when it comes to the demand-side factors, such as shareholders, competitors, bank supervisors, and creditors, but more direct when supply-side considerations—such as technology, human capital, and supply chain—are taken into account. Table 2.1 lists the determinants that they identify as important. The definition of competitiveness that has evolved over the previous two to three decades from the business school literature reflects a multifaceted concept that includes nations as well as enterprises.7 The Institute for Strategy and Competitiveness at Harvard University suggests that “a nation’s prosperity depends on its competitiveness, which is based on the productivity with which it produces goods and services. … Many determinants of competitiveness are regional and local, requiring economic strategies for cities and states, not just nations.� Stéphane Garelli (2011, 49), director of the World Competitiveness Center, provides an academic definition, “Competitiveness of nations is a field of economic theory, which analyzes the facts and policies that shape the ability of ­ Clusters of Competitiveness   •  http://dx.doi.org/10.1596/978-1-4648-0049-8 36 Table 2.1 The Who, What, and How of Firms’ Competitiveness How Who/what Firm level Other firms Institutions (public-private) International agreements Demand-side Shareholders Auditing and accounting standards, n.a. n.a. n.a. factors code of ethics, disclosure rules, minority shareholders’ rights Competitors Competition law and policy, antitrust n.a. n.a. n.a. laws, dealing with treatment of mergers, unilateral behavior of powerful corporations, horizontal agreements, vertical restraints, privatization, deregulation Bank supervisors Prudential and regulatory standards, n.a. n.a. n.a. other financial institution supervision practices Creditors Bankruptcy & secured lending regime, n.a. n.a. n.a. debtor-creditor relations, voting rules for institutional investors Supply-side Adapt, absorb, Devote resources to R&D Spillovers, diffusion Measurement, standards, testing, Intellectual property factors and modify and quality; research and protection agreements, technologies technology laboratories WTO membership Attract, build, and Devote resources to in-firm training Spillovers (hiring workers Educational institutions, skills Exchange and training retain human trained by other firms, joint development fund, vocational agreements capital training arrangements) training institutes Manage logistics Devote resources to supplier and Coordination among firms to Physical infrastructure, quality, & Antitrust, e-commerce and improve the vendor development programs integrate production and certification institutions agreements supply chain information systems Source: Nabi and Luthria 2002. Note: R&D = research and development; n.a. = not applicable; WTO = World Trade Organization. Competitiveness and Its Indicators 37 Table 2.2  Definition of Competitiveness and Its Underlying Elements Definition Comments “Competitiveness of nations is a This is a new field taught and researched since 1980. field of Economic theory …� Its origins can be traced to the Classical Economists. Ricardo’s (1819) theory of comparative advantage underlies competitiveness. “… which analyzes facts and Facts are endogenous, for example, an economy’s natural resources policies …� and geographic location area are a given. Policies affect human effort and are affected by human effort. “… that shape the ability of a Facts as outlined above and policies work to establish the nation to create and maintain competitive framework. an environment …� Incorporating the word “maintain� suggests that the competitive framework should be for the long term. “… that sustains more value The emphasis on “more� suggests that nations continuously strive to creation for its enterprises …� fully exploit their competitiveness potential. “… and more prosperity for its This is the ultimate objective of competitiveness—to raise prosperity people.� that may be defined as a mix of income, standard of living, and quality of life. Using the word “prosperity� allows us to emphasize the noneconomic side of competitiveness and ensure that the economic strategy of a firm, nation, or region is not competitiveness at all costs. Source: Garelli 2011. a nation to create and maintain an environment that sustains more value creation for its enterprises and more prosperity for its people.� Table 2.2 examines the elements underlying the definition. Porter et al. (2008, 44) suggests that “competitiveness, then, is measured by productivity� and that productivity determines the prosperity of an economy. Thus, competitiveness determines prosperity also. This interpretation relies upon a broader concept of productivity than the simple output-per-employee-per- hour construct that is most readily understood in a manufacturing context but is difficult to replicate in the services sector. A framework for the study of competitiveness would include the outcome variable of “prosperity� that arises from the combination of “productivity� (that is, competitiveness) and “endowments.� Microeconomic and macroeconomic fac- tors determine competitiveness. Figure 2.3 summarizes the framework for the study of competitiveness. Endowments Endowments refer to a nation’s assets—its land, people, and natural resources. A nation may be rich in assets—may have, for example, a favorable geographic location or an abundance of natural resources. Furthermore, “it could be considered that infrastructure, industrial power, and even education and skills ­ are assets that have been accumulated by past generations� (Garelli 2011, 496). These may not necessarily be competitive, as a nation may be complacent about its endowments. It is important to differentiate between wealth and competitiveness. Studies of competitiveness should control for endowments so ­ Clusters of Competitiveness   •  http://dx.doi.org/10.1596/978-1-4648-0049-8 38 Competitiveness and Its Indicators Figure 2.3 Outcomes and Determinants of Competitiveness Microeconomic competitiveness Prosperity Quality of the Sophistication State of microeconomic of company cluster business operations and development environment strategy Productivity Competitiveness Macroeconomic competitiveness + Social infrastructure Macroeconomic Endowments and political policies institutions Source: Porter et al. 2008. © World Economic Forum. Used with permission; further permission required for reuse. that the outcome reflects competitiveness, that is, the value added to labor and natural resources arising from productive economic activity, rather than the wealth effects arising from resource abundance. The growth literature suggests a negative effect of natural resource abundance on prosperity. This counterintui- tive finding has been explained by the Dutch disease concept, whereby a nation’s rising prosperity, reflected in increasing exports and an appreciation of its exchange rate, is eroded because of “factors of production moving into local ­ etailing that have lower long-term potential for productivity activities such as r growth� (Porter et al. 2008, 45). Productivity The broadest measure of productivity is gross domestic product (GDP) per capita. But how does this relate to competitiveness, given that this measure is arguably both a cause and effect of competitiveness? Sustainable economic growth in an economy supports competitiveness. Krugman (1994, 35) suggests that “competitiveness is a poetic way of saying productivity.� Competitiveness is measured by productivity. The higher the level of productivity in an economy, the more that economy can support higher wages, positive returns to capital (both human and physical), a strong currency, and a high standard of living. Microeconomic Competitiveness Microeconomic factors impact directly on the productivity of firms. These ­ factors are affected by “companies, academic institutions, and many business associations� as well as government, central and local (Porter et al. 2008, 47). Distinguishing between outcomes and determinants helps both the market and government in providing the necessary conditions to strengthen the ­determinants of competitiveness and facilitate the best possible outcome. Porter et al. (2008) suggests two areas of microeconomic competitiveness in this regard—company Clusters of Competitiveness   •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Competitiveness and Its Indicators 39 competitiveness and business environment ­competitiveness. A third ­area—­cluster development—is also relevant, but data difficulties prevent it from being ­ analyzed separately, and it is therefore considered part of the business environment.8 Macroeconomic Competitiveness Macroeconomic factors also affect the productivity of firms, albeit indirectly. As noted by Porter et al. (2008, 46), “they are necessary but not sufficient for higher productivity.� Fiscal policy is thought to weakly affect long-term differences in productivity across geographic areas. Spending and revenue decisions by govern- ment affect the overall prosperity of an economy and indirectly the productivity of its enterprises. Productivity is also affected by “the sustainability of govern- ment financing over time.� For example, high debt levels need to be financed, with implications for spending and revenue. The effect of fiscal policy on the business cycle will also impact the productivity level of firms; “more cyclicality can increase the periods of time in which companies with financing constraints are unable to finance otherwise-profitable long-term investments� (Porter et al. 2008, 47). Monetary policy has a role to play in ensuring a stable and low rate of inflation. Volatile and high rates of inflation can put off investment decisions that might have led to higher productivity in the long run. Social Infrastructure and Political Institutions Social infrastructure and political institutions (SIPIs) have generated significant research attention in recent decades. Three dimensions have been identified in the Global Competitiveness Index (GCI) for summarizing SIPIs. These are basic human capacity, political institutions, and rule of law. Basic human capacity refers to basic education, health care, and a clean environment. Political institu- tions refer to the rules and regulations that govern the economy. Rule of law refers to the existence of property rights and the ability to protect legal rights against private and public interests (Porter et al. 2008). In summary, the definitions of competitiveness suggest that a number of indi- cators are relevant for measuring this economic concept. As noted, it is an ambiguous concept, and differentiating cause from effect is critical in under- standing the underlying factors of competitiveness at the national, industry, and firm level. The following sections examine a number of indicators—both price and nonprice indicators—differentiating within this broad framework those indi- cators from a macroeconomic perspective and those from a microeconomic perspective. See box 2.1 for a discussion of price and nonprice competitiveness in Armenia. Price Indicators of Competitiveness The narrowest measures of competitiveness are those indicators based on r ­ elative prices or costs. Krugman (2011) writes that “measures of relative costs and prices are, in fact, commonly—and unobjectionably—referred to as competitiveness Clusters of Competitiveness   •  http://dx.doi.org/10.1596/978-1-4648-0049-8 40 Competitiveness and Its Indicators Box 2.1 Price and Nonprice Indicators of Competitiveness: The Case of Armenia Weber and Yang (2011) examine competitiveness in Armenia, based on price and nonprice indicators of competitiveness. The authors note a loss in external competitiveness in Armenia from 2008, specifically its declining share of world exports (figure B2.1.1) and an appreciation in its real effective exchange rate (figure B2.1.2). Nonprice indexes of competitiveness suggest a similar loss in competitiveness. Armenia was ranked 98 out of 139 countries in 2010 by the Global Competitiveness Report. Figure B2.1.3 Figure B2.1.1 Armenia’s Share of World Exports 0.012 0.010 0.008 Percent 0.006 0.004 0.002 0 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 Source: Weber and Yang 2011. Figure B2.1.2 Nominal and Real Effective Exchange Rates for Armenia 160 Nominal and real exchange rates (2005 = 100) 140 120 100 80 2006M1 2007M1 2008M1 2009M1 2010M1 Normal effective exchange rate Real effective exchange rate Source: Weber and Yang 2011. Note: M1 = month 1. box continues next page Clusters of Competitiveness   •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Competitiveness and Its Indicators 41 Box 2.1  Price and Nonprice Indicators of Competitiveness: The Case of Armenia (continued) Figure B2.1.3  GDP Dynamics and Global Competitiveness Rankings for Armenia 180 70 GDP level and GDP annual growth rates (%) 160 75 140 80 120 85 100 90 GCI ranking 80 95 60 100 40 105 20 0 110 –20 115 –40 120 2005 2006 2007 2008 2009 2010 GDP level index (2004=100, LHS) GCI rank (RHS) GDP annual growth rate (LHS) Source: Weber and Yang 2011. Note: GDP = gross domestic product; GCI = Global Competitiveness Index. examines gross domestic product (GDP) and global competitiveness rankings for Armenia. The strong growth record—an increase by “about 12 percent on average during 2005–08� coexisted with a declining Global Competitiveness Index (GCI) ranking (Weber and Yang 2011,  8). The poor index ranking stems primarily from the difficulties in doing business in Armenia. Respondents to the survey cite crime, corruption, theft, disorder, and difficulties in accessing finance as major impediments to doing business (Weber and Yang 2011). indicators.� The European Central Bank (ECB) publishes harmonized ­ competitiveness indicators (HCIs) that provide measures of the euro area countries’ price and cost competitiveness consistent with the real effective ­ exchange rates (REERs) of the euro. The HCIs are available based on consumer price indexes, (2) GDP deflators, and (3) unit labor cost indexes (1) ­ for the whole economy. Those based on consumer price indexes are the most widely used; they offer the best data quality and comparability and are timely. However, they often include goods that are not tradable and omit goods that are tradable and that are affected by indirect taxes and subsidies. Indicators based on GDP deflators may be affected by volatility in quarterly GDP figures. Similarly, unit labor costs may be affected by volatility and are subject to revision.9 In his study of international competitiveness, Siggel (2007) identifies those authors or institutions that use price and cost indicators of competitiveness. Table 2.3 summarizes his discussion. Clusters of Competitiveness   •  http://dx.doi.org/10.1596/978-1-4648-0049-8 42 Competitiveness and Its Indicators Table 2.3 Price and Cost Indicators of Competitiveness Author Indicator Comments Lipschitz and McDonald Real exchange rate, real Indicator(s) are associated with a macroeconomic concept (1991); Marsh and effective exchange rate of competitiveness; indicators are unidimensional in that Tokarick (1994); they measure the degree of misalignment of the currency, Helleiner (1991), IMF which enhances or reduces international competitiveness. Indicator(s) can have a static or dynamic interpretation depending on how they are used. Durand and Giorno Price competitiveness Price ratios are associated with a microeconomic concept of (1987); Helleiner competitiveness common in microeconomic studies of single (1991); Jorgenson and industry competitiveness where the relative industry price Kuroda (1992), OECD relative to one or more foreign competitors is translated by the exchange rate, formally resembling the real exchange rate except that prices relate to one industry only. Hickman (1992); Turner Unit labor costs/relative Cost competitiveness is a microeconomic concept of and Golub (1997) unit labor costs competitiveness and is a unidimensional measure at the industry level. Source: Siggel 2007. Note: IMF = International Monetary Fund; OECD = Organisation for Economic Co-operation and Development. However, the link between relative prices and costs at an international level on the one hand and a country’s economic performance on the other is not always straightforward. Turner and Van’t Dack (1993, 9) note that the interna- tional “relative price and cost position can be both cause and result of a country’s economic performance.� For example, high relative prices and costs will hamper a country’s competitiveness internationally, but the levels may stem from an exchange rate appreciation. The appreciation may arise from firms in the econ- omy competing successfully on nonprice factors such as innovation, flexibility, and high-quality goods. Increasing prices and wages suggest a worsening of com- petitiveness, but they are in fact a symptom of success. Also contributing to the ambiguity is the large number of measures in use for prices and costs and the divergences among these measures. Turner and Van’t Dack (1993) examine narrow measures of international competitiveness based on relative prices and costs expressed in a common cur- rency or the REER.10 Underlying the real effective rate is the nominal effective rate, and while this is not usually used as an indicator of competitiveness,11 its derivation has implications for the construction of the real effective rate. Three issues are of note: the choice of the currency basket, the choice of weights, and the base period. The literature concentrates mostly on the first two issues. The choice of weights depends upon the export-import profile of the host country. In countries where only exporters compete, global weights are used. In these situations, “the currencies of partner countries are weighted in proportion to their share in world trade� (Turner and Van’t Dack 1993, 21). Bilateral weights are relevant in situations where the domestic producer is the sole competitor in each export market; that is, there is no competition from exporters from third ­ ommonly markets. Both global and bilateral weights are special cases of the most c applied weights—double weights. These weights apply when “domestic p ­ roducers Clusters of Competitiveness   •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Competitiveness and Its Indicators 43 of import substitutes face competition from the various foreign producers exporting to the domestic market� (Turner and Van’t Dack 1993, 18), and when exporters to that market face competition from one another. Double weights are applied to exporting activity in which the bilateral exchange rates of a country and its competitor countries are weighted according to (1) each country’s contribution to the total supply of competing goods in each separate domestic ­ market and (2) the relative importance of each market in the given country’s international trade (Turner and Van’t Dack 1993). Turner and Van’t Dack (1993, 18) express this formally as follows Import weight w jm = m ij lm j     i      x ik    ∑ x ik x j y i Export weight w jx =    +  x    x j   yi +   h ∑i xh    k+i  j   yk +   ∑ h k xh     mj  m  x j  x Overall weight w i =   wi +  x + m  wi  x j + mi   j j  ( ) where x ij m ij = exports (imports) of country j to (from) country i x j ( m j ) = total exports (imports) of country j y i = domestic production in country j for its home market The choice of currencies in the currency basket is quite narrow, given the vagaries of currencies linked to major international currencies, the fact that some currencies are nonconvertible and others are convertible at multiple exchange rates, and the fact that nominal exchange rate indexes include only those ­ currencies from countries with stable and moderate rates of inflation. “Up to about two dozen� currencies are included in the currency basket (Turner and Van’t Dack 1993, 15). The double weight system is the most widely used. International institutions such as the International Monetary Fund (IMF), Organisation for Economic Co-operation and Development (OECD), and European Commission use this system in addition to the central bank in the United Kingdom, France, Germany, Italy, Spain, and the Netherlands (Turner and Van’t Dack 1993). Quantitative differences in the derived weights are attributed to the disaggregation in trade when calculating the weights.12 The U.S. Federal Reserve and the Bank of Canada do not use the double weight methodology but instead use the global weights index. The REER is the nominal rate deflated by weighted measures of prices or costs. A distinction is usually made between the REER deflated by prices and that deflated by costs. In markets for homogenous goods, price competitiveness Clusters of Competitiveness   •  http://dx.doi.org/10.1596/978-1-4648-0049-8 44 Competitiveness and Its Indicators is not very relevant, for example. For differentiated goods, both price and cost matter in maintaining market share. There are a number of prices that can be used when computing the REER, each with its own advantages and disadvan- tages; these are discussed by Turner and Van’t Dack (1993) and summarized in table 2.4. A wide range of cost indicators has also been used in constructing the REER. However, as noted by Turner and Van’t Dack (1993, 30), “cost as a notion is far from unambiguous.� Labor costs are the most commonly used largely due to data availability and ease of comparison. A disadvantage is their inability to take account of productivity changes. Marsh and Tokarick (1994, 11) define an index of unit labor costs as “the ratio of an index of hourly compensation per worker in the manufacturing sector to an index of output per man hour.�13 Marsh and Tokarick (1994) discuss five indicators of competitiveness: real exchange rates based on consumer price indexes, export unit values of manufac- turing goods, the relative price of traded to nontraded goods, normalized unit labor costs in manufacturing, and the ratio of normalized unit labor costs to value-added deflators in manufacturing. Each of these is related to an economy’s balance of trade in goods and nonfactor services in a way that has implications for that country’s competitiveness. Their conclusion is that no one indicator is Table 2.4 Price Measures of Competitiveness Price index Advantages Disadvantages Use Relative Obvious choice for gauging Force of international competition will limit observed Most export price competitiveness in differences in export prices international prices market conditions where Calculation is limited to goods actually traded—ignores institutions some degree of pricing goods that are potentially traded produce real independence exists Indexes used for prices are derived from unit value effective indexes based on average value of goods traded and exchange can be heavily influenced by composition of exports rates Timeliness of data—measured export unit values relate calculated to prices set in past on this basis Use of export prices alone is inconsistent with double weighting scheme, especially where domestic producer prices are more relevant for domestically produced and sold goods Consumer Calculated on basis of a May be poor proxies of tradable goods Widely used prices basket of goods that is Consumer prices include goods and services that are fairly comparable across not traded countries Excludes capital goods Data are readily available Affected by indirect taxes, subsidies, and price controls and timely As final goods prices, they do not take into account (traded) intermediate goods Wholesale Sometimes chosen to Prices are based on turnover and tend to overweight IMF prices or approximate more closely raw commodities and semi-manufactured goods industrial prices of tradable goods Sometimes high weight given to imported goods producer Prices reflect primarily in makes the index unsuitable for evaluating the prices the more active industrial competitiveness of domestic production sector Source: Turner and Van’t Dack 1993. Note: IMF = International Monetary Fund. Clusters of Competitiveness   •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Competitiveness and Its Indicators 45 a true representation of an economy’s competitiveness. The authors suggest “­competitiveness indicators should be used in conjunction with other indicators in order to obtain an assessment of competitiveness that is as complete as possible� (Marsh and Tokarick 1994, iii). ­ Nonprice Indicators of Competitiveness Three international annual publications in particular shed light on nonprice com- petitiveness. These are the World Competitiveness Yearbook (WCY) published by the International Institute for Management Development (IMD 2012), the Global Competitiveness Report published by the World Economic Forum (WEF 2009, 2012), and the Doing Business report published by the World Bank Group (2012). The reports include readily measurable indicators of competitiveness as well as indicators that are more qualitative in nature and rely upon survey responses. Survey responses can capture those indicators that are less amenable to quantification, such as an economy’s capacity for technological innovation, its degree of product specialization, and the quality of the products and after-sales service, and also those factors that contribute indirectly to improved competi- tiveness. The following sections examine the indicators of competitiveness from these three publications. The WCY The IMD has published competitiveness rankings for selected OECD countries and newly industrialized countries since 1989 in its annual publication, the WCY. The IMD website explains that “the WCY ranks and analyzes the ability of nations to create and maintain an environment in which enterprises can com- pete.�14 Since 2001, the WCY has relied on the country scores achieved on four factors, which rely themselves on five subfactors related to the national environ- ment (for competitiveness).15 See table 2.5. The 20 subfactors comprise more than 300 criteria—331 in 2011. Each sub- factor has a different number of criteria, although each subfactor has the same weight in the overall calculation of results. Two-thirds of the c­ riteria are “hard data,� that is, they make use of data that can be measured, such as GDP. The remaining one-third comprises survey data. Aggregating the 20 subfactors yields the competitiveness score for each country in the sample. Table 2.5  Factors and Subfactors Comprising the National Environment (World Competitiveness Yearbook) Economic performance Government efficiency Business efficiency Infrastructure Domestic economy Public finance Productivity Basic infrastructure International trade Fiscal policy Labor market Technological infrastructure International investment Institutional framework Finance Scientific infrastructure Employment Business legislation Management practices Health and environment Prices Societal framework Attitudes and values Education Source: IMD, “Research Methodology,� http://www.imd.org/research/centers/wcc/research_methodology.cfm. Clusters of Competitiveness   •  http://dx.doi.org/10.1596/978-1-4648-0049-8 46 Competitiveness and Its Indicators The Global Competitiveness Report The goal of the Global Competitiveness Report, published by the WEF, is to “­provide insight and stimulate discussion among all stakeholders on the best strategies and policies to overcome the obstacles to improved competitiveness.� The WEF uses the GCI, “a comprehensive tool that measures the microeconomic and macroeconomic foundations of national competitiveness,� for its analysis (WEF 2012, 4).16 The index represents a weighted average of variables that are grouped into 12 “pillars� of competitiveness and that summarize a set of ­ institutions, policies, and factors that determine the productivity level of an economy. The weights assigned depend upon the level of development of the economy. The WEF identifies three types of development and categorizes countries as possessing basic requirements, efficiency enhancers, or innovation ­ and sophistication factors (table 2.6). The 12 identified pillars may be grouped according to an economy’s stage of development—that is, some pillars are more relevant for specific stages of development (see figure 2.4). ­ At the basic requirements stage of development, an economy is likely to be factor driven, with large pools of unskilled labor. Its sources of competitiveness stem from its factor endowments. Competition is based on price, and low productivity is reflected in low wages. Maintaining competitiveness relies on ­ well-functioning public and private institutions, a stable macroeconomic ­ environment, and a healthy population with at least a basic education. Institutions (pillar 1) represent the legal and administrative framework of an economy within which individuals, firms, and government interact for the people’s welfare. Institutions are also influenced by the attitude of govern- ­ ment toward markets and the manner in which government conducts its own ­ operations. Bureaucracy, red tape, corruption, and lack of transparency impose large costs on businesses and hinder economic growth and develop- ment. The recent global crisis has revealed the importance of private Table 2.6 Subindex Weights and Income Thresholds for Stages of Development (Global Competitiveness Index) Stage 1: Transition Stage 2: Transition Stage 3: factor from stage 1 efficiency from stage 2 innovation driven to state 2 driven to stage 3 driven GDP per capita (US$) thresholdsa < 2,000 2,000–2,999 3,000–8,999 9,000–17,000 > 17,000 Weight for basic requirements subindex (%) 60 40–60 40 20–40 20 Weight for efficiency enhancers subindex (%) 35 35–50 50 50 50 Weight for innovation and sophistication factors subindex (%) 5 5–10 10 10–30 30 Source: WEF 2012. Notes: GDP = gross domestic product. a. For economies with a high dependency on mineral resources, GDP per capita is not the sole criterion for the determination of the stage of development. Clusters of Competitiveness   •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Competitiveness and Its Indicators 47 Figure 2.4 The 12 Pillars of Competitiveness Pillar Basic requirements 1 Institutions Key for 2 Infrastructure factor-driven 3 Macroeconomic environment economies 4 Health and primary education Efficiency enhancers 5 Higher education and training 6 Goods market efficiency Key for 7 Labor market efficiency efficiency-driven 8 Financial market development economies 9 Technological readiness 10 Market size Innovation and sophistication factors Key for 11 Business sophistication innovation-driven 12 Innovation economies Source: WEF 2012. © World Economic Forum. Used with permission; further permission required for reuse. institutions, including accounting standards, transparency, and mutual trust ­ ompetitiveness. The quality of institutions impacts on competitiveness in c and growth because it influences key investment decisions by one and all in the economy. The quality of the institutional framework is important for investment and for how efficiently an economy distributes its wealth. Good ­ governance across private and public institutions is critical in maintaining competitiveness. Infrastructure (pillar 2) includes transport, telecommunications, and energy. An efficient and well-functioning infrastructure is critical for growth and devel- opment. Well-developed infrastructure reduces the distance between various regions and lowers the costs of operation for everyone. In addition to enabling entrepreneurs to distribute their goods and services on time, moving workers to jobs, and enhancing growth, it helps lower nonincome poverty and reduces regional disparities. A stable macroeconomic environment (pillar 3) is critical for economic growth and development and a necessary albeit not a sufficient condition for competitiveness. Public finance management is a major component of macroeconomic stability and an issue of particular relevance in the wake of the ­ 2008 global financial crisis. Many of the advanced economies now face high Clusters of Competitiveness   •  http://dx.doi.org/10.1596/978-1-4648-0049-8 48 Competitiveness and Its Indicators levels of indebtedness that may have adverse consequences for competitiveness. For example: • Sovereign debt in advanced economies may trigger a global recession in the short run. • Higher debt levels are generally associated with higher interest rates; these create a business environment in which it is difficult to raise finance, thus ­ ­lowering investment. • Governments come under pressure to raise taxes to service debt; these taxes may be distortive or stifle further business activity. In the long run, the impact of public debt depends on how the debt is spent— it will benefit competitiveness if it is used to finance investments that raise productivity, but if it is used to finance consumption, it will burden the economy ­ in the long run, resulting in higher interest payments and debt service payments that take up a larger proportion of the government budget, thus forcing a reduction in spending in other areas. ­ A healthy population with at least a basic education (pillar 4) is vital to a coun- try’s competitiveness and productivity, and malnutrition among children does not create a healthy workforce down the road. Weak and ill workers will be less pro- ductive, and the poor health of citizens increases the costs of health care and thus doing business. Public and private investment in preventive health care and provi- sion of health services is important. Quantity and quality of basic schooling is also an important factor for ensuring healthy practices across ­ generations and for providing the initial building blocks upon which further advancement can take place. Without these, firms will find it difficult “to move up the value chain by producing more sophisticated or value-intensive products� (WEF 2012, 5). As economies develop and become more competitive (reflected, for example, in increasing productivity and rising wages), they will transition to the efficiency- driven stage of development (figure 2.4). Economies in this stage must “begin to develop more efficient production processes and increase product quality because wages have risen and they cannot increase prices� (WEF 2012, 9). The pillars necessary for continuing competitiveness are higher education and train- ing, efficient goods markets, well-functioning labor markets, developed financial markets, the ability to harness the benefits of existing technologies, and a large domestic or foreign market (that is, substantial market size). Higher education and training (pillar 5) are critical for economies that would like to move up the value chain and compete in a globalized world. Highly educated workers are better able to adapt to changing production ­ ­ systems and meet new opportunities that the globalized work environment presents. Higher ­ education is measured by enrollment in secondary and tertiary education and includes qualitative measures of education from evaluations by the business com- munity. Vocational training and on-the-job training are used to measure training. Economies with efficient goods markets (pillar 6) benefit from the right mix of products and services based on their particular supply and demand conditions. Clusters of Competitiveness   •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Competitiveness and Its Indicators 49 Factors that adversely impact those conditions and stymie competitiveness are a lack of healthy market competition, perhaps due to obstructions by government intervention17; protectionism; and insufficient knowledge, so that (for example) sellers may fail to respond to customers’ orientation or level of sophistication. Well-functioning labor markets (pillar 7) suggest that workers are performing their jobs to their best ability and are suited to their jobs. Efficient and flexible labor markets require that workers be able to move easily between jobs, at low cost, and without social disruption owing to wage changes. From the worker’s perspective, well-functioning labor markets ensure a link between worker incen- tives and meritocracy, and equity concerns are reflected in equal opportunities for men and women. A well-functioning, efficient financial sector (pillar 8) ensures that savings by residents and nonresidents are put to their most effective and productive use. A properly regulated, sophisticated financial system offering financial products is imperative for business investment that leads to increased productivity in the economy. The technological-readiness pillar (pillar 9) measures how readily an economy can adopt existing technology to increase industry productivity. Of particular importance is an economy’s capacity to harness ICT. “ICT access and usage are key enablers� (WEF 2012, 7). Foreign direct investment can also play an impor- tant role in introducing new technology and best practice to the host economy. Market size (pillar 10) is important because the larger the size of the market, the more opportunity for firms to exploit economies of scale. Globalization has extended the market to include the international market, and a large body of empirical evidence suggests a positive relationship between openness to trade and economic growth. The third stage of development is the innovation and sophistication factors stage. Future progress on productivity will rely upon the production of new and different goods. Firms must be technologically efficient and engaged in e ­ xpanding the economy’s frontiers of technology and knowledge by innovating. Business sophistication and innovation are critical pillars in this stage. Business ­sophistication (pillar 11) refers to the quality of the overall business networks within a country and the quality of the individual firms’ operation and strategies. The former relies upon a sufficient number of high-quality suppliers; where these are in geographic proximity to the firm, forming a cluster, productivity is heightened. “Branding, marketing, distribution, advanced production processes, and the production of unique and sophisticated products� are responsible for the firm’s business sophistication (WEF 2012, 8). Technological innovation (pillar 12) “is the final ­ pillar of competitiveness� identified by the Global Competitiveness Report. This pillar aims to move the economy further toward its technological and knowledge frontier by technologically efficient firms engaged in innovation. The environ- ment for this relies upon “public and private investment, particularly in research and development (R&D); the presence of high-quality scientific research institu- tions; extensive collaboration in research between university and industry; and the protection of intellectual property� (WEF 2012, 8). Clusters of Competitiveness   •  http://dx.doi.org/10.1596/978-1-4648-0049-8 50 Competitiveness and Its Indicators As noted in the Global Competitiveness Report, the pillars of competitiveness are strongly interrelated; for example, “a strong innovation capacity (pillar 12) will be very difficult to achieve without a healthy, well-educated and trained workforce (pillars 4 and 5) that is adept at absorbing new technologies (pillar 9), and without sufficient financing (pillar 8) for R&D or an efficient goods market that makes it possible to take new innovations to market (pillar 6)� (WEF 2012, 8). The pillars are aggregated into a single index based on weights, and results for each pillar are also published. These results provide information for an economy on gaps in competitiveness. Doing Business: Measuring Business Regulations The Doing Business project was launched in 2002 by the World Bank to look at the regulations facing small- and medium-size businesses throughout their life cycle. Quantitative data measuring business regulation environments across countries are gathered and analyzed in an annual report. The first annual report was published in 2003 and covered 5 indicator sets and 133 economies. According to the Doing Business website, the goal of the report is to “provide an objective basis for understanding and improving the regulatory environment for business around the world.�18 The most recent report (World Bank 2012) covers 11 indicator sets and 183 economies (see table 2.7). The data come from two sources—one from the domestic laws and r ­ egulations and administrative requirements pertaining to businesses in the host country and the second from “time-and-motion indicators that measure the efficiency in achieving a regulatory goal (such as granting the legal identity of a business)� (World Bank 2012, 17). Most of the cost indicators are based on official fee schedules where available. For indicators such as dealing with construction permits, enforcing contracts, and resolving insolvency, the time and cost compo- ­ nents are based on actual practice. However, the respondents are those who are very familiar with these elements of the regulatory environment—such as the professionals or government officials who routinely administer or advise on Table 2.7 Eleven Areas of Business Regulation Measured by Doing Business Start-up Expansion Operations Insolvency Starting a Registering property Dealing with construction Resolving business Procedures, time, and cost permits insolvency Minimum capital Procedures, time, and cost Time, cost, and requirement recovery Getting credit Getting electricity Procedures, time, rate Credit information systems Procedures, times, and cost and cost Movable collateral laws Protecting investors Paying taxes Disclosure and liability in related- Payments, time, and total party transactions tax rate Enforcing contracts Trading across borders Procedures, time, and cost to Documents, time, and cost resolve a commercial dispute Employing workers Source: World Bank 2012. Clusters of Competitiveness   •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Competitiveness and Its Indicators 51 the legal and regulatory requirements covered in each topic.19 “The credit infor- mation survey is answered by officials of the credit registry. Freight f­orwarders, accountants, architects, and other professionals answer the survey related to trad- ing across borders, taxes, and construction permits� (World Bank 2012, 22). A simple averaging is applied to all topics and the components within each topic: each is weighted equally. The indicators are based on standardized case studies where the business is located in the largest city. This approach may prove limiting in some countries where there are differences in regulations. Doing Business also publishes ­ subnational studies for a range of countries and recently published a pilot study on the business environment in the second-largest city in three large economies in order to examine the within-country variations. The standardized case approach assumes a limited liability company or its equivalent. The results from the Doing Business survey correlate nicely with those from other studies examining competitiveness, such as the OECD product market regulation indicators and the GCI (see figures 2.5 and 2.6). The OECD product market regulations inform on the extent to which the regulatory environment promotes or curtails competition. They include mea- sures on “the extent of price controls, the licensing and permit system, the degree of simplification of rules and procedures, the administrative burdens and legal and regulatory barriers, the prevalence of discriminatory procedures and the degree of government control over business enterprises� (World Bank 2012, 18). The correlation between these measures and those on the ease of doing business is .72. Furthermore, the correlation between the ease of doing business and the GCI is .82 (figure 2.6). Figure 2.5 Correlation between Doing Business Rankings and OECD Rankings of Product Market Regulation 40 35 Ranking on OECD product market 30 regulation indicators 25 20 15 10 5 0 20 40 60 80 100 120 140 Ranking on the ease of doing business Source: World Bank 2012. Note: OECD = Organisation for Co-operation and Development. Clusters of Competitiveness   •  http://dx.doi.org/10.1596/978-1-4648-0049-8 52 Competitiveness and Its Indicators Figure 2.6 Correlation between Doing Business Rankings and World Economic Forum Rankings on Global Competitiveness 140 Ranking on Global Competitiveness Index 120 100 80 60 40 20 0 20 40 60 80 100 120 140 160 180 Ranking on the ease of doing business Source: World Bank 2012. Conclusion The chapter examined the concept of competitiveness by first noting its many applications—to the firm, the industry, the region, the nation, and the global economy. It highlighted its growing popularity among policy makers, lawmakers, and researchers. Competitiveness has been given a new impetus in the wake of the most recent financial crisis, in 2008. Stronger competitiveness is particularly relevant in a growth-challenged world. Price and nonprice measures of competitiveness help to classify broadly the concept. Price and cost measures were identified and their advantages and disad- vantages discussed. Nonprice measures have proliferated in the business litera- ture, and annual surveys of competitiveness at the national level have become popular over the last few decades. The role of government in promoting com- petitiveness was referenced throughout the chapter; it is a theme that recurs throughout this book. Notes 1. A substantial new literature has emerged from the business studies discipline. 2. Environmentally friendly choices need not be at the cost of economic competitive- ness. The firm that addresses environmental improvements through innovation and new technology may actually increase its economic competitiveness. For a list of papers that examine the issue of environmental quality and competitiveness, see Institute for Strategy and Competitiveness, “Environmental Quality and Competitiveness,� http://www.isc.hbs.edu/soci-environmental.htm. 3. The theory of comparative advantage (Ricardo 1819) applies this approach at the level of the economy. Clusters of Competitiveness   •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Competitiveness and Its Indicators 53 4. In recognition of this, the World Bank has developed the Trade Competitiveness Diagnostic (TCD) toolkit. The TCD provides “a framework, guidelines, and practical tools needed to conduct an analysis of trade competitiveness. The toolkit can be used to assess the competitiveness of a country’s overall basket of exports, as well as specific traded sectors� and is intended “for policy makers and practitioners involved in analy- sis of trade performance and design of trade and industrial policy� (Guilherme Reis and Farole 2012, 1). 5. Dwyer and Kim (2003) identify the following authors addressing the comparative advantage and/or price competitiveness aspect of competitiveness: Bellak and Weiss (1993); Cartwright (1993); Durand and Giorno (1987); Fagerberg (1988); Fakiolas (1985); Hilke and Nelson (1988); Hodgetts (1993); Porter (1990); Rugman (1991); and Rugman and D’Cruz (1993). 6. Cantwell (2005, 3) writes, “Dynamic accounts of the paths of international trade and investment were revived . . . by the technology gap approach (Posner 1961) and the product cycle model (Vernon 1966). It was only in the 1980s that scholars based at Sussex once again wedded an analysis of structural shifts over time in the pattern of international trade to a more realistic approach to innovation—see Soete 1981; Dosi and Soete 1988; Dosi, Pavitt, and Soete 1990; and Fagerberg’s 1987 paper on structural changes in international trade (reprinted as chapter 7 in Fagerberg 2002).� ­ 7. The World Economic Forum (WEF) defines competitiveness as “the set of institutions, policies and factors that determine the level of productivity of a city or region� (2007). 8. Chapter 5 in this book examines clusters and competitiveness. 9. See European Central Bank, “Harmonised Competitiveness Indicators,� http://www. ecb.int/stats/exchange/hci/html/index.en.html. 10. The real effective exchange rate is the nominal rate (a weighted average of various bilateral exchange rates, with the choice and weights of the bilateral rates reflecting their relative importance to the economic issue being analyzed) deflated by a similarly weighted average of foreign prices or costs, relative to those at home (Turner and Van’t Dack 1993). 11. Turner and Van’t Dack (1993, 13) note that authors such as Rosensweig (1987) advocate the use of the nominal effective exchange rate as a measure of competitive- ­ ness given “its timeliness, its greater frequency, the ease of data collection and of cross-country comparability, and the avoidance of measurement errors in the price or ­ cost series.� 12. Turner and Van’t Dack (1993) note that the International Monetary Fund (IMF) uses highly disaggregated manufacturing data when applying the methodology, while the European Commission includes all traded goods and services. 13. “A real exchange rate indicator is then computed by dividing the index of unit labor costs in the home country by the index of unit labor costs for the sixteen industrial countries for which data are collected by the IMF� (Marsh and Tokarick 1994, 11). 14. See http://www.imd.org/research/centers/wcc/research_methodology.cfm. 15. Up to 2001, the World Competitiveness Yearbook published a competitiveness score for its sample of countries based upon eight factors that comprised a number of subfac- tors each. The eight factors were domestic economy, internationalization, government, finance, infrastructure, management, science and technology, and people. 16. “The first version of the Global Competitiveness Index was published in 2004� (WEF 2012, 44). Clusters of Competitiveness   •  http://dx.doi.org/10.1596/978-1-4648-0049-8 54 Competitiveness and Its Indicators 17. Examples include “distortionary or burdensome taxes� and “restrictive and ­ discriminatory rules on foreign direct investment that limit foreign ownership� (WEF 2012, 7). 18. See http://www.doingbusiness.org/about-us. 19. Estimates are made by “practitioners with significant and routine experience in the transaction� (World Bank 2012, 22). References Bellak, C. J., and A. Weiss. 1993. “A Note on the Austrian ‘Diamond.’� Management International Review 33 (2): 109–18. Boltho, A. 1996. “The Assessment: International Competitiveness.� Oxford Review of Economic Policy 12 (3): 1–16. Cantwell, J. 2005. “Innovation and Competitiveness.� In Handbook of Innovation, edited by J. Fagerberg, D. C. Mowery, and R. R. Nelson, 543–67. Oxford, UK: Oxford University Press. Cartwright, W. R. 1993. “Multiple Linked ‘Diamonds’ and the International Competitiveness of Export-Dependent Industries: The New Zealand Experience.� Management International Review 33: 55–70. Dosi, G., K. L. R. Pavitt, and L. L. G. Soete. 1990. The Economics of Technical Change and International Trade. London: Harvester Wheatsheaf. Dosi, G., and L. L. G. Soete. 1988. “Technical Change and International Trade.� In Technical Change and Economic Theory, edited by G. Dosi, C. Freeman, R. Nelson, G. Silverberg, and L. L. G. Soete, 401–30. London: Frances Pinter. Durand, M., and C. Giorno. 1987. “Indicators of International Competitiveness: Conceptual Aspects and Evaluation.� OECD Economic Studies 9: 147–82. Dwyer, L., and C. Kim. 2003. “Destination Competitiveness: Determinants and Indicators.� Current Issues in Tourism 6 (5): 369–414. Fagerberg, J. 1988. “International Competitiveness.� Economic Journal 98: 355–74. ———. 2002. Technology, Growth and Competitiveness: Selected Essays. Cheltenham, UK: Edward Elgar. Fakiolas, T. 1985. “Basic Causes of Soviet Industry’s Low International Competitiveness.� Journal of Economic Studies 12 (5): 39–52. Garelli, S. 2011. “The Fundamentals and History of Competitiveness.� In IMD World Competitiveness Yearbook 2011, appendix 3. Lausanne, Switzerland: International Institute for Management Development. Guilherme Reis, J., and T. Farole. 2012. Trade Competitiveness Diagnostic Toolkit. Washington, DC: World Bank. Helleiner, G. K. 1989. “Transnational Corporations and Direct Foreign Investment.� In Handbook of Development Economics, vol. 2, chapter 27. Amsterdam: Elsevier Science Publishers BV. ———. 1991. “Increasing International Competitiveness: A Conceptual Framework.� In Increasing the International Competitiveness of Exports from Caribbean Countries, edited by Y. Wen and J. Sengupta, 17–26. Washington, DC: World Bank. Hickman, B. G. 1992. “International Productivity and Competitiveness: An Overview.� In  International Productivity and Competitiveness, edited by B. G. Hickman, 3–32. New York: Oxford University Press. Clusters of Competitiveness   •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Competitiveness and Its Indicators 55 Hilke, J., and P. Nelson. 1988. US International Competitiveness: Evolution or Revolution? New York: Praeger. Hodgetts, R. M. 1993. “Porter’s Diamond Framework in a Mexican Context.� Management International Review 33 (special issue): 41–54. IMD (International Institute for Management Development). 2012. IMD World Competitiveness Yearbook 2012. Lausanne, Switzerland: IMD. Jorgenson, D. W., and M. Kuroda. 1992. “Productivity and International Competitiveness in Japan and the United States, 1960–1985.� In International Productivity and Competitiveness, edited by B. G. Hickman, 203–29. New York: Oxford University Press. Krugman, P. 1994. “Competitiveness—A Dangerous Obsession.� Foreign Affairs 73 (2): 28–44. http://www.pkarchive.org/global/pop.html. ———. 2011. “Competitiveness.� The Conscience of a Liberal (blog), January 22. http:// krugman.blogs.nytimes.com/2011/01/22/competitiveness. Lall, S. 2001. “Competitiveness Indices and Developing Countries: An Economic Evaluation of the Global Competitiveness Report.� World Development 29 (9): 1501–25. Lipschitz, L., and D. McDonald. 1991. “Real Exchange Rates and Competitiveness: A Clarification of Concepts and Some Measurements for Europe.� IMF Working Paper 91/25, International Monetary Fund, Washington, DC. Marsh, L. W., and S. P. Tokarick. 1994. “Competitiveness Indicators: A Theoretical and Empirical Assessment.� Working Paper 94/29, International Monetary Fund, Washington, DC. Nabi, I., and M. Luthria. 2002. Building Competitive Firms: Incentives and Capabilities. Washington, DC: World Bank. Porter, M. E. 1990. The Competitive Advantage of Nations. New York: Free Press. Porter, M. E., M. Delgado, C. Ketels, and S. Stern. 2008. “Moving to a New Global Competitiveness Index.� In The Global Competitiveness Report 2008–2009, 43–63. World Economic Forum, Geneva, Switzerland. Posner, M. V. 1961.“International Trade and Technical Change.� Oxford Economic Papers 13 (3): 323–41. Ricardo, David. 1819. On the Principles of Political Economy and Taxation. London, UK: John Murray. Rosensweig, J. A. 1987. “Constructing and Using Exchange Rate Indexes.� Economic Review (Summer): 4–16. Rugman, A. M. 1991. “Diamond in the Rough.� Business Quarterly 55 (3): 61–64. Rugman, A. M., and J. R. D’Cruz. 1993. “The ‘Double Diamond’ Model of International Competitiveness: The Canadian Experience.� Management International Review 33 (special issue): 17–39. Siggel, E. 2007. “International Competitiveness and Comparative Advantage: A Survey and a Proposal for Measurement.� Paper prepared for CESifo Venice Summer Institute, Venice International University, Italy, July 20–21. Soete, L. L. G. 1981. “A General Test of Technological Gap Trade Theory.� Weltwirtschaftliches Archiv 117 (4): 638–60. Turner,A., and S. Golub. 1997.“Towards a System of Unit Labor Cost-Based Competitiveness Indicators for Advanced, Developing and Transition Countries.� Staff Studies for the World Economic Outlook, 47–60. Washington, DC: International Monetary Fund. Clusters of Competitiveness   •  http://dx.doi.org/10.1596/978-1-4648-0049-8 56 Competitiveness and Its Indicators Turner, P., and J. Van’t Dack. 1993. “Measuring International Price and Cost Competitiveness.� Economic Papers No. 39, Bank for International Settlements, Geneva, Switzerland. Vernon, R. 1966. “International Investment and International Trade in the Product Cycle.� Quarterly Journal of Economics 80 (2): 190–207. Weber, A., and C. Yang. 2011. “Armenia: An Assessment of the Real Exchange Rate and Competitiveness.� Working Paper WP/11/20, International Monetary Fund, Washington, DC. WEF (World Economic Forum). 2009. The Global Competitiveness Report 2008–2009. Geneva: World Economic Forum. ———. 2010. The Global Competitiveness Report 2010–2011. Geneva: World Economic Forum. ———. 2012. The Global Competitiveness Report 2012–2013. Geneva: World Economic Forum. World Bank. 2012. Doing Business 2012: Doing Business in a More Transparent World. Washington, DC: World Bank. Clusters of Competitiveness   •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Chapter 3 National Competitiveness Paul Krugman (1994) criticized the concept of national competitiveness nearly 20 years ago when he attested that the term was meaningless as applied to national economies. He was reacting to the view, increasingly propounded by governments and academics, that nations were in competition with one another and that only the most competitive would succeed. Indeed, he cautioned about the dangerous obsession with competitiveness applied to national economies. Elsewhere indexes of national competitiveness were being developed to describe international competitive performance.1 The best known of these is the Global Competitiveness Index (GCI) maintained by the World Economic Forum (WEF). The WEF defines national competitiveness as “the set of institutions, policies, and factors that determine the level of productivity of a country� (WEF 2009, 3). Competitiveness at the national level is both a static and a dynamic concept. On the one hand, productivity determines the prosperity of a nation, a prosperity that will be shared among its citizens. On the other hand, productiv- ity also determines the returns on investment in an economy, which in turn determine the growth rate of that economy. Globalization, liberalization, and rapid technical change have contributed to the enormous emphasis on national competitiveness. The following sections examine what is meant by national competitiveness and explore the policy implications of increasing a nation’s competitiveness domesti- cally and internationally. We present global competitiveness rankings as summa- rized by the International Institute for Management Development (IMD), the GCI from the WEF, and the Doing Business report from the World Bank. Defining National Competitiveness National competitiveness may be defined as “a field of economic theory, which analyzes the facts and policies that shape the ability of a nation to create and maintain an environment that sustains more value creation for its enterprises and more prosperity for its people� (Garelli 2011, 49). This definition informs the competitiveness measures developed by the WEF and the IMD. The definition provides for a broad measure of competitiveness, incorporating macroeconomic Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8   57   58 National Competitiveness and microeconomic factors, and is discussed more fully below in the context of the country competitiveness rankings produced annually by the WEF and IMD. In addition, other more narrow price- and cost-based measures of competi- tiveness are widely used to measure national competitiveness. These various measures incorporate elements such as labor productivity, consumer prices, unit labor costs, terms of trade, and Balassa’s index of revealed comparative advantage, with the consumer price index (CPI)–based real effective exchange rate (REER) the most widely used measure (Bakardzhieva, Ben Naceur, and Kamar 2010).2 The CPI-based REER takes account of price/inflation differences between trad- ing partners after allowing for nominal effective exchange rate movements and after weighting individual country components according to trade or other such weights. Basically, in the literature, “movements in REERs provide an indication of the evolution of a country’s aggregate external price competitiveness.�3 All other things held constant, an appreciation in the REER means a loss in competitiveness for the country, while a depreciation means the opposite. ­ However, all other things are rarely constant, and an appreciation may not always indicate a loss of competitiveness. For example, the Balassa-Samuelson effect suggests that countries with higher productivity growth in the traded goods sector experience higher prices in the nontraded goods sector and thus an appre- ciation in the REER as the underlying equilibrium exchange rate rises. (This effect, which is transmitted through wage movements in the nontraded sector, does not imply declining competitiveness, as higher prices reflect higher underlying pro- ductivity. For instance, although price levels are higher in, say, Germany, than in most developing countries, this does not mean that Germany is less competitive, as it is more productive.) Of interest in measuring true ­ competitiveness gains or losses, therefore, is the distance that the REER moves from its equilibrium value, which itself changes over time for various reasons (including not only productivity movements but also terms of trade changes, the accumulation of assets, and so on). Bearing these caveats in mind, we take a quick look, in figure 3.1, at ­movements in measured REER competitiveness across a set of wealthy, highly productive countries. Taking 2000 as the starting point, we can see that, all other things equal, there have been major changes in the REER over a short period (just over a decade). Among this sample, Switzerland has seen the greatest appreciation in its REER (that is, the greatest measured loss of competitiveness), while Japan has seen the largest REER depreciation (that is, the greatest measured gain in competi- tiveness). Of course, not all is equal, and Switzerland’s equilibrium real exchange rate has likely become stronger over this period given its major accumulation of international reserves, while Japan’s has probably become weaker, as its public debt position has worsened, growth has stagnated, and its external position has worsened. Another major issue is that in any chosen base year, not all c ­ ountries are equal in terms of starting competitiveness. So REER measures are good at captur- ing changes in competitiveness rather than competitiveness levels themselves. (The WEF and IMD measures do aim to capture competitiveness levels.) Nevertheless, despite the caveats, REERs are good s ­ummary measures of competitiveness moves and are therefore widely used by policy makers and economists alike. Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 National Competitiveness 59 Figure 3.1 REER Competitiveness Gains and Losses, Selected Countries, 2000–11 130 120 110 Index of REERs 100 90 80 70 60 00 01 02 03 04 05 06 07 08 09 10 11 20 20 20 20 20 20 20 20 20 20 20 20 Year United States Denmark Singapore Japan Netherlands Finland United Kingdom Switzerland Germany Sweden Source: Calculations based on International Monetary Fund, “International Financial Statistics,� http://elibrary-data.imf​ .org/ FindDataReports.aspx?d=33061&e=169393. Note: REERs = real effective exchange rates. Competitiveness Rankings A country’s competitiveness may be viewed as its competitive position relative to other countries. It has become important for a nation to assess its competitive- ness in the increasingly globalized world. A favorable macroeconomic environ- ment is a necessary but insufficient condition for competitiveness; it also matters how productive the nation’s enterprises are. Thus, microeconomic factors are critical. Each year, the IMD and the WEF publish rankings of national competi- tiveness based on macroeconomic and microeconomic factors and subjective assessments by business owners. In addition, the World Bank also publishes the Doing Business report. The report ranks countries according to the ease of doing business in their economies. The most recent report (World Bank 2013) covers 11 indicator sets and 183 economies. The following sections examine the com- petitiveness rankings of countries for 2012 from the reports published by the IMD, WEF, and the World Bank. World Competitiveness—IMD Rankings Fifty-nine economies were assessed by the IMD in 2011. Figure 3.2 presents the rankings for the top 20 countries, with the previous year’s ranking in parenthesis. Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 60 National Competitiveness Figure 3.2  World Competitiveness Scoreboard 2011: Top 20 Economies 100.000 (2) Hong Kong SAR, China 1 100.000 (3) United States 1 98.557 (1) Singapore 3 94.063 (6) Sweden 4 92.588 (4) Switzerland 5 92.011 (8) Taiwan, China 6 90.782 (7) Canada 7 90.219 (15) Qatar 8 89.259 (5) Australia 9 87.824 (16) Germany 10 86.475 (11) Luxembourg 11 86.418 (13) Denmark 12 86.313 (9) Norway 13 85.707 (12) Netherlands 14 84.380 (15) Finland 15 84.120 (10) Malaysia 16 81.629 (17) Israel 17 81.619 (14) Austria 18 81.100 (18) China 19 80.278 (22) United Kingdom 20 79.799 (20) New Zealand 21 78.499 (23) Korea, Rep. 22 77.599 (25) Belgium 23 77.101 (21) Ireland 24 77.827 (28) Chile 25 0 10 20 30 40 50 60 70 80 90 100 Index score Source: IMD, http://www.vi.is/files/IMD%202011%20-%20listar_831280280.pdf. Note: The top 20 of 59 economies ranked by the IMD are shown here, from most to least competitive. The scores shown on the y-axis are indexes (0 to 100) generated for the purpose of constructing charts and graphics. For the first time since the rankings began, the top position was held jointly by Hong Kong SAR, China, and the United States, which both ranked slightly ahead of Singapore, the top ranked country in 2010. Sweden gained ground and now occupies the 4th position, while Germany gained six places to occupy the 10th position overall. Emerging market countries—for example, Qatar, the Republic of Korea, and Turkey—continue to gain in competitiveness, while the aftereffects of the 2008 global recession leave just four big economies in the top 20. The story behind the ranking suggests a “greater self-reliance of countries. [World competitiveness] increasingly emphasizes re-industrialization, exports, and a more critical look at delocalization.�4 Competitiveness Trends—The GCI The GCI has ranked countries since 2005. Figure 3.3 presents an overview of competitiveness trends from 2005 to 2011 for the United States, China, advanced economies, and emerging and developing economies. Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 National Competitiveness 61 Figure 3.3 Competitiveness Trends, 2005–11 7 6 5 GCI score 4 3 2 1 2005–06 2006–07 2007–08 2008–09 2009–10 2010–11 2011–12 GCI edition United States China Advanced economies Emerging and developing economies Source: WEF 2012. Note: GCI = Global Competitiveness Index. The analysis is based on a constant sample composed of the 113 economies already covered in 2005. Country classification is derived from the International Monetary Fund (IMF) and reflects the situation as of April 2011. Weights for the computation of group-weighted averages are based on each economy’s share of gross domestic product in its group. Data are taken from IMF, World Economic Outlook, April 2011, http://www.imf.org/external/pubs/ft/weo/2011/01/pdf/text.pdf. The weighted average overall GCI score is computed for 80 emerging market and developing economies and for 33 advanced economies. There is some convergence of the two groups, but it is gradual. The GCI score for the emerging ­ market and developing economies was 4.1 in 2005, increasing to 4.4 in 2011, while the respective scores for the advanced market economies was 5.4, ­declining to 5.2. Thus, the spread between the two groups has narrowed from 1.3 to 0.8. The reduction in spread is primarily due to the contrasting experience in China and the United States, the two largest economies. The United States was ranked first overall in 2005, falling to fifth place in 2011. China on the other hand improved its ranking, experiencing a 0.5 increase in its GCI over the study period. Figure 3.3 suggests a gradual catching up of the emerging and developing economies, with some stagnation among the advanced economies. WEF (2012) notes that four advanced economies suffered a large loss in their GCI score over the study period: the United States (−0.4), Greece (−0.3), and Ireland and Iceland (−0.2 each). However, score loss is not characteristic of all advanced economies—both Switzerland and Sweden have gained 0.3 points since 2005 (WEF 2012). Table 3.1 examines the ranking for the top 10 countries in 2011–12 and compares this with their ranking between 2005–06 and 2008–09. Switzerland ­ ranked first in 2011–12, followed by Singapore, Sweden, and Finland, with the United States in the fifth position as noted above. Three of these countries Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 62 National Competitiveness Table 3.1  Global Competitive Index for Top 10 Countries for 2011–12 versus 2005–06 and 2008–09 Rankings GCI 2011–12 GCI 2005–06 GCI 2008–09 Country/economy Rank 1/142 Score Rank 1/117 Rank 1/134 Switzerland 1 5.74 4 2 Singapore 2 5.63 5 5 Sweden 3 5.61 7 4 Finland 4 5.47 2 6 United States 5 5.43 1 1 Germany 6 5.41 6 7 Netherlands 7 5.41 11 8 Denmark 8 5.40 3 3 Japan 9 5.40 10 9 United Kingdom 10 5.39 9 12 Source: WEF 2006, 2009, 2012. Note: GCI = Global Competitiveness Index. (the United States, Switzerland, and Singapore) were in the top five slots in 2005–06 and 2008–09, with Denmark replacing Sweden in 2005–06 and Finland in 2008–09. The ranking has changed over time, with Singapore moving to second place in 2011–12 from fifth place in 2005–06 and 2008–09. Germany and Japan have been consistently ranked in the top 10 for the years chosen, while the Netherlands and the United Kingdom are usually there also. Examining the country rankings across the 12 pillars of competitiveness ­ outlined in chapter 2 clarifies the GCI score and rankings. As would be expected, all countries score highly on these pillars with some exceptions. Poor macroeco- nomic stability is an area of weakness for several countries: the United States, which has had repeated fiscal deficits leading to large levels of public indebted- ness; the United Kingdom, which had double-digit fiscal deficits in 2010 and a large public debt (77 percent of gross domestic product [GDP] in 2010) coupled with a comparatively low national savings rate (12.3 percent of GDP in 2010); and Japan, which had high budget deficits and the highest public debt of the sample (220 percent of GDP in 2010). Table 3.2 shows the country rankings across the 12 pillars for the top 10 economies. The Global Competitiveness Report identifies five regions—Europe and North America, Asia and the Pacific, Latin America and the Caribbean, the Middle East and North Africa, and Sub-Saharan Africa. There are no regional rankings, but it is possible to identify both top performers and those that have not fared so well among the regional country groupings. The Global Competitiveness Report highlights those areas in which there is room for improvement, that is, where a ­ country is underperforming and requires further efforts in order to attain greater competitiveness. As one would expect, the areas for improvement increase as the country rankings decrease. Thus, the economies ranked in the top 40 have few areas for improvement, but once we cross that threshold, the areas for improve- ment increase. One further point is that not all pillars are relevant, depending on an economy’s level of growth and development as captured in the three Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Table 3.2 Twelve Pillar Rankings for GCI Top 10 Countries, 2011–12 Health Higher and education Goods Labor Financial Macroeconomic primary and market market market Technological Market Business Institutions Infrastructure environment education training efficiency efficiency development readiness size sophistication Innovation Switzerland 6 5 7 8 3 5 1 7 1 39 3 1 Singapore 1 3 9 3 4 1 2 1 10 37 15 8 Sweden 2 13 13 18 2 7 25 11 2 31 2 2 Finland 4 19 20 1 1 21 15 9 12 54 9 3 United States 39 16 90 42 13 24 4 22 20 1 10 5 Germany 19 2 30 23 7 26 64 39 14 5 4 7 Netherlands 10 7 36 7 8 9 23 23 5 18 5 12 Denmark 5 10 31 28 6 16 6 17 4 53 6 10 Japan 24 15 113 9 19 18 12 32 25 4 1 4 United Kingdom 15 6 85 14 16 19 7 20 8 6 8 13 Source: WEF 2012. Note: GCI = Global Competitiveness Index. 63 64 National Competitiveness ­ategories of basic requirements, efficiency enhancers, and innovation and c ­sophistication factors. Insufficient or lagging development in institutions is of concern in Italy, Turkey, the Russian Federation, Ukraine, and Greece (table 3.3). Public insti- tutions (government inefficiency, corruption, undue influence) hamper competitiveness in Greece, Turkey, and Italy. Little improvement has occurred ­ in the weak institutional framework in Ukraine and Russia. Rigidities in labor markets and weak financial markets also hamper competitiveness in many of the countries in the Europe and North America region. Labor market concerns include strict hiring and firing rules in France and Spain, rigid labor markets in Italy and Portugal hindering employment creation, a disconnect between sala- ries and p ­ roductivity in Portugal and Spain, and inefficient labor markets in Turkey and Greece (table 3.3). Insufficiently developed financial markets in Italy, a low national savings rate in Spain, and a high level of debt in Portugal hinder finance for business development in these countries. Financial markets in Iceland, Greece, and Ireland have been weakened, while Russian markets remain ­ unstable, with poor assessments of the banks. Macroeconomic instabil- ity is a threat to competitiveness and a factor adversely affecting the competi- tiveness rankings in a number of European countries: persistent deficits and high levels of public debt characterize the macroeconomies in Belgium, Ireland, and Spain, and the o ­ ngoing sovereign debt crisis in Greece hampers competitiveness on many levels. Table 3.4 and WEF 2012 examine the competitiveness rankings and underly- ing pillars in selected economies in Asia and the Pacific region. The disparity in competitiveness rankings is highest in Asia and the Pacific. This region is home to Table 3.3  GCI and Pillar Rankings for North America and Europe Region, Selected Countries, 2011–12 Pillar Country GCI rank 1 2 3 4 5 6 7 8 9 10 11 12 Belgium 15 27 17 60 2 5 14 44 28 11 26 11 15 France 18 28 4 83 16 20 38 68 18 13 7 14 17 Ireland 29 23 29 118 12 22 13 17 115 17 56 22 23 Iceland 30 25 14 131 5 9 40 10 108 3 128 28 19 Spain 36 49 12 84 44 32 66 119 64 28 13 34 39 Poland 41 52 74 74 40 31 52 58 34 48 20 60 58 Italy 43 88 32 92 20 41 59 123 97 42 9 26 43 Portugal 45 51 23 111 34 35 62 122 78 19 45 50 32 Turkey 59 80 51 69 75 74 47 133 55 55 17 58 69 Russian Federation 66 128 48 44 68 52 128 65 127 68 8 114 71 Ukraine 82 131 71 112 74 51 129 61 116 82 38 103 74 Greece 90 96 45 140 37 46 107 126 110 47 42 77 88 Source: WEF 2012. Note: Numbers in bold type denote areas for improvement. Pillars are as follows: 1= institutions; 2 = infrastructure; 3 = macroeconomic environment; 4 = health and primary education; 5 = higher education and training; 6 = goods market efficiency; 7 = labor market efficiency; 8 = financial market development; 9 = technological readiness; 10 = market size; 11 = business sophistication; 12 = innovation. Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 National Competitiveness 65 Table 3.4  GCI and Pillar Rankings for Asia and the Pacific Region, Selected Economies, 2011–12 Pillar Country GCI rank 1 2 3 4 5 6 7 8 9 10 11 12 Hong Kong SAR, China 11 9 1 8 27 24 3 3 2 6 28 19 25 Taiwan, China 13 31 20 22 11 10 11 33 24 24 16 13 9 Australia 20 13 24 26 10 11 22 13 6 22 19 29 22 Malaysia 21 30 26 29 33 38 15 20 3 44 29 20 24 Korea, Rep. 24 65 9 6 15 17 37 76 80 18 11 25 14 China 26 48 44 10 32 58 45 36 48 77 2 37 29 Thailand 39 67 42 28 83 62 42 30 50 84 22 47 54 India 56 69 89 105 101 87 70 81 21 93 3 43 38 Indonesia 46 71 76 23 64 69 67 94 69 94 15 45 36 Vietnam 65 87 90 65 73 103 75 46 73 79 33 87 66 Philippines 75 117 105 54 92 71 88 113 71 83 36 57 108 Pakistan 118 107 115 138 121 122 93 136 70 115 30 76 75 Source: WEF 2012. Note: Numbers in bold type denote areas for improvement. For pillars, see note to table 3.3. GCI = Global Competitiveness Index. Singapore, which is ranked 2nd globally, and to Japan, which is ranked 9th, but also to Timor-Leste at 131st. Hong Kong SAR, China, is an anomaly, ranking 11th globally (and third in the region) but featured among the top three in infrastruc- ­ ture (first), goods market (third), labor market (third), and financial market (sec- ond). Improving its competitiveness will require a higher participation rate in education and improvements to its innovative capacity. Taiwan, China, is ranked 13th in 2011, a ranking consistent with earlier years. The economy ranks in the top 10 positions in just two of the pillars—education (10th) and innovation (9th). The economy has the largest number of granted patents worldwide (from the United States Patent and Trademark Office) on a per capita basis. It features an excellent education system and a high-end manufacturing sector characterized by high-quality business clusters and research and development (R&D). Two sources of weakness are rigidity in its labor markets and insufficiently developed public and private institutions. Rigidities and labor market inefficiency in Indonesia ham- per competitiveness there, with a similar story in the Philippines and Pakistan. Three areas of concern in Australia are innovation, business sophistication, and infrastructure. The increase in commodity trade in recent years has placed significant demands on its transport infrastructure, and some areas, in particular the seaports, are feeling the strain. The success of the innovation path being pursued by the Malaysian economy depends upon the quality of the education and technological adoption by business and the population in general—two areas of concern. Concerns about institutions hamper the competitiveness profile of the remaining countries in the sample. The overall quality of institutions was assessed ­ unfavorably in Korea, the Philippines, and Pakistan. Many of the institutional aspects related to business were assessed poorly in the case of China (business Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 66 National Competitiveness ethics, corporate accountability), India and Vietnam (corruption and onerous regulation), and Indonesia (corruption and bribery), while poor public institutions were noted in Thailand. Concerns about infrastructure were also dominant among the sample of countries in this region. In particular, the infrastructure in India is considered grossly inadequate for the country’s developmental and growth needs, and the improvements adopted since 2006 have been insufficient. Road transport in Vietnam, port facilities and electricity supply in Indonesia, and air and sea transport in the Philippines are considered lacking, despite improvements, while there has been no sign of improvement in the infrastructure in Pakistan. The Philippines showed one of the largest improvements in competitiveness in 2012, moving 10 places to a GCI of 75; this gain was achieved by significant improvements in many of the pillars, although as we have noted, many ­ challenges remain. The Latin America region has rebounded quickly from the 2008 global financial crisis, aided by continuing efforts to maintain macroeconomic stability, ­ high international demand for the region’s commodities, and the large internal market (WEF 2012). However, the long-term viability of the recovery is of ­ concern, particularly in light of the region’s poor institutional record. Poor insti- tutional quality is identified as a factor hampering competitiveness in 7 of the 12 countries in the regional sample (table 3.5), with República Bolivariana de Venezuela having the worst record in the global sample. Poor quality of public institutions was also identified in Panama and Belize, the first time that Belize has been included in the global sample. The quality of public and private institutions is a cause for concern in Colombia and Argentina, while the improvements made in the private institutions in Mexico and Peru are not matched by similar improvements in public institutions. The second regional cause for concern is the weak record of innovation. Greater achievement in innovation is a necessary factor for economies to move toward higher stages of development. Poor ­ Table 3.5  GCI and Pillar Rankings for Latin America and the Caribbean Region, Selected Countries, 2011–12 Pillar Country GCI rank 1 2 3 4 5 6 7 8 9 10 11 12 Chile 31 26 41 14 71 43 25 39 37 45 46 39 46 Barbados 42 18 22 126 17 25 56 35 29 29 134 41 49 Panama 49 75 38 41 79 78 46 115 27 40 85 46 72 Brazil 53 77 64 115 87 57 113 83 43 54 10 31 44 Mexico 58 103 66 39 69 72 84 114 83 63 12 56 63 Costa Rica 61 53 83 109 39 47 57 55 91 56 83 35 35 Uruguay 63 35 49 59 47 42 77 118 79 49 87 83 55 Peru 67 95 88 52 97 77 50 43 38 69 48 65 113 Colombia 68 100 85 42 78 60 99 88 68 75 32 61 57 Argentina 85 134 81 62 56 54 137 131 126 64 24 79 78 Belize 123 120 100 88 53 112 121 82 111 118 140 116 135 Venezuela, RB 124 142 117 128 84 67 142 142 132 92 41 124 126 Source: WEF 2012. Note: Numbers in bold type denote areas for improvement. For pillars, see note to table 3.3. GCI = Global Competitiveness Index. Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 National Competitiveness 67 performance on the innovation pillar was noted as a cause for concern in Chile, Peru, Mexico, and República Bolivariana de Venezuela. The potential for progress on the innovation pillar is stymied by poor rankings on the education pillar in Chile, Mexico, and Peru. Although República Bolivariana de Venezuela boasts an impressive tertiary education enrollment rate, the overall quality of education is weak and hinders the innovation potential of the country. Figure 3.4 examines the trends in the innovation pillar score and compares the Latin America region with the Organisation for Economic Co-operation and Development (OECD) average and China. The stable performance of Latin America has remained below the OECD average and has failed to converge on the more developed economies, in contrast to China. The Global Competitiveness Report calls for “a higher allocation of public and private resources toward educa- tion and training activities and R&D� (WEF 2012, 35). The competitiveness rankings in the Middle East and North Africa region sug- gest a divide between the Gulf economies and the others. The competitiveness gap may be further exacerbated by the political and social turbulence from early 2011 (WEF 2012). The rankings depicted in table 3.6 were assembled before the period termed the Arab Spring occurred, except in the case of the Arab Republic of Egypt and Tunisia. Both economies suffered in their rankings, with Egypt dropping 13 places and Tunisia 8 places. The Republic of Yemen was added to the global sample in 2011 and was ranked 138th. Qatar was ranked the most competitive in the region, on the back of strong macroeconomic performance, business sophistication, and innovation. The finan- cial sector is an area of concern noted by the business community, which is apprehensive about the soundness of the banking system and the unprotected Figure 3.4 Trends in the GCI Innovation Pillar Score, 2005–11 7 GCI innovation pillar score (1–7) 6 5 4 3 2 1 2005–06 2006–07 2007–08 2008–09 2009–10 2010–11 2011–12 GCI edition OECD China Latin America Source: WEF 2012. Note: The Latin American average includes Argentina, Barbados, Chile, Colombia, Costa Rica, Mexico, Panama, Peru, Puerto Rico, and Uruguay. Together these countries represent more than 90 percent of the regional gross domestic product. GCI = Global Competitiveness Index; OECD = Organisation for Economic Co-operation and Development. Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 68 National Competitiveness Table 3.6  GCI and Pillar Rankings for Middle East and North Africa Region, Selected Countries, 2011–12 Pillar Country GCI rank 1 2 3 4 5 6 7 8 9 10 11 12 Qatar 14 14 27 5 22 50 17 22 19 33 59 12 18 Saudi Arabia 17 12 25 12 61 36 4 50 16 43 23 17 26 Israel 22 33 53 53 36 27 33 24 10 21 51 16 6 United Arab Emirates 27 22 8 11 41 33 10 28 33 30 43 23 28 Tunisia 40 41 52 38 38 44 44 106 76 58 63 52 37 Egypt, Arab Rep. 94 74 75 132 96 107 118 141 92 95 27 72 103 Yemen, Rep. 138 140 132 130 127 138 133 129 142 139 78 134 142 Source: WEF 2012. Note: Numbers in bold type denote areas for improvement. For pillars, see note to table 3.3. GCI = Global Competitiveness Index. Table 3.7  GCI and Pillar Scores—Sub-Saharan Africa, Selected Countries, 2011 Pillar Country GCI rank 1 2 3 4 5 6 7 8 9 10 11 12 South Africa 50 46 62 55 131 73 32 95 4 76 25 38 41 Mauritius 54 40 54 79 55 68 28 67 42 61 110 44 89 Rwanda 70 21 101 61 112 119 49 8 54 109 129 84 56 Botswana 80 32 92 82 120 93 68 52 44 101 99 101 79 Namibia 83 43 58 63 114 113 71 57 36 99 120 95 92 Kenya 102 114 103 117 118 94 80 37 26 98 77 59 52 Ghana 114 61 110 139 124 109 72 79 61 113 81 99 98 Tanzania 120 85 130 129 113 131 112 73 85 126 82 104 73 Nigeria 127 111 135 121 140 114 73 70 86 106 34 64 62 Zimbabwe 132 97 127 136 123 118 124 130 104 128 133 120 117 Source: WEF 2012. Note: Numbers in bold type denote areas for improvement. GCI = Global Competitiveness Index. rights of lenders and borrowers. The institutions pillar is a cause of concern in the majority of the countries surveyed, from security concerns in Israel, to corrup- tion, government favoritism, and a judiciary that is less independent than in the past in Tunisia. Institutional rankings have also fallen in the United Arab Emirates perhaps due to the severity of the economic crisis there (WEF 2012), and the Republic of Yemen has a very weak institutional framework—public and private. Poor outcomes for health and education, pillar 4, hamper competitiveness in the Republic of Yemen and Saudi Arabia, while the quality of education and the poor representation in the math and science areas are cause for concern in Israel. Labor market rigidities and inefficiencies adversely affect competitiveness in Egypt and Saudi Arabia. Countries in Sub-Saharan Africa lag behind the rest of the world in competi- tiveness, and many areas of concern are noted in table 3.7. South Africa and Mauritius are top ranked in the region, but even here a number of pillars require further effort, especially in South Africa. Pillar 4, health and primary education, Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 National Competitiveness 69 is a main area of concern for almost all of the countries; in particular, poor health and the high rate of communicable diseases are issues in this region. The higher education pillar is also difficult for a number of countries. South Africa, for example, has a university enrollment rate of just 15 percent, and enrollment rates and overall quality are judged as insufficient in Mauritius, Rwanda, and Botswana. Tanzania, with commendable primary enrollment rates, has the lowest rates of secondary and tertiary enrollment in the world (WEF 2012). Rigidities and inef- ficiencies in the labor market constrain competitiveness in South Africa, Mauritius, Ghana, and Zimbabwe. Despite improvements in the institutions ­ pillar in South Africa, Kenya, and Nigeria, security concerns remain a factor and present an obstacle to doing business in these countries. In addition, insufficient protection of property rights is an issue in Zimbabwe and Nigeria. A number of countries are not making sufficient use of technologies to improve productivity; for example, adoption rates for information and communication technology are very low in Ghana, Nigeria, and Tanzania, with low rates of mobile phone ­ penetration in Namibia and Tanzania. The Global Competitiveness Report summarizes the 2011 country rankings by noting the complexity that characterizes national competitiveness, depen- dent as it is upon an “an array of reforms in different areas that affect the longer-term productivity of a country� (WEF 2012, 44). The rankings facili- tate the prioritizing of policy reforms as each country can identify its own strengths and weaknesses in achieving economic growth, development, and competitiveness. In the past, both the IMD’s World Competitiveness Yearbook and the WEF’s Global Competitiveness Report have been criticized for their methodology and the subjectivity of their findings (Lall 2001). Arguably the changes in methodology that these two publications have imposed throughout their life span have addressed the criticisms. Doing Business Report The World Bank/International Finance Corporation publication Doing Business also ranks countries according to the regulatory practices pertaining to small- and medium-size companies. Now in its 10th year, the Doing Business report provides an important insight to trends in regulatory reform. The main findings from the 2013 report indicate the following: • Smarter business regulation supports economic growth. • Simpler business registration promotes greater entrepreneurship and firm productivity. • Lower-cost registration improves formal employment opportunities. • An effective regulatory environment boosts trade performance. • Sound financial market infrastructure—courts, creditor and insolvency laws, and credit and collateral registries—improves access to credit (World Bank 2013). Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 70 National Competitiveness Table 3.8 Top 10 Economies for Ease of Doing Business in 2013, versus 2012 and 2011 Rankings Economy Rank 2013 Rank 2012 Rank 2011 Singapore 1 1 1 Hong Kong SAR, China 2 2 2 New Zealand 3 3 3 United States 4 4 4 Denmark 5 5 5 Norway 6 6 7 United Kingdom 7 7 6 Korea, Rep. 8 8 15 Georgia 9 16 17 Australia 10 15 11 Source: World Bank 2012, 2013. The aggregate ranking on the ease of doing business for small- and medium- size companies in 185 countries is based on indicator sets for 10 areas of the firms’ life cycles.5 The indicator sets measure and benchmark regulations in these areas: starting a business, dealing with construction permits, getting electricity, registering property, getting credit, protecting investors, paying taxes, trading across borders, enforcing contracts, and resolving insolvency (World Bank 2013). Table 3.8 presents the top 10 countries ranked by ease of doing business for 2013 and compares their ranks in the previous 2 years. There has been no change in the top five economies over the period, with Singapore in the number 1 position, followed by Hong Kong SAR, China; New Zealand; the United States; ­ and Denmark. Georgia moved from a rank of 17th in 2011 to 16th in 2012 and to 9th in 2013. Australia also made similar strides upward between 2012 and 2013, and Korea also progressed through the rankings. A number 1 ranking does not imply that the country ranked first in all of the 10 indicators. The ease of doing business is an aggregate measure, an average of the 10 indicators identified above. Figure 3.5 illustrates the dispersion around the average. Thus, any conclu- sions from table 3.8 should also take into account the dispersion across the mea- sures. For example, Singapore’s rankings range from 1 in trading across borders to 36 in registering property. The Doing Business report notes that 58 percent of the surveyed countries implemented at least one institutional or regulatory reform, making it easier to do business; 23 countries implemented reforms in three or more areas. Of these 23 countries, 10 moved ahead quite significantly through the rankings (table 3.9). The 10 indicators underlying the aggregate “ease of doing business� can be classified into two groups—a group that summarizes the strength of legal ­ institutions relevant to business regulation and a group that illustrates the complexity and costs of regulatory processes. The two sets of indicators form the ­ axes in figure 3.6. Regions in the figure’s northeast quadrant combine strong legal institutions and business-friendly regulation, whereas regions in the southwest quadrant have weak legal institutions and the least business-friendly regulation. Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 71 Average ranking 100 120 140 160 180 20 40 60 80 0 Sin New gapore Unit Zealan ed d Kore States a, Re Geo . p Aus rgia tra Taiw Icela lia Source: World Bank 2013. all topic rankings rankings highest 3 an, C nd Average of Average of Average of Mau hina rit E ius lowest 3 topic topic rankings Gerstonia Sau m any d Switi Arabia zerla n Latv d ia Net Japa herla n Slov nds Belgenia iu Bah m Arm rain enia Om an P Rwa eru Slov nda ak R Israe epu l bli Lux Mexic c emb o ourg Hun St. V gar ince Mon Belaruy nt a tene s nd t gro he G rena Fiji dine Turk s G ey Sey hana che Cze Mong lles Kyrgch Rep olia yz R ublic epu Van blic uatu It Kuwaly Mar Barb ait shall ado Isla s Solo Namnds mon ibia Islan St. K d itts Moldo s and va Gua Nevis tem Economy Uru ala g Viet uay na Jord m a Beli n z Malte N a Figure 3.5 Variation in Individual Economies’ Regulatory Environment (dispersion around average) Leb epal ano Palan Egy pt, A Guyanu rab a Rep Kirib . Ethio ati Nica pia r Indo agua n Ug esia Arg anda Ban ntinae g Phil ladesh ippin Nige es Bhu ria Ecu tan a Uk dor Tajikraine Gam istan Syria bia, n Ar the ab R Sudan Sier epubli r c Burk a Leon ina e Faso Tim or-L Iraq Com este oros Mala Buru wi A ndi Mau lgeria ritan ia T Sen ogo e Djib gal o Ang uti ola Côte Nig d’Iv er Cen o tral Eritrire Afric ea an R Chad epu blic 72 National Competitiveness Table 3.9 Ten Economies Showing Most Improvement in Ease of Doing Business, 2011–13 Economy 2011 Rank 2012 Rank 2013 Rank Poland 59 62 55 Sri Lanka 98 89 81 Ukraine 149 152 137 Uzbekistan 164 166 154 Burundi 177 169 159 Costa Rica 121 121 110 Mongolia 89 86 76 Greece 101 100 78 Serbia 88 92 86 Kazakhstan 58 47 49 Source: World Bank 2013, 2012. Note: The economies shown improved in three or more areas as measured by Doing Business. Figure 3.6 Regions Ranked by Strength of Legal Institutions and Complexity and Cost of Regulatory Processes Stronger Stronger legal institutions but Stronger legal institutions and more complex and expensive simpler and less expensive regulatory processes regulatory processes Eastern Europe 29 and Central Asia Average ranking on ease of doing business OECD high income 73 Size of bubble reflects Strength of legal institutions number of economies Latin America East Asia and Caribbean and Pacific 97 86 South Asia 121 98 Sub-Saharan Middle East Africa and North Africa 140 Weaker legal institutions and Weaker legal institutions but more complex and expensive simpler and less expensive Weaker regulatory processes regulatory processes Complex and Complexity and cost Simple and expensive of regulatory processes inexpensive Source: World Bank 2013, 4. Note: OECD = Organisation for Economic Co-operation and Development. Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 National Competitiveness 73 Figure 3.6 shows the OECD and other high-income economies in the north- east quadrant, with an average rank of doing business of 29. The Middle East and North Africa region and the East Asia and Pacific region have quite efficient regulatory processes but lag behind when it comes to the strength of their legal institutions. The Latin America and the Caribbean region is also part of this group. The Eastern Europe and Central Asia region has fairly strong legal institu- tions but more complex and expensive regulatory processes. By comparison, Sub-Saharan Africa and South Asia have weaker legal institutions and more complex and expensive regulatory processes. Business regulation reform has an impact on economic outcomes. With more years of data now available, it is possible to examine the impact of business ­ regulation reform on a number of variables, although “credibly pinning down the magnitude of this effect is more difficult� (World Bank 2013, 11). Low-income countries that implemented reforms over a 5-year period experienced an increase in their growth rate of 0.4 percentage points in the following year (World Bank 2013). Figure 3.7 shows the effect of business regulatory reform on business start-up. This area of research has increased in recent years and has shown that “simpler entry regulations encourage the creation of more new firms and new jobs in the formal sector� (World Bank 2013, 11). Figure 3.7 shows noticeable increases in business registrations after reforms have taken place. The Doing Business report provides valuable insights for policy makers and planners on the state of business regulation reform across countries. Now in its Figure 3.7 Impact of Regulatory Reform on Registration of New Firms 50 Number of newly registered 40 firms (thousands) 30 20 10 0 –3 –2 –1 0 1 2 3 Years before reform Years after reform Chile Morocco Bangladesh Sweden Kenya Rwanda Source: World Bank 2013. Note: All 6 economies implemented a reform making it easier to start a business as measured by Doing Business. The reform years vary by economy and are represented by the vertical line in the figure. For Bangladesh and Rwanda, it is 2009; for Chile, 2011; for Kenya, 2007; for Morocco, 2006; and for Sweden, 2010. Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 74 National Competitiveness 10th year, it has amassed a wealth of data that can be combined with other variables to show the economic impact of business regulatory reforms. ­ Conclusion The chapter looked at competitiveness rankings across countries. The two main sources are the annual publications from the IMD and the WEF. We looked also at the annual Doing Business report from the World Bank. The latter provides insight into business regulation and reforms. The GCI from the WEF has the largest coverage in terms of countries and indicators. Data are gathered for 12 pillars that summarize all aspects of the macroeconomy and microeconomy and that result in an overall ranking, the GCI. Switzerland occupied the top posi- tion in 2011–2012, while the United States, after 2 years in the number position, moved to number 5. 1 ­ Notes 1. The other well-known published ranking is the index prepared by the International Institute for Management Development (IMD) discussed below. There are many unpublished reports “prepared by governments, consultants, and research institutions� (Lall 2001, 1501). 2. Bakardzhieva, Ben Naceur, and Kamar (2010) cite Eyraud (2009), Bennett and Zarnic (2008), and Monfort (2008). This explanation is from the Organisation for Economic Co-operation and 3. Development, “Glossary of Statistical Terms,� http://stats.oecd.org/glossary/detail. asp?ID=2243. 4. The quotation is from Stéphane Garelli, in IMD, “IMD Announces the 2011 World Competitiveness Rankings and the Results of the ‘Government Efficiency Gap.’� Press release, May 17. 2011. http://www.imd.org/news/IMD-announces-the-2011-World- Competitiveness-Rankings-and-the-results-of-the-Government-Efficiency-Gap.cfm. 5. The Doing Business report also includes regulations on employing workers, which is not included in the 2013 aggregate ranking (World Bank 2013). References Bakardzhieva, D., S. Ben Naceur, and B. Kamar. 2010. “The Impact of Capital and Foreign Exchange Flows on the Competitiveness of Developing Countries.� Working Paper WP/10/154, International Monetary Fund, Washington, DC. Bennett, H., and Z. Zarnic. 2008. “International Competitiveness of the Mediterranean Quarter: A Heterogeneous-Product Approach.� Working Paper WP/08/240, International Monetary Fund, Washington, DC. Eyraud, L. 2009. “Madagascar: A Competitiveness and Exchange Rate Assessment.� IMF Working Paper WP/09/107, International Monetary Fund, Washington, DC. Garelli, S. 2011. “The Fundamentals and History of Competitiveness.� In IMD World Competitiveness Yearbook 2011, Appendix 3. Lausanne, Switzerland: International Institute for Management Development. Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 National Competitiveness 75 IMD (International Institute for Management Development). 2011. “IMD Announces the 2011 World Competitiveness Rankings and the Results of the ‘Government Efficiency Gap.’� Press release, May 17, 2011. http://www.imd​.org/news/IMD-announces-the- 2011-World-Competitiveness-Rankings-and-the-results-of-the-Government- Efficiency-Gap.cfm. ———. 2012. IMD World Competitiveness Yearbook 2012. Lausanne, Switzerland: IMD. IMF (International Monetary Fund). “International Financial Statistics, IFS—IMF eLibrary Data.� http://elibrary-data.imf.org/FindDataReports.aspx?d=33061&e=169393. Krugman, P. 1994. “Competitiveness—A Dangerous Obsession.� Foreign Affairs 73 (2): 28–44. http://www.pkarchive.org/global/pop.html. Lall, S. 2001. “Competitiveness Indices and Developing Countries: An Economic Evaluation of the Global Competitiveness Report.� World Development Report 29 (9): 1501–25. Monfort, B. 2008. “Chile: Trade Performance, Trade Liberalization, and Competitiveness.� Working Paper WP/08/128, International Monetary Fund, Washington, DC. WEF (World Economic Forum). 2006. The Global Competitiveness Report 2005–2006. Geneva: World Economic Forum. ———. 2009. The Global Competitiveness Report 2008–2009. Geneva: World Economic Forum. ———. 2012. The Global Competitiveness Report 2012–2013. Geneva: World Economic Forum. World Bank. 2012. Doing Business 2012: Doing Business in a More Transparent World. Washington, DC: World Bank. ———. 2013. Doing Business 2013: Smarter Regulations for Small and Medium-Size Enterprises. Washington, DC: World Bank. Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Chapter 4 Innovation Policy for Competitiveness The pursuit of competitiveness through innovation is a hallmark of the modern knowledge-driven, globalized economy, primarily in the developed world but also extending to the developing economies wishing to catch up. Innovation is the basis of sustainable economic growth and a key driver of competitiveness. It also plays an important role in promoting economic convergence, increasing welfare, creating new jobs, and destroying old ones. The effects of innovation and competitiveness translate into economic growth at the macroeconomic level. Innovation and competitiveness have thus become major objectives of national policy. An understanding of the factors affecting the innovative efforts of firms and industries and the interactions among these is critical to informing policy for innovation and competitiveness. Innovation drives competitiveness at many levels—the firm, groups of firms in an industry, the region, and the nation—with substantial scope for interac- tion across these levels. The increasing interaction among firms—the “death of distance�—arises from interfirm knowledge flows.1 Knowledge has become ­ easier to share and adopt because of globalization that has been driven by tech- ­ nological change and rapid advances in new technologies such as information and communication technologies (ICTs). Globalization and ICTs have facili- tated new forms of competition and opened new markets for innovative prod- ucts and services. Government, universities, and alliances among companies contribute to the codification of knowledge, that is, available to all, a process that increasingly benefits from ICT. Following an examination of the definition and measurement of innovation, the chapter highlights some examples of the positive relationship between inno- vation, economic growth, and competitiveness. Drawing on that discussion, it then examines the elements of an effective innovation policy. Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8   77   78 Innovation Policy for Competitiveness Innovation: Definition and Measurements The increasing importance ascribed to innovation as a key driver of growth and competitiveness has refocused attention on how this concept is defined and measured. A broader concept of innovation extends to the nature, role, and ­ determinants of innovation and moves beyond the simple definition that focused primarily on the introduction of a new product or a new process, for example. Furthermore, in terms of measuring innovation, efforts have moved beyond a focus on spending for research and development (R&D) to large-scale statistical surveys that measure how firms innovate.2 “Innovation is the implementation of a new or significantly improved product (good or service), or process, a new marketing method, or a new organizational method in business practices, workplace organization or external relations� (Organisation for Economic Co-operation and Development [OECD] 2009, 12). This definition arose from innovation surveys that have been carried out by the OECD since 1992. It has been modified twice to reflect the changing nature of innovation. Initially, the innovation surveys were confined to firms in the manu- facturing sector and described product and process innovations, but they later evolved to cover service firms and organizational and marketing innovations, resulting in the identification of four types of innovation: product, process, mar- keting, and organizational. The blueprint for the innovation surveys is the Oslo Manual,3 which has been adopted in the European Union, Japan, the Republic of Korea, Mexico, New Zealand, Norway, Switzerland, Turkey, the Russian Federation, South Africa, and most of Latin America, though not in the United States (OECD 2009, 12). The Oslo Manual contends that innovation can be “new to the firm, new to the market or new to the world� (OECD 2010, 1). An innovation new to a firm may already be in place in other firms. New to the market suggests that a new product or process is being introduced to its market for the first time, and new to the world indicates that the innovation is new to all markets and industries. This concept of innovation suggests a much broader notion than that encapsu- lated by R&D activity (see figure 4.1). Figure 4.1 illustrates that new-to-market product innovators are a feature of innovative firms that have in-house R&D but also firms that do not have in-house R&D. Furthermore, a broader interpretation of innovation suggests greater scope for policy, as innovation is not just about R&D but also extends to “organizational changes, training, testing, marketing and design� (OECD 2010, 1). The innovation surveys collect information on the firms’ inputs and outputs, tangibles and intangibles that relate to their innovative activities. ­ The surveys also capture details about the nature of innovation in each firm, concentrating on R&D, collaboration and links with other firms or public research organizations, the sources of knowledge, the reason for innovating, and the obstacles to innovation. Table 4.1 suggests some simple innovation indicators that may be derived from innovation surveys based on the Oslo Manual. Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Innovation Policy for Competitiveness 79 Figure 4.1 New-to-Market Product Innovators as a Percentage of Innovative Firms by R&D Status, 2004–06 70 60 50 Percent 40 30 20 10 0 04, act –07, n ma . (200 m a a m ) rg le n ds 01) 07) ) ) blic y ark y and gal g) –04 –07 –04 No g) stri oni ede pai Ital rwa Chi gdo giu bou rlan urin nuf 02– urin –20 06– nm u u 5 Irel Au 002 Est 006 S 002 t Sw Rep Bel Por ea, d Kin em the ma (20 De 9 (20 act d (2 9 o (2 a (2 (19 Lux ch Ne nuf ada lia Rep ite fric lan Cze xic stra an Un Can Me th A Ice Jap Au Kor Sou Innovative firms without R&D Innovative firms with in- house R&D Source: OECD 2010. Note: R&D = research and development; OECD = Organisation for Economic Co-operation and Development. Table 4.1 Simple Innovation Indicators Type of innovation Indicator(s) Technological innovation Share of firms that introduced a product innovation Share of firms that introduced a process innovation Share of firms that introduced either a product or a process innovation (“innovative firms�) Share of firms that developed in-house technological innovations (product or process) Share of firms that introduced a new-to-market product innovation Nontechnological innovation Share of firms that introduced a marketing innovation Share of firms that introduced an organizational innovation Share of firms that introduced a nontechnological innovation (marketing or organization) Inputs Total expenditure on innovation (as percentage of total turnover) Expenditure on innovation by type of expenditure (machinery acquisition, external knowledge, R&D, etc.; as percentage of total expenditure on innovation) Share of firms that performed R&D Share of firms that performed R&D on a continuous basis Outputs Share of turnover from product innovation (as percentage of turnover) Share of turnover from new-to-market product innovations (as percenatge of turnover) Key policy-relevant Share of firms that were active on international markets (outside the home country) characteristics Share of firms that cooperated with foreign partners on innovation Share of firms that cooperated on innovation activities Share of firms that cooperated with universities/higher education or government research institutes Share of firms that received public financial support for innovation Share of firms that applied for one or more patents (to protect innovations) Source: OECD 2009. Note: R&D = research and development; OECD = Organisation for Economic Co-operation and Development. Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 80 Innovation Policy for Competitiveness The first group of indicators in table 4.1 has to do with technological innova- tion and specifically with product and process innovation. Technology lies behind product and process innovations, whether these have been developed in house or outside the firm. Technological innovation captures both the product and process innovation activities in the firm. Product innovations are the final ­commercialization of innovation, while process innovations represent improvements in firms’ internal processes, as a result of either knowledge acquired through new ­ technology or in-house developments (OECD 2009). The final two indicators in ­ this group differentiate between creative activities and diffusion. The latter derives from in-house technological innovations, while the former captures inventive activity through the introduction of a new product or process. The second group of indicators, nontechnological innovations, summarizes the marketing and organizational innovations. These areas suggest a much broader concept of innovation and provide scope for policy intervention. The third group shows measures of innovation inputs, including expenditure on innovation by type of innovation that allows us to differentiate between ­ creative activities, knowledge being developed in house versus knowledge being acquired externally, and R&D expenditures. The input measures differentiate between ongoing expenditures on R&D and expenditures confined to a specific sector for a specific period (intramural). The fourth group captures output measures of innovation—those that ­ measure the output of any product innovations and those that measure the output of product innovations new to the market. ­ The final group comprises indicators that focus on internationalization and are directly relevant for policy. Participation in foreign markets and efforts to access international knowledge are both vital for maintaining and increasing competitiveness. Data from the innovation surveys provide valuable information for the design of innovation policy. The indicators reflect a focus on internationalization that is critical for competitiveness, and also on firms’ interaction with other firms, research organizations, government, and universities. Intellectual property rights are a prime focus of policy and are reflected in the patent indicator listed in the last row of table 4.1. Each indicator by itself conveys information about innovation in each ­ country—that is, one can ascertain the share of firms in each country with a product or process innovation.4 Of greater interest, particularly to policy m ­ akers, are the composite indicators that classify and distinguish different types of inno- vative firms. Composite indicators combine answers to several questions and provide a better measure of the diversity of innovation taking place within the enterprise. OECD (2009) identifies four composite indicators based on the indi- cators identified in table 4.1: • Output-based innovation modes classify innovative firms according to the novelty of their innovations and whether innovation was conducted in house ­ or mainly by others.5 Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Innovation Policy for Competitiveness 81 • Innovation status classifies firms according to the inventiveness of their innovation activities and whether they engage in collaboration.6 ­ • Technological and nontechnological innovation examines the combination of product-process innovation with organizational and marketing innovations.7 • Dual innovators identify firms that are active in both goods and service innovations.8 These composite indicators can be used for benchmarking purposes, although one needs to be aware of the limitations of the data. The surveys are not fully harmonized across the participating countries and the responses may be subject to interpretation differences. Moreover, further work needs to be done to ascer- tain the statistical significance of the differences in data across countries. However, indicators derived from innovation surveys have not featured that strongly in policy. R&D indicators are still the most widely used. OECD (2009) cites a number of reasons why this continues to be the case, including the ­ reliability of R&D measures, their role in national science and technology policies, their wide acceptance as an indicator of innovation, and the lack of international comparability across the innovation surveys. That said, the composite indicators derived from the innovation surveys provide a detailed picture of economy-wide innovation activities within the firm and significantly broaden the scope for policy to assist innovative efforts at the level of the firm and the economy. Innovation, Growth, and Competitiveness World Bank (2010) states that innovation has always been a key part of eco- nomic and social development, and it refers to four key effects of innovation: “It is the main source of economic growth, it helps improve productivity, it is the foundation of competitiveness, and it improves welfare� (World Bank 2010, 6). Petrakos, Arvanitidis, and Pavleas (2007) refer to the strong association between innovation and economic growth, citing the work of Fagerberg (1987), Lichtenberg (1992), and Ulku (2004) in this regard. Including innovation as a regressor in the empirical models of growth improves the explanatory power of these models. Innovation increases productivity and growth through improved technology arising from new products and processes. Moreover, endogenous growth theory maintains that investment in innovation, human capital, and knowledge results in economic growth, as expressed by Q = f(K, L, R&D, HC), where Q = output, K = capital, L = labor, R&D proxies for innovation, and HC proxies for human capital. The endogenous growth equation proves a better approximation for growth in developed economies. Developing economies “do not do much R&D� and produce new products and processes by importing the knowledge from devel- oped economies (World Bank 2010, 41). Growth equations in developing Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 82 Innovation Policy for Competitiveness Table 4.2 Extent of Economic Growth beyond Growth Predicted by Rates of Capital Accumulation, Selected Economies, 1960–89 Actual minus predicated growth Economy Investment/GDP (%) rate of GDP per capita Algeria 35.0 −0.026 Gabon 40.0 −0.030 Greece 24.2 0.008 Hong Kong SAR, China 27.3 0.031 Ireland 22.2 0.011 Jamaica 25.0 −0.037 Korea, Rep. 24.9 0.032 Panama 24.0 0.002 Portugal 23.7 −0.002 Singapore 34.3 0.017 Taiwan, China 25.0 0.047 Source: Nelson and Pack 1999; cited in Cantwell 2005. Note: GDP = gross domestic product. economies should therefore include imports of capital goods and components as well as foreign direct investment (FDI). Knowledge affects total factor productiv- ity (TFP), that is, the residual for the growth in output not explained by the growth in factor inputs. Part of capital accumulation derives from the innovative efforts of the firm. Nelson and Pack (1999)9 show that countries that sustained high rates of capital accumulation, that is, investment in gross domestic product (GDP) of 20 percent or more, achieved growth rates of GDP in excess of what might have been expected from the rates of capital accumulation alone. Table 4.2 shows 11 coun- tries that had high rates of investment to GDP over the 1960 to 1989 period. The right-hand column shows the residuals from a regression of 101 countries of GDP per capita on investment share, the initial level of GDP per capita in 1960 (a proxy for a catching-up effect), the growth of population (to capture the avail- ability of labor supplies), and the relevant cohort of population educated to at least secondary school standards. The results show that the East Asian countries and economies—Hong Kong SAR, China; Korea; Singapore; and Taiwan, China—achieved growth rates in excess of what might have been expected, given their favorable rates of ­ capital accumulation. Cantwell (2005, 7) suggests that what was different in these economies “was their greater ability to innovate, to upgrade and restructure their indigenous industries, and to learn and absorb more effectively from foreign technologies.� In addition to these studies, the growth literature that focuses on cross-country convergence, or catching up, identifies innovation as a key factor in explaining dif- ferences across countries. Figure 4.2 shows how innovation contributed to growth in two countries—Ghana and Korea—from 1960 to 2005. The disparity in growth performance was primarily due to TFP or knowledge accumulation (figure 4.2).10 Roughly two-thirds of the difference in output between the two countries was Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Innovation Policy for Competitiveness 83 Figure 4.2 Contribution of Innovation to Growth in Ghana and the Republic of Korea, 1960–2005 14,000 12,000 Real GDP per capita (2000 US$) 10,000 Di erence in 8,000 output due to TFP growth or knowledge 6,000 accumulation in Korea, Rep. 4,000 Di erence in 2,000 output due to growth in labor and capital in Korea, Rep. 0 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 Korea, Rep. Ghana Source: World Bank 2010. Note: GDP = gross domestic product; TFP = total factor productivity. Table 4.3 Conventional Breakdown of Sources of Growth, 1970–2000 Average annual growth of GDP Average annual growth of Average annual growth of Indicator per worker capital-labor ratio total factor productivity Low income 0.17 0.25 −0.07 Lower-middle income 1.01 0.61 0.40 Upper-middle income 0.99 0.59 0.40 New tigers 3.79 1.70 2.09 Old tigers 4.89 2.37 2.52 High income 1.95 1.00 0.95 Source: Hulten and Isaksson 2007. Note: Old tigers refers to Hong Kong SAR, China; the Republic of Korea; Singapore; and Taiwan, China. New tigers refers to China, India, Indonesia, Malaysia, and Thailand. GDP = gross domestic product. attributed to innovation or the technology-related improvements pursued by Korea, with the remainder arising from the growth in capital and labor. Looking at the World Bank classification by income group, Hulten and Isaksson (2007) noted that capital deepening appears to be more important as an explanation of growth in countries at lower levels of income, while TFP is more important for those countries that grew fastest (tiger economies; see table 4.3). Capital deepening, or the average annual growth rate in the capital-labor ratio, accounted for a greater proportion of output or the average annual growth of GDP per worker in the lower-income economies. Furthermore, TFP has been credited with accounting for the difference in levels of development across countries. Isaksson (2007), in reviewing the lit- erature on TFP, highlighted the factors captured in TFP. He identified these Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 84 Innovation Policy for Competitiveness Table 4.4 Level of Productivity in Countries of Various Incomes, 1970–2000 Indicator Log of GDP per worker Log of capital-labor ratio Log of TFP Low income 7.76 2.61 5.15 Lower-middle income 9.08 3.14 5.93 Upper-middle income 9.76 3.45 6.31 New tigers 8.09 2.78 5.31 Old tigers 9.83 3.48 6.35 High income 10.57 3.81 6.77 Source: Hulten and Isaksson 2007. Note: Old Tigers refers to Hong Kong SAR, China; the Republic of Korea; Singapore; and Taiwan, China. New tigers refers to China, India, Indonesia, Malaysia, and Thailand. GDP = gross domestic product; TFP = total factor productivity. Table 4.5  Decomposition of the Predicted Growth in National Market Shares from an Estimated Empirical Model of Cross-Country Competitiveness, 1961–73 Country Japan United Kingdom United States Growth in technological capabilities 66.9 6.9 −0.6 Rise in relative unit labor costs −0.9 0.8 1.6 Initial technological capabilities (catch-up) 20.9 15.9 7.3 Investment as a share of GDP, and growth of world demand 16.5 −39.8 −38.2 Total growth in market share (predicted by model) 103.3 −16.2 −29.8 Source: Fagerberg 1988. Note: GDP = gross domestic product. ­actors as competition, the rule of law, enforcement of contracts, R&D, and f capital accumulation. Using productivity data from countries of various incomes over the period 1970–2000, Hulten and Isaksson (2007) show that the share of TFP growth in output per worker, that is, the log of TFP, is always greater than that of capital deepening, the log of capital-labor ratio for all countries (see table 4.4). Fagerberg (1987, 1988) suggests that innovation is one of the key factors affecting differential growth rates among countries. Table 4.5 examines the results from a model of international competitiveness that decomposes the pre- dicted growth in national market shares into four elements for Japan, the United Kingdom, and the United States. What stands out in this table is the growth in indigenous technological capabilities in Japan, which accounts for the largest share in its competitiveness over the 1961–73 period. By contrast, the loss of world trade shares by the United Kingdom and the United States may be attrib- uted to weak capital accumulation (investment as a share of GDP and growth of world demand) arising from the high share of military spending in these two economies. The link between innovation and competitiveness focuses on longer-term tech- nological competitiveness. Innovation has played and continues to play a key role in changing or transforming the way in which the world’s technological systems are structured. Figure 4.3 charts the four poles around which technological sys- tems are structured—energy, matter, life, and time—and also charts how these poles are affected by the transformations taking place in the world’s technological Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Innovation Policy for Competitiveness 85 Figure 4.3 Major Technical Systems from the Middle Ages through the Present Middle Ages Industrial Revolution Present Optronics, computers and communication Time scale Taylorism Converging Polymers, Electricity, Materials Energy Steel, Combustion cement nanotechnology supraconductivity Microbiology Human-biosphere relations Biotechnology, ecological equilibrium Source: World Bank 2010. systems. Change depends on how quickly humans adapt to the new technologies. The world moved over time from the agricultural revolution in the Middle Ages through the Industrial Revolution to the “cognitive revolution� today (World Bank 2010). This cognitive revolution is taking the form of a knowledge economy char- acterized by rapid developments in science and technology. These developments call for new skills and higher levels of education in order to exploit the innovation potential that the advances are unearthing. Investment in this knowledge, whether through education or through R&D, is critical for economic progress and prosperity. Access to knowledge is key for innovation. ­ Longer-term technological competitiveness supposes that the faster growth of output and exports achieved by innovation, along with new lines of value cre- ation, drives up the domestic currency, reflecting an increase in international competitiveness. This type of technological competitiveness, classified as neo- Schumpeterian,11 pertains when countries (or firms) that are most successful in innovation achieve a sustainable increase in the share of world trade (or in the share of the relevant world market) and also expand the overall magnitude of world trade and the world market (Cantwell 2005). Traistaru-Siedschlag et al. (2006) identify a number of researchers who have shown that innovating firms are more likely to export and have a higher share of exports than those firms that do not engage in R&D. Traistaru-Siedschlag et al. (2006, 7) cite the work of “Kumar and Siddhartan (1994) for Indian firms, Braunerhjelm et al. (1996) for Swedish manufacturers, Nassimbeni (2001) and Basile (2001) for Italian plants, Özçelik and Taymaz (2004) in the case of Turkey.� Although it remains an under- researched area, Traistaru-Siedschlag et al. (2006) note other studies showing the positive effect of exporting on innovation.12 The Commission on Growth and Development (2008) identified 13 econo- mies worldwide that had achieved at least 25 years of consecutive growth above 7 percent.13 One of the five main elements identified for this successful growth pattern was their participation in the global economy and the importance of innovation and technology in their economic development. “To put it very simply, they imported what the rest of the world knew, and exported what it ­ wanted� (Commission on Growth and Development 2008, 22). Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 86 Innovation Policy for Competitiveness The ability of an economy to innovate by upgrading and restructuring its indus- tries based on learning from foreign technologies has focused attention on the firm and the industry. The scope for interaction between the firm and the industry feeds into the competitiveness of the local area, the region, and the nation. The scope for interaction between the firm and its location has concentrated the research effort along two similar paths. At one level, the research has focused on the geography of this interaction, and academic attention has concentrated on “innovative regions and milieux,� “high-tech areas,� “clusters of knowledge based industries,� and “knowledge spillovers.�14 At a second and complementary level, the research reflects the fact that innovation requires a range of complementary activities that include organizational changes, training, testing, and marketing. It also takes into consideration the fact that innovation is a highly collaborative endeavor requiring input from many participants and thus best undertaken where these stakeholders converge. Here the academic and policy work has adopted a “system of innovation� 15approach that concentrates on the entire process of inno- vation rather than just one element, such as the supply of technology. The innova- tion system approach informs policy for innovation, as discussed in the following sections. Policies for Innovation A new global context heralds an unprecedented role for innovation. The follow- ing section will examine both the role for innovation in the wake of the 2008 financial crisis and its role in meeting the economic and social challenges of the modern age against the background of increasing global interdependence. The share of world merchandise trade to world GDP increased from 32 percent in 1990 to 51 percent in 2011, while trade in services increased from 7.5 percent of GDP in 1990 to 11.4 percent in 2011.16 Figure 4.4 shows that cross-border capital flows increased from less than $1 trillion in 1990 to $11 trillion in 2007. Furthermore, FDI inward stock as a percentage of GDP increased from 9.6 ­percent in 1990 to 30.3 percent in 2010. These global developments provide a rich context for the development of policy to promote innovation. The legacy of the 2008 financial crisis was manifest in the weak and sluggish recovery in the developed economies, in contrast with the stronger growth in the emerging markets (see figure 4.5). Recovery from the recent financial crisis depends upon new sources of economic growth. Many of the traditional sources of growth are declining in importance. Stagnating populations in the developed economies have implications for labor inputs in long-run economic growth, while physical capital inputs face diminishing returns and may be insufficient to strengthen long-term growth (OECD 2010). Technology can be a means of but- tressing growth and achieving sustainable growth over the longer term. Developed economies work at or near the technology frontier, and developing economies have the potential to catch up by acquiring existing knowledge. Most of this knowledge is already in the public domain and can be acquired through formal modes as well as through informal channels such as copying and reverse Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Innovation Policy for Competitiveness 87 Figure 4.4 Cross-Border Capital Inflows, 1985–2009 12 24 10 20 Percentage of world GDP 8 16 US dollars, trillion 6 12 4 8 2 4 0 0 –2 –4 1985 1990 1995 2000 2005 Year Gross capital inflows Total in percentage of world GDP (RHS) Emerging markets Advanced economies Low-income countries Source: IMF 2010. Note: GDP = gross domestic product. engineering. The success of these efforts depends upon the developing economy having its own capabilities to acquire, use, and create knowledge. There is a clear role for policy in meeting these challenges. Innovation provides the foundation for new industries, businesses, and jobs by improving competitiveness and economic growth. It is already an important contributor to growth in some countries. Figure 4.6 shows the contribution made to labor productivity growth by innovation. First, investment in intangible assets, that is, investment in R&D, software, databases, and skills, accounts for just as much as investment in physical capital in the majority of countries. Investment in intangible assets and multifactor productivity (MFP) accounted for between two-thirds and three-quarters of labor productivity growth between 1995 and 2006 in Austria, Finland, Sweden, the United Kingdom, and the United States. Market failure, or indeed the absence of markets, acts as a constraint on the development of innovations. There is increasing pressure to deal with the various social challenges such as climate change, health, food security, and access to water. These challenges are global in nature and cannot be dealt with by one single country. The challenges require commitment and coordination at an inter- national level. The pricing of externalities, such as carbon emissions, is an impor- tant catalyst for innovation. Tax policies can also help by providing a signal and fostering a market for innovation. Removing subsidies for environmentally harm- ful substances can also help (OECD 2010). The distribution and use of existing technologies are key for economic and social challenges. They are particularly important for developing economies, where simple technologies can significantly increase welfare. World Bank (2010) Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 88 Innovation Policy for Competitiveness Figure 4.5  Global Growth in Real GDP, 2000–12 (quarterly change from prior year) a. Emerging and advanced economies b. United States, Japan, and euro area 10 Emerging 6 United States economies 8 3 6 World Real GDP growth (percent) Real GDP growth (percent) 0 4 2 Japan Euro –3 area 0 –6 –2 Advanced economies –9 –4 –6 –12 2000 2002 2004 2006 2008 2010 2012 2000 2002 2004 2006 2008 2010 2012 c. Asia d. Emerging Europe, Brazil, and Latin America 16 China 16 India 12 12 Real GDP growth (percent) Real GDP growth (percent) Emerging Europe 8 8 4 4 ASEAN-4 Brazil 0 0 NIEs Latin America –4 –4 –8 –8 2000 2002 2004 2006 2008 2010 2012 2000 2002 2004 2006 2008 2010 2012 Source: IMF 2011. Note: IMF = International Monetary Fund; ASEAN = Association of Southeast Asian Nations; NIEs = newly industrialized economies. In panel a, emerging economies are Argentina, Brazil, Bulgaria, Chile, China, Colombia, Hungary, India, Indonesia, Latvia, Lithuania, Malaysia, Mexico, Peru, Philippines, Poland, Russian Federation, South Africa, Thailand, Turkey, and Venezuela, RB; advanced economies are those that report quarterly data—Australia; Canada; Czech Republic; Denmark; Euro area; Hong Kong SAR, China; Israel; Japan; Korea, Rep.; New Zealand; Norway; Singapore; Sweden; Switzerland; Taiwan, China. In panel c, ASEAN countries are Indonesia, Malaysia, Philippines, and Thailand. In panel d, emerging European countries are Bulgaria, Hungary, Latvia, Lithuania, and Poland; Latin American countries are Argentina, Brazil, Chile, Colombia, Mexico, Peru, and Venezuela, RB. Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Innovation Policy for Competitiveness 89 Figure 4.6 Innovation Accounts for a Large Share of Labor Productivity Growth, Percentage Contributions, 1995–2006 7 6 5 4 Percent 3 2 1 0 ic ic en ce m lia es ria n ce y n ly k d an ar pa ai bl bl Ita an do at ee an ra st ed Sp nm pu pu rm Ja St st Au nl ng Gr Fr Sw Au Fi Re Re Ge De d Ki ite ak h d Un ec ite ov Cz Un Sl Labor productivity growth Contribution from tangible capital Contribution from intangible capital Source: Wyckoff 2010. Table 4.6 Percentage of Rural and Urban Population with Access to Clean Water, 1990 and 2004 Total Rural Urban Location 1990 2004 1990 2004 1990 2004 Region East Asia and Pacific 71.8 78.5 61.4 69.8 97.3 91.9 Europe and Central Asia 91.7 91.7 83.4 79.8 97.0 98.7 Latin America and the Caribbean 82.8 91.0 50.0 73.0 92.6 96.0 Middle East and North Africa 87.5 89.5 78.9 80.8 96.1 96.3 South Asia 70.6 64.4 64.9 81.3 88.6 93.6 Sub-Saharan Africa 48.9 56.2 36.1 42.4 81.9 80.1 World 76.4 82.7 63.2 72.2 95.2 94.5 Countries High income 99.8 99.5 99.1 98.5 99.8 99.8 Low and middle income 72.1 79.9 60.6 70.5 93.3 92.8 Low income 64.3 75.0 56.7 69.4 87.0 88.1 Source: World Bank 2010. highlights three areas in which existing technology can improve welfare—­ vaccines, access to water, and sanitation. Table 4.6 shows the percentage of rural and urban population with access to clean water in 1990 and 2004. The technol- ogy to provide clean water is relatively simple and has benefitted from improve- ments in the past decade, but roughly 20 percent of the population of low- and middle-income countries continue to lack access to clean water. The disparity is even more pronounced between rural and urban dwellers—roughly 30 percent of rural dwellers do not have access to clean water, compared to 7 percent of urban dwellers. Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 90 Innovation Policy for Competitiveness Table 4.7 Pace of Dissemination of Major Technologies, 1748–2000 Period in which technology was initially discovered Technology 1748–1900 1900–50 1950–75 1975–2000 Number Transportation 21 Shipping (steam) 83 57 Shipping (steam motor) 180 93 Rail (passenger) 126 99 Rail (freight) 124 153 Vehicle (private) 96 123 Vehicle (commercial) 63 109 Aviation (passenger) 60 103 Aviation (freight) 60 Communications Telegram 91 77 Telephone 99 156 Radio 69 154 Television 59 156 Cable TV 50 98 Personal computer 24 134 Internet use 23 151 Mobile phone 16 150 Manufacturing Spindle (ring) 111 50 Steel (OHF) 125 50 Electrification 78 155 Steel (EAF) 92 91 Synthetic textiles 36 75 Medical (OECD only) Cataract surgery 251 19 X-ray 93 27 Dialysis 33 29 Mammography 33 18 Liver transplant 28 29 Heart transplant 28 27 CAT scan 18 29 Lithotripter 15 26 Average (excluding medical) 106.9 60.9 23.5 16.0 Average (including medical) 118.9 61.3 25.7 15.5 Source: World Bank 2010. Note: The table indicates the number of years elapsed between the time the technology was invented and the time it had reached 80 percent of reporting countries. CAT = computer-assisted tomography; EAF = electric arc furnace; OECD = Organisation for Economic Co-operation and Development; OHF = open hearth furnace. Table 4.7 examines the speed at which major global technologies were imple- mented. Two key trends emerge. First, the speed at which the major technologies were disseminated over countries has increased over time. For example, for over 80 percent of the countries surveyed, key innovations developed between 1748 and 1900 took slightly more than 100 years on average to disseminate; those Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Innovation Policy for Competitiveness 91 developed between 1900 and 1950 took an average of 61 years; those between 1950 and 1975 took an average of 24 years; and those between 1970 and 2000 took an average of 16 years (World Bank 2010). The speed at which innovations are being adapted has increased over time. The importance and rate of adaptation of innovations and inventions are the subjects of recent literature that examines the extent to which innovation and new technology are driving economic growth.17 The second trend is that while there has been improvement in the distribution of technology to the capital and major cities of developing countries, the rate of dissemination within these countries is slow (World Bank 2010). This was evi- dent from table 4.6, which showed that the rate of rural access to clean water was markedly different from the urban access rate. Of concern is that access to basic technologies such as electricity and paved roads also remains outside of the reach of many. How Can Government Help? The discussion so far has highlighted a broad view of innovation that argues for well-specified policies across a range of areas. Innovation depends on a favorable economic environment that encompasses education, governance, and infrastruc- ture. These areas may be problematic for developing countries, “but experience shows not only that proactive innovation policies are possible and effective but also that they help create an environment for broader reforms� (World Bank 2010, 2). Government has a key role to play. Governments bring about the regu- lations and markets that enable firms to innovate. They can launch programs in education and training, in product and labor markets, in public research institu- tions, and in policies for networking and knowledge exchange between firms and markets. “Pro-growth tax reforms can also help to strengthen growth and innova- tion� (OECD 2010, 2). Government plays both a direct and an indirect role when it comes to ­ fostering innovation. For example, in the area of technology promotion, government can play a direct role through supporting innovative efforts for ­ space exploration or defense. Indirectly, government can create a favorable climate for innovation through enactment of supporting laws. Macroeconomic, business, and governance conditions all determine the capabilities for innova- tion in each country. At the same time, each country will have its own needs for innovation. Policy changes in these areas can bridge the gap between capa- bilities and needs. Moreover, specific innovation policy of and by itself can be an important trigger for change in areas that may be lacking. Figure 4.7 presents the various factors that influence a developing country’s capabilities in innovation. Developed countries are assumed to be at the tech- nological frontier and through various links—trade, FDI, and diaspora and other networks—affect positively the capacity for innovation in the developing countries. These transmission channels can be enhanced by policies to create ­ competencies and build an innovation-friendly business climate. These policies Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 92 Innovation Policy for Competitiveness Figure 4.7  Determinants of Technology Upgrading in Developing Countries Technological frontier Transmission Foreign Diaspora channels Trade direct and other investment networks Governance and the business climate absorptive capacity Policies to Technological Basic technological literacy - Create competencies - Build infrastructure - Foster an innovation- Finance of innovative firms friendly business climate Proactive policies Dynamic Technological absorption Spillover Returns to effects effects scale magnify technology transfer Domestic technological achievement Source: World Bank 2010. target the technological absorptive capacity of the developing economy, and through spillover effects and returns to scale they result in technology transfer that increases domestic technological achievement in the developing economy. Figure 4.7 illustrates the involvement of both the private and public sector in what the World Bank (2010, 8) has termed the “innovation system.� In this system, public and private organizations work together to foster the t ­ ­ echnological, commercial, and financial competencies and inputs required for innovation. The government can facilitate the innovation system in a number of ways: • supporting innovators through appropriate incentives and mechanisms; • removing obstacles to innovative initiatives; • establishing responsive research structures; and • forming a creative and receptive population through appropriate educational systems (World Bank 2010, 8). The World Bank (2010) illustrates the role of the government in promoting innovation in figure 4.8. Innovation policy requires input from many different areas, including education, trade, investment, decentralization, and finance. The approach suggests a holistic role for government involving many depart- ments in order to achieve a “fundamentally horizontal and interdepartmental innovation policy� (World Bank 2010, 9). Moreover, subnational governments have a key role to play, as innovation takes place in firms and enterprises at the local level. Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Innovation Policy for Competitiveness 93 Figure 4.8  Growing Innovation: The Government as Gardener Watering (finance, support to innovators) Removing weeds (competition, deregulation) Nurturing soil (research, information) Preparing the ground (education) Source: World Bank 2010. The OECD (2010) proposes a similar innovation strategy that is built upon five priorities for government: • empowering people to innovate; • unleashing innovation in firms; • creating and applying knowledge; • applying innovation to address global and social challenges; • improving the governance and measurement of policies for innovation. The principles underlying these priorities are shown in table 4.8, where we also include the strategies under the “gardening� approach suggested by the World Bank (2010). Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 94 Innovation Policy for Competitiveness Table 4.8 OECD and World Bank Policy Principles for Innovation OECD World Bank Priority Principles Priority Principles Empowering people to Education Preparing the ground Education innovate Training Knowledge transmission Unleashing innovation Foster entrepreneurship Removing weeds Competition in firms Enhance access to finance Deregulation Build foundations for innovation with sound framework conditions Foster markets for innovative goods, services, and processes Creating and applying Foster strong and effective public research Nurturing soil Research knowledge Invest in a knowledge-supporting infrastructure Information Foster efficient knowledge flows, networks, and markets Unleash innovation in the public sector Applying innovation to Foster international cooperation Watering Finance address global and Tackle key challenges through innovation: climate Support to social challenges change, health, and food security innovators Bridge the gap in economic development through innovation Improving the Link science, technology, and innovation policies to governance and economic growth measurement of Develop data infrastructure to measure the policies for innovation determinants and impact of innovation Account for the role of innovation in public sector Promote new statistical methods and interdisciplinary approaches to data collection Promote measurement of innovation for social goals and social impacts of innovation Source: OECD 2010; World Bank 2010. Note: OECD = Organisation for Economic Co-operation and Development. The elements of innovation policy discussed by recent OECD and the World Bank publications on innovation strategy—and encapsulated in table 4.8—­ suggest a “whole of government� approach encompassing cross-department cooperation and extending to local and regional as well as national governments. The differing country contexts in terms of needs and capabilities suggest the fallacy of a one- size-fits-all policy. Indeed, the World Bank (2010, 48) notes that “innovation agendas in the developed and in the developing world will differ significantly.�18 The economic, cultural, and social settings particular to each country suggest a need to understand the “specific motivations and behavior as people innovate, create new things, adapt their institutions, and manage their businesses� (World Bank 2010, 69). Berdegué (2005) and Gupta (2007) stress the importance of aligning innovation policies with social settings. Empowering People to Innovate or Preparing the Ground Both education and training systems are key when it comes to “empowering people to innovate� or “preparing the ground.� The OECD argues for flexible Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Innovation Policy for Competitiveness 95 systems of education and training that provide people with the foundation for learning and developing broad ranges of skills and that make it possible for them to upgrade their skills and adapt to changing market conditions. The OECD study identifies universities and vocational training colleges as “essential nodes in the innovation system,� which produce students with the capacity for lifelong learning as well as being the “anchor for clusters of innovative activity� (OECD 2010, 10). On-the-job training has an important role to play in lifelong learning. It is particularly relevant in developing economies, where it is respon- sible for “more skills development than all other types of training combined� (World Bank 2010, 16). Skills such as “critical thinking, creativity, communica- tion, user orientation and team work in addition to domain-specific and linguis- tic skills� are identified by the OECD study as being critical for innovation (OECD 2010, 9). The World Bank study calls for skills development in the informal economy, “which can represent 30 percent or more of nonfarm employment in a number of developing economies� (World Bank 2010, 16). Vocational training colleges should work more closely with businesses through engaging workers and employers in curriculum development. Entrepreneurship education should be part of the curriculum, and female participation in science and technology sub- jects should be encouraged. Labor market policy should provide incentives for women to enter the workplace, such as availability of child care and tax and benefit systems, and should support workplace practices that favor women’s participation in the labor market (OECD 2010). Investing in a well-educated workforce may be a double-edged sword for developing economies, since it poses the risk of a brain drain. Therefore, policy in these economies should strive for a “brain circulation� that would connect talented migrants with their home coun- try as “creators of enterprises, openers of new markets, sources of venture capital, or facilitators of institutional reforms� (World Bank 2010, 17). Unleashing Innovation or Removing the Weeds “Unleashing innovations� or “removing the weeds� calls for policies that will put in place framework conditions to support competition that fosters innovation. Under this rubric, policies would aim to mobilize funding for firms through well- functioning financial markets and ease the access to finance for new firms. In addition, policies would aim to foster open markets and competitive markets predicated on healthy risk taking and creative activity. Policies would focus on entrepreneurship and on small- and medium-size firms, particularly new firms. The World Bank study notes that many areas of government will be involved in establishing framework conditions. These tasks are “particularly necessary, but difficult, in developing country contexts� (World Bank 2010, 13). Well-functioning financial markets, which provide access to finance for new and small innovative firms with the necessary early-stage financing and ­networks for business angels and venture capital, are critical for promoting innovation and a prime area for policy. Access to finance can be a constraint for many firms, particularly for the innovative firm that may require a long-term horizon. Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 96 Innovation Policy for Competitiveness Seed capital and start-up capital by business angel funds provide more than just financing; they also offer advice and experience. Government can encourage networks of business angels. Government can play a further role in ensuring that “information on intellectual assets is consistent and comparable over time and across companies� (OECD 2010, 13). This step will help investors to make better decisions about investment opportunities. ­ Building sound framework conditions targets the overall environment for business creation and development and has become increasingly important in the global environment. Stable macroeconomic policies help to reduce uncertainty. Innovation would also benefit from open international markets that facilitated the exchange and spread of knowledge. Government policies can help speed up the knowledge-adoption process by removing tariffs or other restrictions on acquiring global knowledge. It could subsidize early adopters of innovation and provide support services by launching information and publicity campaigns and provide demonstration and extension services. Furthermore, it could introduce regulations requiring adoption of global innovation in certain areas, for example reducing pollution or carbon use. Finally, it could invest in human capital in order to improve the absorption capacity for new technologies. Microeconomic poli- cies help foster open and competitive markets that are critical for innovation. Taxation policies affect investment decisions at the household and firm level and can thus affect innovation. Policies to foster entrepreneurship may take a variety of forms that target areas of difficulty for entrepreneurs. For example, policies may focus upon reducing barriers to firm entry and exit. New firms or young firms are critical for innovation, bringing new ideas to market and taking advantage of existing ­ technology or other opportunities that may have been neglected by older, more established firms. OECD (2010) stresses the importance of new firms for bring- ing new ideas to markets, as illustrated by their patent filings (see figure 4.9). Figure 4.9 Patenting Activity of Young Firms, 2005–07 33.5 13.5 4.5 4.2 3.0 2.8 1.6 1.6 0.9 0.7 0.7 0.4 0.3 25 20 Share of countries in PCT filed by firms (%) 15 Percent 10 5 0 es y e s om en d ly m ria n ay k nd an ar c ai an Ita iu at an rw st ed Sp nm d rm la nl lg St Au ng Fr No Sw er Fi Be Ge De d th Ki ite Ne d Un te i Un Share of patents filed by young firms (%) Source: OECD 2010. Note: Young firms are those under five years old. Data are for Patent Cooperation Treaty patents. OECD = Organisation for Economic Co-operation and Development; PCT = Patent Cooperation Treaty. Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Innovation Policy for Competitiveness 97 Policy may also focus on the tax system, another area that entrepreneurs identify as adversely affecting their decision to become self-employed. Policies may target a more neutral tax treatment and thus help to foster entrepreneurship. ­ The high rate of failure endemic to new enterprises need not be such a bad thing in an environment that efficiently allows for the reallocation of resources from declining to innovative firms. Policies should foster open and competitive markets. The World Bank’s Doing Business surveys can identify obstacles to ­ innovation. Governments have a role to play in fostering markets for innovation. OECD (2010, 15) identifies the following three areas in which government can play an active role: “getting prices right; opening markets for competition; and devising innovation-inducing standards and regulations.� Creating and Applying Knowledge or Nurturing the Soil The third priority identified by the OECD, and one that loosely coincides with the World Bank’s innovation strategy of “nurturing the soil,� is creating and ­ applying knowledge. Policies here would target knowledge development, sharing, and transmission from the creation and governance of public research institutes to the fostering of networks and markets that enable knowledge sharing and diffusion. Also under this priority are policies targeted at establishing an effective ­ system of intellectual property rights and those that ensure coherence among multilevel sources of funding for R&D. In addition, policies should promote innovation in the public sector to bring about an enhancement of public service delivery and an improvement in efficiency. The World Bank (2010) advocates that developing countries invest in their own capability to acquire, use, and create knowledge. Developing countries can dramatically improve their position by acquiring existing knowledge. Most of this knowledge is in the public domain already. Other knowledge can be acquired through formal means, and some may be acquired through informal copying and reverse engineering. The latter is the most important source of technological catch up, especially in the case of rapidly growing economies such as the United States in the 1800s; Japan, Korea, and Taiwan, China, in the 1900s; and China and India now. Applying Knowledge to Address Global and Social Challenges or Watering the Soil Applying innovation to address global and social challenges is another area for policy. Under this heading, we include challenges from climate change, global health, and food security. Market failure or the absence of a market for many of these challenges suggests a role for policy in helping to achieve sustainable ­ solutions. Moreover, the scale and scope of these challenges indicates a role for government involvement at an international level. “Proven co-operation strategies include joint investment in basic and pre-competitive research; mapping of R&D needs; technology transfer initiatives; ­ Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 98 Innovation Policy for Competitiveness and scholarships and fellowships for international researchers and students� (OECD 2010, 20). An example of a multilateral effort is the United Nations Clean Development Mechanism (CDM). Academic partnerships—cross-country teaming of higher education establishments in science and ­ technology—also facilitate technology transfer with implications for local innovation systems (OECD 2010). The solutions to meet global challenges are long term in nature and thus call for policy that is predictable and provides long-term incentives. Policies should promote private sector involvement by being flexible, and they should focus, where possible, on direct solutions to global problems. For example, “in addressing climate change, a tax on carbon will be more effective for inducing an optimal innovation path than a tax on fuel or electricity use� (OECD 2010, 22). Innovation can play a significant role in bridging the gap in economic develop- ment at the developing economy level. Poor framework conditions and low social and human capital in developing economies present a challenge for innovation. The domestic economy needs to be very good at tapping global knowledge through all its forms—trade, technology, licensing, FDI, and participation in global value chains; foreign education and training; participation in trade fairs; global research networks, technical publications and databases; and copying and reverse engineering. Using catch-up strategies is easier than pushing back the global frontier, so the domestic economy must exploit these catch-up opportunities. Finally, the domestic economy needs to build up its capability not just for education in general but also for entrepreneurship, business management, and science and engineering. It also requires institutions and mechanisms capable of providing re-skilling and up-skilling to keep abreast of new technology and busi- ness needs. Improving the Governance and Measurement of Policies for Innovation Finally, policy coherence, good governance of policies, and an improved measure- ment framework for innovation are essential in developing innovation policies. A “whole of government� approach is needed that includes medium- and long-term policies overseen by high-level officials at the local, regional, national, ­ and international levels. Cooperation across government departments and com- plementary policies are necessary to foster an environment in which innovation can flourish. Implementing the right framework conditions will have positive spin-off benefits for the coordination of policies at the regional and local level. Improving measures of innovation is critical for policy making. The OECD (2010, 24) notes that the “current innovation indicators are too focused on the inputs of the innovation process rather than on its outcomes.� It puts forward a wide range of innovation indicators for policy making and suggests that this approach be taken up at the national and international level. Table 4.9 summarizes the elements of policy for innovation and highlights individual policy instruments that may be adopted. Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Table 4.9 Summary of Innovation Policy Elements Priority Principles Instruments Empowering people to Provide education and training Education policy: Change curriculum and pedagogy; evaluate teachers on ongoing basis; innovate Foster knowledge transmission engage employers and workers in curriculum development; ensure independence, competition, excellence, entrepreneurial spirit, and flexibility in universities; provide entrepreneurial education. Labor market policy: End nontransparent hiring and promotion in scientific institutions; open labor markets to foreign students; ensure tax regime does not penalize mobile skilled workers. Unleashing innovation in Foster entrepreneurship Labor market policies: Simplify and reduce start-up regulations and administrative burdens; firms Enhance access to finance foster open and competitive markets to facilitate the reallocation of resources that occurs Build foundations for innovation with sound when firms fail and new firms emerge. framework conditions Finance policies: Amend bankruptcy laws to facilitate restructuring of businesses, paying Foster markets for innovative goods, services, and attention to risk management and moral hazard. Adapt changes to develop a more processes neutral tax policy. Lower regulatory barriers so high-growth firms do not spend needed capital on bureaucracy; restore the health of the financial sector; develop well- functioning venture capital markets; securitize innovation-related assets (intellectual property); ease new and small firms’ access to debt finance and equity finance, e.g., through risk-sharing schemes with the private sector; develop seed capital and start- up financing by business angel funds and networks. Macroeconomic policies: Ensure sound policies for the macroeconomy; promote openness to trade (reduce tariff barriers, dismantle nontariff barriers, and liberalize capital markets); conclude the WTO’s Doha Development Agenda; promote investment, fiscal discipline, and strong and stable output growth. Microeconomic policies: Develop policies for competition,a tax policies, framework for intellectual property rights. Get prices right; develop standards and regulations governing public procurement, which provides important signals on future demand to the private sector. table continues next page 99 100 Table 4.9  Summary of Innovation Policy Elements (continued) Priority Principles Instruments Creating and applying Foster strong and effective public research Microeconomic policies: Create a strong and effective public research system; finance public knowledge Invest in a knowledge-supporting infrastructure research to better facilitate funding of multidisciplinary research; tie part of funding to Foster efficient knowledge flows, networks, and societal objectives; recognize that private investment may not take place when time markets horizon is long and outputs are not immediately marketable. Unleash innovation in the public sector Education policies: Grant greater autonomy to universities and public research organizations; establish guidelines for collaborative arrangements between universities and public laboratories; establish criteria for evaluating research performance; attract well-trained technology transfer personnel. Technology policies: Promote development of ICT; adopt the new standard for Internet protocol (IPv6); promote the relationship between broadband networks and energy, health, transport, and education; foster the integration of ICT investments in physical infrastructure such as buildings, roads, transport systems, health facilities, and electricity grids; protect intellectual property rights. Promote knowledge transfer across borders through tax treaties; review cross-country differences in regulations and commercial law. Develop knowledge networking infrastructure.b Competition policies: Ensure that the patent system is not used anticompetitively. table continues next page Table 4.9  Summary of Innovation Policy Elements (continued) Priority Principles Instruments Applying innovation to Foster international cooperation Technology policiesc: Develop a new model for the governance of multilateral cooperation address global and Tackle key challenges—climate change, health, and on international science, technology, and innovation, one focusing on the “setting of social challenges food security—through innovation work priorities, funding and institutional arrangements to support that work,� and on Bridge the gap in economic development through “procedures to ensure access to knowledge, transfer of technology and capacity building� innovation (OECD 2010, 20). Macroeconomic policies: Remove trade barriers that limit technology transfer across borders and develop mechanisms that enhance technology transfer and the development of knowledge markets;d strengthen framework conditions in developing countries— education, basic infrastructure (transport, rural energy, irrigation); modernize agriculture, carry out poverty reduction, develop ICT, strengthen institutions. Education policies: Foster academic partnerships, cross-border higher education, and scientific cooperation. Finance policies: Develop new financing mechanisms to provide incentives for innovation, e.g., venture capital, public-private partnerships Microeconomic policies: Develop pricing policies for environmental externalities; develop tax policies; implement standards; make use of subsidies. Improving the governance Link science, technology, and innovation policies to “A whole-of-government approach to policies for innovation is needed to encourage and measurement of economic growth innovation in its many forms. It requires stable platforms for coordinating actions, policies for innovation Develop data infrastructure to measure the policies with a medium- and long-term perspective, and attention from policy makers determinants and impact of innovation at the highest level. It also calls for coherence and complementarities between the local, Account for the role of innovation in the public sector regional, national and international levels� (OECD 2010, 23). Promote new statistical methods and interdisciplinary approaches to data collection Promote measurement of innovation for social goals and social impacts of innovation Source: OECD 2010. Note: WTO = World Trade Organization; ICTs = information and communication technologies; OECD = Organisation for Economic Co-operation and Development. a. The OECD has developed a Competition Assessment Toolkit to help governments. See http://www.oecd.org/daf/competition/competitionassessmenttoolkit.htm. b. “Some good practice exists (for example, in networking R&D [research and development] for emerging infectious diseases) but significant scale-up is required.� OECD 2010, 19. c. OECD 2010, 20. d. For example: “voluntary patent pools and other collaborative mechanisms for reducing transaction costs to access intellectual property.� OECD 2010, 20. 101 102 Innovation Policy for Competitiveness Conclusion The chapter discussed the definition and measurement of innovation and high- lighted examples of the positive relationship between innovation, economic growth, and competitiveness. It then examined the elements of an effective innovation policy capable of meeting the economic and social challenges of the modern age. The concept of innovation has evolved from a fairly narrow definition empha- sizing new products and processes to a broader systemic definition that empha- sizes the flow of technology and knowledge among people, enterprises, and institutions. The expanded definition calls for a more detailed measurement rubric in addition to the traditional focus on R&D spending alone. The expanding use of innovation surveys assists in providing both input and output measures of innovation, although these surveys are not yet widely used. Innovation is necessary for growth, both for developed economies seeking to push the technology frontier further and for developing countries wanting to catch up. The literature examining the relation of innovation to growth and com- petitiveness illustrates innovation’s significant contribution, while developments in new growth theory have enabled innovation to feature as a key explanatory variable in the endogenous growth models. Harnessing innovation for growth and competitiveness is critical in the modern, knowledge economy, particularly one recovering from recession and facing global challenges. Unsurprisingly, a system of innovation, associated with today’s economy, requires a “system of government� policy, or a “whole of government� approach. This includes coordinated demand- and supply-side policies at the local, national, and international level. There is no one-size-fits-all set of policies; the policies adapted will depend on the needs and capabilities of the underlying economies. Developing countries can learn from others but should also develop the capabil- ity to do some frontier work by investing in R&D and joining global research networks in, for example, nanotechnology and biotechnology. This step paves the way toward adapting potentially new and exciting technologies that may be key to new technological revolutions. Deciding how much to invest and in what areas depends of course on the underlying capabilities and ambitions of the country, and it also depends upon the strength of the country’s entrepreneurship. Governments have a key role to play in getting innovation out of the universities and research labs and into production and use. Pursuing competitiveness through innovation is an increasingly important objective of policy, given the preeminent role of innovation in the modern, knowledge-driven economy. Notes 1. “The ‘death of distance’ opens opportunities: new markets and narrower forms of specialization in ‘fragmented’ production and global value chains.� (S. Lall n.d.) 2. Organisation for Economic Co-operation and Development (OECD 2009) notes the shortcomings of research and development (R&D) as a measure of innovation, citing its focus on inputs, technological doings, and manufacturing activity. Patent data have Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Innovation Policy for Competitiveness 103 also been used as measures of innovation “but they cannot measure the full extent of innovative activities and suffer from some well-known limitations� (OECD 2009, 12). 3. The Oslo Manual was developed by the OECD in 1992 “to harmonise and ensure the quality of innovation surveys� (OECD 2009, 12). 4. Cross-country comparisons, however, “should be undertaken with caution given that there are differences in both response rates and in the methods used by countries to adjust for non-responses� (OECD 2009, 29). 5. Five indicators comprise the composite indicator for output-based innovation modes. (1) New-to-market international innovators collect data on enterprises that have introduced a product or process new to international markets or have developed a new product or process in house. (2) New-to-market domestic innovators collect data on those enterprises that have introduced product innovations new to domestic mar- kets; innovations are partly developed in house. (3) International modifiers represent those enterprises that have some in-house development activities but whose product and/or process innovations already exist on international markets. These may or may not be new to domestic markets. (4) Domestic modifiers refer to those innovating enterprises that operate only on domestic markets. Product or process innovations already exist on the domestic market but are new to the innovating enterprise. (5) Adopters are those enterprises that adopt the innovation of others (OECD 2009). 6. Inventive activities or formal innovations are measured by in-house R&D or a patent application. Collaboration is captured by the degree to which enterprises’ innovations were developed with or solely by others or the level of cooperation on innovations (OECD 2009). 7. Enterprises are classified according to four groups. (1) Technological innovators ­ engaging in product and/or process innovation only. (2) Nontechnological innovators engaging in marketing and/or organizational innovation only. (3) Technological and nontechnological innovators. (4) No innovations implemented (OECD 2009). 8. “An analysis of dual innovators can help provide a picture of how prevalent service innovation is in manufacturing enterprises (and conversely, the prevalence of goods innovation in the services sector)� (OECD 2009, 40). 9. Cited in Cantwell (2005). 10. Total factor productivity (TFP) is the growth in output that does not come from the growth in inputs; it is a proxy for innovation here. TFP and multifactor productivity (MFP) are used as measures of innovation in growth regressions. Hall (2011) surveys a number of innovation indicators that establish a quantitative link between produc- tivity growth and innovation. 11. Fagerberg (2005) identifies the Marx-Schumpeter model of innovation in which Schumpeter holds that technological competition, or competition through innova- tion, is the driving force of economic development. When a firm in a given sector introduces an innovative product, it will be rewarded, and other firms will seek to emulate this. The initial advantage enjoyed by the first firm will eventually be eroded, and the effects on growth caused by the innovation will slow down. Schumpeter held that imitators would also innovate and bring about a process of innovation diffusion, that is, one important innovation sets the stage for a plethora of subsequent innova- tions. The interdependencies between the initial and induced innovations imply that innovations and growth concentrate in certain sectors and certain geographic areas. This process of innovation underlies much of the subsequent research on industrial growth and international trade and competitiveness. Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 104 Innovation Policy for Competitiveness 12. Specific works cited are Salomon and Shaver (2005); Salomon (2006); Criscuolo, Haskel, and Slaughter (2005); and Castellani and Zanfei (2006). 13. These were Hong Kong SAR, China; Japan; The Republic of Korea; Malta; Singapore; Taiwan, China; Botswana; Brazil; China; Indonesia; Malaysia; Oman; and Thailand. 14. This list of four research areas is given by Traistaru-Siedschlag et al. (2006) who cite key sources for each as follows: for “innovative regions and milieux,� see Camagni (1991); Ratti, Bramanti, and Gordon (1997); and Crevoisier (2001). For “high-tech areas,� see Keeble and Wilkinson (1999, 2000). For “clusters of knowledge based industries,� see Cooke (2002). For “knowledge spillovers,� see Audretsch and Feldman (1996); and Bottazzi and Peri (2003). 15. An innovation system is “a network of organizations within an economic system that are directly involved in the creation, diffusion and use of scientific and technological knowledge, as well as the organizations responsible for the coordination and support of these processes� (SciDev Net: http://www.scidev.net/en/editorials/systems-of- innovation-their-time-has-come.html). 16. Data are from the World Bank, World Development Indicators database; see http:// data.worldbank.org/indicator/TG.VAL.TOTL.GD.ZS/countries/1W?display=graph (for merchandise trade) and http://data.worldbank.org/indicator/BG.GSR.NFSV​ .GD.ZS/countries/1W?display=graph (for trade in services). 17. The Economist summarizes some of the recent research that is pessimistic about inno- vation; see “Has the Ideas Machine Broken Down?� January 12, 2013. http://www​ .economist.com/news/briefing/21569381-idea-innovation-and-new-technology- have-stopped-driving-growth-getting-increasing. See also Gordon (2012). 18. “The drivers for innovation in the developed world have been centered on getting more (performance and productivity) from less (physical, financial, human capital) for more (profit, value to the shareholder). In contrast, the drivers in the developing world are to get more (performance, productivity) from less (cost) for more and more (people).� World Bank 2010, 48. References Audretsch, D., and M. Feldman. 1996. “Innovative Clusters and the Industry Life Cycle.� Review of Industrial Organization 11 (2): 253–73. Basile, R. 2001. “Export Behaviour of Italian Manufacturing Firms over the Nineties: The Role of Innovation.� Research Policy 30: 1185–201. Berdegué, Julio A. 2005. “Pro-Poor Innovation Systems.� International Fund for Agricultural Development, Rome. http://www.ifad.org/events/gc/29/panel/e/julio.pdf. Bottazzi, L., and G. Peri. 2003. “Innovation and Spillovers in Regions: Evidence from European Patent Data.� European Economic Review 47 (4): 687–710. Braunerhjelm, P., K. Ekholm, L. Grundberg, and P. Karpaty. 1996. “Swedish Multinational Corporations: Recent Trends in Foreign Activities.� Working Paper 462, Research Institute of Industrial Economics, Stockholm. Camagni, R. 1991. “Local ‘Milieu,’ Uncertainty and Innovation Networks: Towards a New Dynamic Theory of Economic Space.� In Innovation Networks, edited by R. Camagni, 121–44. London: Belhaven Press. Cantwell, J. 2005. “Innovation and Competitiveness.� In Handbook of Innovation, edited by J. Fagerberg, D. C. Mowery, and R. R. Nelson, 543–67. Oxford: Oxford University Press. Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Innovation Policy for Competitiveness 105 Castellani, D., and A. Zanfei. 2006. Multinational Firms, Innovation and Productivity. London: E. Elgar. Commission on Growth and Development. 2008. The Growth Report: Strategies for Sustained Growth and Inclusive Development. Washington, DC: World Bank. Cooke, P. 2002. Knowledge Economies: Clusters, Learning and Cooperative Advantage. London: Routledge. Crevoisier, O. 2001. “Der Ansatz des kreativen Milieus.� Zeitschrift für Wirtschaftsgeographie 45: 246–56. Criscuolo, C., J. E. Haskel, and M. J. Slaughter. 2005. “Global Engagement and the Innovation Activities of Firms.� Working Paper 11479, National Bureau of Economic Research, Cambridge, MA. The Economist. 2013. Has the Ideas Machine Broken Down?� January 12, 2013. http:// www.economist.com/news/briefing/21569381-idea-innovation-and-new- technology-have-stopped-driving-growth-getting-increasing. Fagerberg, J. 1987. “A Technology Gap Approach to Why Growth Rates Differ.� Research Policy 16 (2-4): 87–99. ———. 1988. “Why Growth Rates Differ.� In Technical Change and Economic Theory, edited by G. Dosi, C. Freeman, R. R. Nelson, G. Silverberg, and L. L. G. Soete, 432–57. London: Frances Pinter. ———. 2005. “Innovation: A Guide to the Literature.� In Oxford Handbook of Innovation, edited by J. Fagerberg, D. C. Mowery, and R. R. Nelson, 1–27. Oxford: Oxford University Press. Gordon, R. J. 2012. “Is US Economic Growth Over? Faltering Innovation Confronts the Six Headwinds.� Policy Insight 63, Centre for Economic Policy Research, London. Gupta, Anil K. 2007. “Towards an Inclusive Innovation Model for Sustainable Development.� Paper presented at the Global Business Policy Council of A. T. Kearney, Dubai, United Arab Emirates, December 9–11. http://www.sristi.org/.../ Towards%20an%20inclusive%20innovation%model%20for%20sustainable%20 ­development.doc. Hall, B. H. 2011. “Using Productivity Growth as an Innovation Indicator.� European Commission. http://ec.europa.eu/commission_2010-2014/geoghegan-quinn/hlp/ documents/20120309-hlp-productivity-innovation_en.pdf. Hulten, C., and A. Isaksson. 2007. “Why Development Levels Differ: The Sources of Differential Economic Growth in a Panel of High and Low Income Countries.� NBER Working Paper 13469, National Bureau of Economic Research, Cambridge, MA. IMF (International Monetary Fund). 2010. “The Fund’s Role Regarding Cross-Border Capital Flows.� Strategy, Policy and Review Department and the Legal Department, IMF, Washington, DC. http://www.imf.org/external/np/pp/eng/2010/111510.pdf. ———. 2011. World Economic Outlook. Washington, DC: IMF. Isaksson, A. 2007. “Determinants of Total Factor Productivity: A Literature Review.� Geneva, Switzerland: United Nations Development Organization. Keeble, D., and F. Wilkinson. 1999. “Collective Learning and Knowledge Development in the Evolution of Regional Clusters of High-technology SMEs in Europe.� Regional Studies 33 (special issue): 295–303. ———, eds. 2000. High-technology Clusters, Networking and Collective Learning in Europe. Aldershot, UK: Ashgate. Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 106 Innovation Policy for Competitiveness Kumar, N., and N. S. Siddhartan. 1994. “Technology, Firm Size, and Export Behavior in Developing Countries.� Journal of Developing Studies 31 (2): 288–309. Lall, S. n.d. “Competitiveness, FDI, Trade and Innovation: A Global Perspective,� slide presentation, Economics Department, University of Oxford. http://www.slideworld​ .com/slideshows.aspx/Competitiveness-FDI-trade-and-innovation-A-global- ppt-320737. Lichtenberg, F. 1992. “R&D Investment and International Productivity Differences.� Working Paper 4161, National Bureau of Economic Research, Cambridge, MA. Nassimbeni, G. 2001. “Technology, Innovation Capacity, and the Export Attitude of Small Manufacturing Firms: A Logit/Tobit Model.� Research Policy 30: 245–62. Nelson, R. R., and H. Pack. 1999. “The Asian Miracle and Modern Growth Theory.� Economic Journal 109 (457): 416–36. OECD (Organisation for Economic Co-operation and Development). 2009. Innovation in Firms: A Microeconomic Perspective. Paris, France: OECD. ———. 2010. “Ministerial Report on the OECD Innovation Strategy. Innovation to Strengthen Growth and Address Global Challenges.� Key Findings. May. Paris, France: OECD. http://www.oecd.org/sti/45326349.pdf. Özçelik, E., and E. Taymaz. 2004. “Does Innovativeness Matter for International Competitiveness in Developing Countries? The Case of Turkish Manufacturing Industries.� Research Policy 33:409–24. Petrakos, G., P. Arvanitidis, and S. Pavleas. 2007. “Determinants of Economic Growth: The Experts’ View.� DYNREG Working Paper 20/2007. http://www.esri.ie/research/ research_areas/international_economics/dynreg/papers/Working_Paper_No._20.pdf. Ratti, R., A. Bramanti, and R. Gordon, eds. 1997. The Dynamics of Innovative Regions: The GREMI Approach. Aldershot, UK: Ashgate. Salomon, R. 2006. “Spillovers to Foreign Market Participants: Assessing the Impact of Export Strategies on Innovative Productivity.� Strategic Organization 4 (2): 135–64. Salomon, R., and J. M. Shaver. 2005. “Learning by Exporting: New Insights from Examining Firm Innovation.� Journal of Economics and Management Strategy 14 (2): 431–60. Traistaru-Siedschlag, I., ed., G. Murphy, M. Schiffbauer, G. Petrakos, L. Resmini, C. Pitelis, G. Maier, M. Trippl, P. Nijkamp, P. van Hemert, J. Vilrokx, A. Rodríguez- Pose, J. Damjian, and C. Kostev. 2006. “Dynamic Growth Regions, Innovation and Competitiveness in a Knowledge Based World Economy: A Survey of Theory and Empirical Literature.� Workpackage No. 1, DYNREG, Economic and Social Research Institute. http://www.esri.ie/research/research_areas/international_­ economics/­dynreg/papers/DYNREG_D1.1.pdf. Ulku, H. 2004. “R&D Innovation and Economic Growth: An Empirical Analysis.� Working Paper WP/04/185, International Monetary Fund, Washington, DC. World Bank. 2010. Innovation Policy: A Guide for Developing Countries. Washington, DC: World Bank. https://openknowledge.worldbank.org/bitstream/handle/10986/2460/5 48930PUB0EPI11C10Dislosed061312010.pdf?sequence=1. Wyckoff, A. 2010. “OECD’s Innovation Strategy: Key Findings and Policy Messages.� Slide presentation, Organisation for Economic Co-operation and Development, Paris, France. http://www.oecd.org/site/innovationstrategy/45154092.ppt. Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Chapter 5 Competitiveness and Clusters “A cluster is a system of interconnection between private and public sector enti- ties. It usually comprises a group of companies, suppliers, service providers, and associated institutions in a particular field, linked by externalities and comple- mentarities� (World Bank 2009, vii). Clusters usually have a specific spatial dimension as well because interlinked firms often concentrate in a specific geo- graphic area. First proposed by Michael Porter in 1990, cluster development has been embraced by policy makers and academics as a means for stimulating an area’s economic development and growth. It has become increasingly important in the context of globalization, which has left many regions and nations strug- gling to remain competitive. Governments and private sector entities, acting either as a cluster initiative (CI) organization or through a cluster-based competi- tiveness project, support links among firms and industries at a regional level to promote an area’s growth and competitiveness. Initially associated with devel- oped economies, cluster-based competitiveness projects have since 2000 also been implemented in developing economies. Public policy, through regional policy as well as policy for science and technol- ogy and industry, has implications for cluster development and competitiveness. The optimal form and depth of policy to promote cluster development remain subject to debate. Policy for cluster development at the regional level has focused mostly on lagging regions. Science and technology policies focus mainly on pro- moting growth efforts among technology companies and on supporting research and development (R&D), while industrial policies strive to promote an area’s growth, perhaps by focusing on small and medium enterprises (SMEs). The dif- ferent types of policies are not mutually exclusive; regional, science and technol- ogy, and industrial policies may all share the common goal of innovation, which is critical for long-term productivity growth. The chapter examines the background to cluster development and competi- tiveness and then goes on to discuss some CIs and cluster-based competitiveness projects. The final section examines the policy implications of CIs. Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8   107   108 Competitiveness and Clusters Background to Clusters Agglomeration or clustering occurs at many geographic levels and can take many forms. Scale externalities and knowledge spillovers promote agglomeration, which leads to different outcomes depending upon whether these spillovers operate at the general level or at the level of related firms and industries. Outcomes also depend upon whether spillovers improve static efficiency and flexibility in general or innovation and upgrading of competitiveness specifically (see table 5.1; Ketels, Lindqvist, and Sölvell 2008). Agglomeration spillovers that arise from economic activity in general and that promote efficiency and flexibility at the urban level lead to metropolises and to industrial districts at the level of technologically related industries (table 5.1). These agglomeration effects are also referred to as economies of scale and scope. A more dynamic, global world, where firms and industries are involved in inno- vation and upgrading, leads to clusters that rely upon knowledge creation and innovation as well as the traditional flows of goods and services. More generally, at the level of the region, this innovation and upgrading can lead to the concept of the creative region.1 “Agglomerations of economic activity in general, and clusters in particular, are natural economic and social phenomena, both in earlier times and in the modern economy� (Ketels, Lindqvist, and Sölvell 2008, 3). Cortright (2006) traces clusters’ theoretical background to the social sciences. He begins with the contributions from the neoclassical school of economic thought and pro- gresses to contributions from social scientists emphasizing nonmarket social forces and relationships, such as customs/traditions, technological change, and social networks (see table 5.2). There are many types of clusters, each with its own characteristics. Ketels, Lindqvist, and Sölvell (2008, 3) suggest that clusters differ on a number of dimensions: • Well-established clusters versus clusters that are just emerging; • Large and dense clusters with a multitude of related industries and associated organizations and institutions, as opposed to thinner and smaller clusters; • Manufacturing-oriented clusters such as automotive versus more service-­ oriented clusters such as financial services; • Science-driven clusters and clusters in traditional sectors; • Clusters with strong external links and global reach (hot spots) as opposed to clusters with a mere regional reach. Table 5.1  Four Types of Economic Agglomerations Economic activity in general Technologically related industries Efficiency (scale) and flexibility Metropolises Industrial districts Innovation and upgrading Creative regions Clusters Source: Ketels, Lindqvist, and Sölvell 2008. Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Competitiveness and Clusters 109 Table 5.2 Social Science Contribution to Understanding of Clusters Summary Contributors Neoclassical Alfred Marshall Marshall is credited with providing the first clear description of industry clusters. Marshall (1920) He identified 3 key reasons (labor market pooling, supplier specialization, knowledge spillovers) for why industrial clusters would emerge and in doing so identified “external economies�—productive benefits that are not captured by the individual firms that create them Regional science Regional scientists refined Marshall’s idea that firms benefit from being in close Isard (1956) proximity and identified localization economies (gains from proximity to similar Hoover and firms) and urbanization economies (gains from proximity to dissimilar firms); the Giarratani concept of space was reintroduced into thinking about the economy; interest in (1948) the field waned after the 1960s, due in large part to its lack of a theoretical basis Chinitz (1961) Jane Jacobs Although not an economist by training, Jacobs’s view that the creation and Jacobs (1969) development of new products and technologies as the source of economic development occurred most successfully in cities where inhabitants cluster and generate new ideas and her broadening of urbanization economies to include other types of diversity New economic Rekindling of interest in Marshall’s theory about why firms locate in geographic Fujita, geography agglomerations; models indicate the geographic clusters of firms likely to Krugman, form when increasing returns to scale are strong; firms have power to set and prices; transportation costs are low; and customers, suppliers, and workers are Venables geographically mobile (1999) Urban and Economists study the spatial aspects of a variety of economic problems; debate Henderson regional about the relative importance of localization and urbanization economies; no (1997) economics consensus on whether industrial specialization or diversity is more important to Glaeser et al. regional growth (1992) The social and institutional traditiona Business Analyzes the organization of production within and between firms. During the first organization half of the 20th century, the organization of production was dominated by “mass production� or Fordist production systems. Large firms could use economies of scale in production and in marketing to achieve lower costs and dominate markets In several areas, groups of small firms flourished in highly specialized markets; Piore and Sabel small firms were competitive through flexible specialization; groups of firms (1984) in industrial districts were supported by a variety of institutions and culture of cooperation that enabled them to mimic or offset many of the advantages of scale economies. Termed the second industrial divide Geography Geographers and urban and regional planners have taken an interest in industrial and urban districts and clusters and their role in city growth and development by and regional emphasizing the nature of the relationships among firms in a region as a source of planning clustering and juxtaposing local and global interactions in determining the role of cities in development Michael Porter Theory draws on neoclassical and social and institutional traditions as well as from Porter (1990, and business business strategy. Describes industry clusters as the product of four factors (factor 2001, 2008) strategy conditions, demand conditions, related and supporting industries, and firm strategy and competitiveness) termed “the diamond of competitive advantage.� The diamond explains why clusters are more competitive than individual firms Economic Examines the operation, development, and promotion of clusters producing Rosenfeld development practical insights into the nature of industrial agglomeration (1997) practitioners Source: Cortright 2006. a. This approach emphasizes the effects of social forces and relationships (such as customs, technological change, organizations, and social networks) that cannot be fully reduced to market decisions of individuals. Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 110 Competitiveness and Clusters Figure 5.1 An Agribusiness Cluster Seeds nurseries Crop-processing equipment Fertilizer, insecticides, herbicides State government and Transportation donor agencies Farm equipment Packaging services Storage facilities Public relations advertising Irrigation technology Growers Processors Specialized publications Clusters of other Financial services agricultural products Clusters of buyer/ Educational, research, and consumer industries trade operations Source: World Bank 2009. An industrial cluster represents an agglomeration of diverse actors—firms, suppliers, service providers, and related companies—in a specific industry. Figure 5.1 shows a typical cluster in the agribusiness sector. In addition to the entities directly involved in the agribusiness sector, there are more tangential entities such as educational, research, and trade operations and state government and donor agencies. Geographic proximity as well as synergy from the activities of the various actors generates positive economic benefits such as “access to specialized human resources and suppliers, knowledge spill- overs, [and] pressure for higher performance in head-to-head competition� (World Bank 2009, 2). Industrial clusters were used as a vehicle for productivity and as a means of enhancing the competitiveness of an area, a region, or a nation. The cluster concept is not new, but it is Porter’s version that has gained widespread currency ­ among policy makers, academics, and industrial organizations.2 Porter’s definition of competitiveness grows out of an understanding of productivity as something that arises from “successful innovation in processes, or ­ products, or both.�3 His discussion of clusters and competitiveness (Porter 2008, 9) emphasizes the following: • Clusters increase productivity and efficiency. • Clusters stimulate and enable innovations. • Clusters facilitate commercialization and new business formation. Clusters increase productivity and efficiency by facilitating efficient access to specialized inputs, services, employees, information, institutions, training programs, and other public goods. The existence of clusters increases the likeli- hood that opportunities for business will be recognized and also provides an Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Competitiveness and Clusters 111 environment in which businesses can come together to share knowledge and/ or create knowledge. Clusters help in bringing ideas to market (commercial- ization) because the opportunity for new products or processes is more appar- ent in clusters. Spin-off companies and start-ups are encouraged by the presence of other companies and the availability of skills and suppliers, for example. Porter (2008, 6) defines the cluster as “a geographically proximate group of interconnected companies and associated institutions in a particular field, linked by commonalities and complementarities (external economies).� A cluster may contain • an end-product industry or industries; • downstream or channel industries; • specialized suppliers; • providers of specialized services; • related industries (those with important shared activities, labor technologies, channels, or common customers); • supporting institutions, including financial, training, standard setting, and research institutions as well as trade associations. Examples of clusters in the United States are shown in table 5.3. The clusters listed are the three highest-ranking clusters in terms of share of national employ- ment. The data are from the cluster mapping project at Harvard Business School. Table 5.3 Regional Specialization—Clusters in the United States, 2008, Selected Geographic Areas Geographic area Cluster(s) Atlanta Construction materials Transportation and Logistics Business services Boston Analytical instruments Education and knowledge Communications equipment creation Chicago Communications equipment Processed food Heavy machinery Denver Leather and sporting goods Oil and gas Aerospace vehicles and defense Houston Oil and gas production and Chemical products Heavy construction services services Los Angeles Apparel Building fixtures, equipment, Entertainment and Services Pittsburgh Construction materials Metal manufacturing Education and knowledge creation Raleigh-Durham Communications equipment Information technology Education and knowledge creation San Diego Leather and sporting goods Power generation Education and knowledge creation San Francisco-Oakland- Communications equipment Agricultural products Information technology San Jose Bay Area Seattle-Bellevue-Everett Aerospace vehicles and defense Fishing and fishing products Analytical instruments Wichita Aerospace vehicles and defense Heavy machinery Oil and gas Source: Porter 2008. Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 112 Competitiveness and Clusters Figure 5.2  Determinants of Innovative Capacity Innovative capactiy Company innovation Common innvovation orientation infrastructure Quality of linkages Cluster-specific conditions Source: Porter 2008. © Michael E. Porter. Used with permission; further permission required for reuse. The success of industrial clusters depends among other things on their capac- ity to innovate, whether it be the technological innovation that characterizes the information technology (IT) clusters in Silicon Valley and Bangalore or the ­ creative innovation representative of the fashion design clusters in Paris and Mumbai (World Bank 2009). The determinants of innovative capacity are outlined in figure 5.2. The innovative capacity of the company depends very ­ much upon the quality of the links between the innovative orientation and potential of the company and the cluster-specific conditions. Productivity growth for Porter arises from the interactions between the four factors in his diamond model: firm strategy, structure, and rivalry; input factor conditions; demand conditions; and the presence of related and supporting industries. The diamond is shown in figure 5.3. The interface between firms in a geographic area is primarily one of competi- tive rivalry although collaboration can also be important.4 Geographical proxim- ity affects competitiveness in three ways: 1. It increases productivity—firms can operate with lower levels of stock because of the local presence of specialized suppliers, and they have access to special- ized skills and human resources, aided by specialized and local training providers. 2. It increases the capacity for innovation by facilitating interaction and the dis- semination of knowledge—competition between firms raises the incentive to innovate, which in turn raises the capacity to adapt to changes and external shocks. 3. It stimulates and enables new business formation through spin-off enterprises that face lower barriers to entry than in other localities—this in turn creates Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Competitiveness and Clusters 113 Figure 5.3 Productivity and the Business Environment Context for firm strategy and rivalry Local rules and incentives that encourage investment and productivity For example, incentives for capital Factor (input) investments, intellectual conditions property protection Vigorous local competition Demand High-quality, e cient, and specialized industries inputs to business Openness to foreign and Natural endowments local competition Demanding and sophisticated Human resources local customers and needs Capital availability Challenging quality, safety, Physical infrastructure and environmental standards Administrative infrastructure (e.g., registration, permitting) Information infrastructure (e.g., economic data, corporate Related and supporting disclosure) industries Scienti c and technological infrastructure Capable, locally based suppliers and supporting industries Presence of clusters instead of isolated rms Source: Porter et al. 2008. © World Economic Forum. Used with permission; further permission required for reuse. a positive feedback loop through more competition, innovation, and so on (see note 3). Industrial clusters are not confined to one particular geographic area but may instead span regional or national boundaries. In fact, Porter (2008) noted that a region’s clusters were also likely to be present in neighboring regions. Export- oriented clusters tend to have a lower share of employment but higher average wages, productivity, and innovation (table 5.4). Traded clusters—those clusters made up of traded industries—account for just over 29 percent of employment, compared to 70 percent for local clusters. However, the average wage in the traded clusters was almost $50,000, compared to just over $30,000 in the local clusters. Globalization has been positive for cluster development. As markets globalize, firms have a choice of where to locate. As resources flow to given areas, the role of clusters is reinforced and regional specialization ensues. Ketels, Lindqvist, and Sölvell (2008) suggest that this process leads to clusters becoming increasingly specialized and increasingly connected with other clusters. Moreover, clusters that are successful are more likely to participate in the global marketplace and connect Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 114 Competitiveness and Clusters Table 5.4 Composition of Regional Economies, United States, 2004 Natural resource–driven Traded clusters Local clusters industries Number 40 19 n.a. Share of employment (%) 29.3 70.0 0.7 Employment growth 1990–2004 (%) 0.7 2.4 −1.2 Average wage $49,367 $30,416 $35,815 Relative wage (%) 137.2 84.5 99.5 Wage growth (%) 4.2 3.4 2.1 Relative productivity 144.1 79.3 140.1 Patents per 10,000 employees 23.0 0.4 3.3 Number of SIC 590 241 48 Source: Porter 2008. Note: n.a. = not applicable; SIC = standard industrial classification. Figure 5.4  Dimensions of Clusters and Economic Policy Clusters as a Clusters as a means of platform to implementing inform public economic policy policies Clusters initiatives that mobilize public-private e orts to improve the business enviroment and capture spillovers Source: Porter 2008. © Michael E. Porter. Used with permission; further permission required for reuse. with other clusters providing complementary activities. As shown in figure 5.3, the existence of one cluster may beget another. World Bank (2009, 3) identifies the optics cluster in Arizona as one that subsequently gave rise to clusters in “plastics, aerospace, environment technologies, information technologies and biosciences.� The role of government is important, particularly in regard to the regulatory environment for business and national policies that affect education and skills. Porter (2008) suggests that the old model of economic development, predicated on government driving economic development through policy decisions and incentives, is redundant. The new model understands economic development as a collaborative process involving government at multiple levels, as well as com- panies, teaching and research institutions, and institutions for collaboration. Clusters have a significant role to play in this new model, as shown in figure 5.4. Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Competitiveness and Clusters 115 Cluster Initiatives “Cluster initiatives are organized efforts to increase the growth and competitive- ness of clusters within a region, involving cluster firms, government and/or the research community� (Sölvell, Lindqvist, and Ketels 2003). Findings from the Global Cluster Initiative Survey (Sölvell, Lindqvist, and Ketels 2003) suggest that “almost all Cluster Initiatives have a dedicated facilitator and many (68%) have some sort of office. Many (78%) spend time and efforts to build a frame- work of shared ideas about why the Cluster Initiative is beneficial and how it is supposed to work� (Ketels, Lindqvist, and Sölvell 2008, 7).5 Intelligence on how things are supposed to work comes from an examination of the clusters’ own strengths and capabilities. Most CIs formulate a plan. Dialogue between the vari- ous stakeholders is key to forming new partnerships between cluster leaders and the various public sector organizations. The different CIs have different objectives, but each engages in formally orga- nized efforts with the government and the private sector. Examples of objectives include the following: • Facilitating market development through joint market assessment, marketing, and brand building; • Encouraging relationship building (networking) within the cluster, within the region, and with clusters in other locations; • Promoting collaborative innovation—research, product and process develop- ment, and commercialization; • Aiding the innovation diffusion, that is, the adoption of innovative products, processes, and practices; • Supporting the cluster’s expansion through attracting firms to the area and supporting new business development; • Sponsoring education and training activities; • Representing cluster interests before external organizations such as regional development partnerships; national trade associations; and local, state, and fed- eral governments (Mills, Reynolds, and Reamer 2008). The range of objectives is quite broad, and CIs tend to cover four to five main objectives. Older CIs tend to be more narrowly focused compared to younger CIs (Ketels, Lindqvist, and Sölvell 2008). World Bank (2009) provides an overview of one approach to developing a CI. As figure 5.5 shows, the first stage of development involves mapping and engage- ment with stakeholders; in the second stage, 10 cluster tools are applied to identify gaps in the cluster’s competitive position and to aid in developing col- ­ laboration and collective business strategies among the cluster members.6 In the third stage, leader cluster members then form partnerships with various public sector organizations to expedite policy reform in areas such as industrial develop- ment, infrastructure development, research, innovation, and training. Finally, in the fourth stage, the industrial links formed through clusters provide a solid basis Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 116 Competitiveness and Clusters Figure 5.5 One Approach to Developing a Cluster Initiative Stage 1 Stage 2 Stage 3 Stage 4 Cluster Diagnostics Implementation Postproject mapping and and strategy of strategic, sustainability initial engagement formulation policy, and institutional initiatives Conduct economy-wide Apply the 10 cluster Secure ownership from Ensure that cluster can cluster mapping; tools to ascertain cluster's key cluster leaders in handle resources indepen- identification and competitive position, terms of time, ideas, and dently beyond the life of engagement with develop collaboration cost sharing; conduct the project; do due key cluster stakeholders among cluster members, public-private dialogues diligence and formalize and develop collective on policy and institutional the institutional structure business strategies bottlenecks for imple- of the cluster mentation of business strategies on cluster competitiveness Source: World Bank 2009. for the formulation and sequencing of policy reforms. CIs become in effect a tool for government in pursuing policy reform because “together they [CIs and policy reform] may create positive externalities by informing government of the policy implications and possible business responses� (World Bank 2009, 5). The GCIS (2003) suggested that CIs are initiated by government in 32 ­ percent of cases, by industry in 27 percent, or equally by both in 35 percent. Slightly over half of financing comes primarily from government (54 percent of cases), while companies are the most influential parties in the governance of CIs (Ketels, Lindqvist, and Sölvell 2008). While clusters may evolve naturally, a CI can hasten the process and concentrate on the areas in which policy and institu- tional impediments may be hindering competitiveness. Some key areas of focus include “market information, workforce development, supply chain improve- ments, quality standards, branding, forward integration, and process improve- ments� (World Bank 2009, 4). The number of CIs has increased significantly over time. The 2003 GCIS identified over 500 CIs across Europe, North America, Australia, and New Zealand, and the number has likely grown since then.7 The Cluster Initiative Greenbook (Sölvell, Lindqvist, and Ketels, 2003) reports the following: • The performance of CIs is measured along three dimensions—innovation and international competitiveness; cluster growth; and goal fulfillment. In all, 85 percent agree that the CI has improved the competitiveness of the cluster it was set up to serve, and 89 percent report that the CI helped the cluster grow. About 81 percent of CIs have met their goals. • The national social, political, and economic setting within which each CI is implemented is important for the performance.8 • CIs serving strong clusters of national and regional importance are more successful than those serving weaker clusters. ­ Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Competitiveness and Clusters 117 • CIs initiated through a competition process to get government financing perform significantly better in terms of increasing international ­ competitiveness.9 • CIs limited to domestic companies perform worse than those that are not limited. • CIs with offices and budgets sufficient to conduct significant projects without seeking separate funding perform better than those that lack them.10 • For the facilitator, having a broad network of contacts is the most important success factor, but the facilitator’s qualities are more important for competi- tiveness performance than for growth performance. • CIs that build a clear, explicit framework, based on the cluster’s own strengths, and that spend time to share this framework with all parties, are clearly more successful in promoting cluster competitiveness than those that do not take these steps. • Generally, disappointing results for CIs, including the failure to generate changes, are related to poor consensus, weak frameworks, facilitators lacking strong networks, lack of offices and sufficient budgets, and neglected brand building. Disappointing CIs tend to involve less important clusters. • Government policy and other setting factors also influence performance indi- rectly, by affecting the objectives CIs pursue and some process issues. For example, in countries where local government decision makers are important, CIs tend to pay more attention to various competitiveness-related objectives, such as promoting new technology and monitoring technical trends. The setting of each CI depends upon the country’s underlying level of eco- nomic development, which has implications for the range of objectives and the manner in which each CI is initiated, financed, and organized. CIs are therefore country- and industry-specific but are most common in developed and transition economies. They tend to focus on technology-intensive areas “like IT, medical devices, production technology, communications equipment, biopharmaceuti- cals, and automotive� (Ketels, Lindqvist, and Sölvell 2008,6). CIs and cluster-based competitiveness projects have been associated with advanced economies since the mid-1990s, and have been part of the economic development framework for developing and transition economies since the year 2000. International donor organizations have become involved in and have initi- ated CIs. In fact, donor-initiated CIs are located in the most challenging settings, even in relation to CIs in developing and transition economies. Donor-initiated CIs, which operate in locations where there is little national policy support for CIs, help circumvent some of the trust issues prevalent in developing and transi- tion economies.11 Because policy is more likely to be centralized in developing and transition economies and preoccupied with macroeconomic issues, “there is usually little policy support relating to competitiveness and clusters� (Ketels, Lindqvist, and Sölvell, 2006, 5). “Cluster Initiatives in developing countries face very different challenges and often have different types of specific objectives compared to those in transition Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 118 Competitiveness and Clusters Table 5.5 Comparison of Cluster Initiatives by Level of Economic Development Type of economy Measure Developing Transition Advanced Objectives CIs focus on supply chain Donor-initiated CIs have a CIs focus on innovation and development, export narrower range—export business environment promotion promotion and increasing improvement Increasing value-added, value-added improving business environment Activities Upgrading human resources, Lobbying for changes in Firm formation; high developing supply chain, and business environment; importance of joint R&D working out joint logistics management training; supply chain development Membership and 71% of CIs have an office; 37% Fewer companies CIs are larger, 51% have more resources have a website; median of 3 participating in CIs— than 20 firms participating staff members median number is 18 with and median is 25 companies just 40% of CIs having 75% of CIs have an office; 79% more than 20 companies have a website; median of 2 62% of CIs have an office; staff members 41% have a website; median of 2 staff members Cluster focus Focus on “basic� industries More of a mix of industry Sometimes a tendency to favor types; but donor initiators “high-tech� industries focus on “basic� industries and agriculture Role of government CIs often have an international Largest share of funding Dominating role of government and financing initiator; government comes from business that leaves business on the initiatives are also frequent; sector sidelines of CIs is a concern those initiated by business are Most of financing for CIs is less frequent provided by government International funding is usually the main source of income for CIs Performance Developing economies score CIs in transition economies CIs in advanced economies best in acquiring funds and report their best results score best in increasing improving the business in acquiring funds innovativeness environment, followed by from government export promotion and international organizations, improving business environment and increasing innovativeness Source: Compiled from Ketels, Lindqvist, and Sölvell 2006. Note: CI = cluster initiative; R&D = research and development. economies, and there is no simple linear relationship from developing to transi- tion to advanced economies�12 (Ketels, Lindqvist, and Sölvell 2006, 5). Table compares CIs in advanced, developing, and transition economies on a range 5.5 ­ of measures. The information is compiled “from a survey of 1,400 cluster initia- tives, including comprehensive data from 450 CIs that completed the Global Cluster Initiative Survey [in] 2005,� as reported in Ketels, Lindqvist, and Sölvell (2006, 5). Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Competitiveness and Clusters 119 The main findings from the global survey may be summarized by the following: Each CI must find the approach that will be most effective by taking into consideration the level of development of the economy (developing, transition, or advanced) and the barriers to competitiveness faced by each cluster. Removing barriers to competitiveness depends upon the country’s economic policy agenda. The environment for CIs is more accepting, if clusters are accepted as a tool for economic development and if competitiveness is part of an economic d ­ evelopment plan—locally, regionally, or nationally. Greater centralization of decision making in transition and developing economies has implications for cluster development and CIs. First, clusters are essentially a local phenomenon that benefit from the involvement of local and regional government. CIs will be compromised by insufficient decision-making power at the local level (see bars “a� in figure 5.6). Second, “competitiveness and clusters play less of a role in economic policy� in transition economies than in others (bars “b� and “c� in figure 5.6; Ketels, Lindqvist, and Sölvell 2006, 29). This may be due to a greater emphasis on macroeconomic policy in these coun- tries. On the other hand, developing, transition, and advanced countries all consider competitiveness an important issue in the economic policy debate (bars “d� in figure 5.6). CIs in transition and developing economies operate in a more challenging envi- ronment than those in advanced economies because of the low levels of trust and an economic policy that is less oriented toward competitiveness and clusters. The overall success of the cluster depends upon trust between the various participants in the cluster. Trust improves as economic development progresses, although trust between firms and government in transition economies is lower than in develop- ing economies (figure 5.7). CIs are stymied in low-trust environments, both in Figure 5.6 Policy Setting in which CIs Are Conducted 7 Reply scale (average) 6 5 4 3 2 1 a. Centralization b. National c. Cluster d. Economic strategy policy debate Developing economy Transition economy. Advanced economy Source: Ketels, Lindqvist, and Sölvell 2006, 29. © Christian Ketels, Göran Lindqvist, and Örjan Sölvell. Used with permission; further permission required for reuse. Note: CI = cluster initiative. The sets of bars show responses to survey questions by cluster initiative facilitators, indicating agreement on a scale of 1 (disagree completely) to 7 (agree completely). (a) Economic development policy is driven by initiatives at the national government level, not at the local/regional level. (b) The national government has a clear strategy for improving competitiveness. (c) Cluster policies are a core element in economic development policy. (d) Competitiveness is a key issue in the economic policy debate. Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 120 Competitiveness and Clusters Figure 5.7 Level of Trust between Firms and between the Private and Public Sector 7 Level of trust (average) 6 5 4 3 2 1 a. Firms’ trust b. Firms’ trust c. Firms’ trust d. Governments’ in firms in government in academia trust in firms Developing economy Transition economy Advanced economy Source: Ketels, Lindqvist, and Sölvell 2006. © Christian Ketels, Göran Lindqvist, and Örjan Sölvell. Used with permission; further permission required for reuse. Note: The sets of bars show responses to survey questions, indicating agreement on a scale of 1 (disagree completely) to 7 (agree completely). (a) Firms’ trust in other firms; (b) firms’ trust in government initiative; (c) firms’ trust in academia; and (d) Government trusts in firms. Figure 5.8 Objectives Considered Most Important for the CI 100 respondents (%) 75 Share of 50 25 0 de e ts n en ss en s co ce ds ar e ai y m se st irm ad alu se liz io ch ppl or nm e n du un oy ea d ns t t t s ch ro in at re ia en st xp v ve t f su tio e kf vi us pl ncr ic erc ov se uc h. R m in ac e en e b ee nn p se ea em m em . I d ttr lo i. S e ad om ea ti ov cr an A ve or In cr pr g. De ac . C pp In a. od Im j d. b. Su pr f. c. Developing economy Transition economy Advanced economy Source: Ketels, Lindqvist, and Sölvell 2006. © Christian Ketels, Göran Lindqvist, and Örjan Sölvell. Used with permission; further permission required for reuse. Note: CI = cluster initiative. The sets of bars show the percentage of survey respondents who ranked the objective as one of the three most important. how effectively they can operate and in beginning to operate in the first place. Among advanced economies, where trust is highest, CIs can develop an action plan from the beginning. CIs also represent an important vehicle for increasing trust over time. CIs in developing and transition countries usually have different objectives from those in advanced economies. The survey found that CIs in developing and transition economies emphasize value added and exports, whereas CIs in advanced economies emphasize innovation and improving the business envi- ronment. Figure 5.8 shows which objectives are considered among the most important by respondents categorized by economy type. Ketels, Lindqvist, and Sölvell (2006) suggest that the differences in objectives may be related to the lack of local government involvement in developing and transition economies. Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Competitiveness and Clusters 121 Firms in these economies focus on aspects of competitiveness that they can affect, such as in-house activities. Selecting the right cluster for each CI will depend among other things on the type of industry and the strength of cluster. The type of industry is given when the CI is initiated by the business sector itself. The industry needs to be selected for CIs when the government or donor is the initiator. For developing countries in particular, and for transition economies, “basic� industries—agriculture, food, and basic manufacturing—are the most common type of industry in CIs and particularly where donors are the initiators. Figure 5.9 shows the target indus- tries. High-tech industries are more dominant in advanced country CIs. Transition country CIs are spread across the various industrial sectors. The pattern shown in figure 5.9 may reflect the general industry profile in the underlying economies; that is, we would expect there to be a lot of basic industries in developing econo- mies and a lot of high-tech industries in the advanced economies. Whether this pattern reflects a bias on behalf of the CIs is a separate research question and a valid one. What is understood by the concept of a “strong� cluster differs depending on the economic development of the underlying country. CIs target strong clusters, and in advanced economies this usually means clusters with a strong competitive position and capacity to innovate (see figure 5.10). Donors in transition countries tend to target those clusters that are less developed than those targeted by government and/or the business sector. Figure 5.10 shows the various dimensions ­ that can measure the strength of a cluster. CIs target clusters that have global market reach, economic importance, and growth potential, with little difference across the remaining dimensions, except, as noted above, the competitive position and the capacity for innovation. Differences do arise when comparing the clusters targeted by donors to those targeted by government and/or the business sector in developing and transition counties (figure 5.11). Figure 5.9 Target Industries Selected by Donors or Government for the Purposes of CIs 100 respondents (%) 75 Share of 50 25 0 Argriculture, Capital "High tech," Tourism food, basic intensive advanced manufacturing manufacturing services Developing economy Transition economy Advanced economy Source: Ketels, Lindqvist, and Sölvell 2006. © Christian Ketels, Göran Lindqvist, and Örjan Sölvell. Used with permission; further permission required for reuse. Note: CI = cluster initiative. The sets of bars show the percentage of survey respondents (government or donors) targeting industries (for the purposes of cluster initiatives) in developing, transition, and advanced economies. Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 122 Competitiveness and Clusters Figure 5.10 Cluster Strength 7 6 Average reply 5 4 3 2 1 th n ity ea al e y ry st nd t n en cit nc io t r ob ai al ow ur du a ch s sit nm ch pa rta iv rie in e d ke Gl at Gr R po ca po e m ro g lat ar b. a. u d. ve i e al er im nv tin e iv or . R v st iti at se ic of lu t h ov pe om m es i. C ls nn m ve sin n Co co Le j. I pp Bu E g. f. su c. e. Developing economy Transition economy Advanced economy Source: Ketels, Lindqvist, and Sölvell 2006, 33. © Christian Ketels, Göran Lindqvist, and Örjan Sölvell. Used with permission; further permission required for reuse. Note: Facilitators from the cluster initiatives (CIs) were asked about the strength of their clusters measured along many different dimensions. The sets of bars show responses to survey questions, indicating ratings on a scale of 1 (weak) to 7 (strong). (a) How strong is the cluster based on rivalry (i.e., who are the major players in your industry?)? (b) How strong is the cluster based on global market reach (sales to global markets)? (c) How strong is the cluster based on its economic importance (to the nation)? (d) How strong is the cluster based on its growth performance? (e) How strong is the cluster based on the business environment? (f ) How strong is the cluster based on levels of the value chain? (g) How strong is the cluster based on its competitive position? (h) How strong is the cluster based on the range of related and supporting industries? (i) How strong is the cluster based on its level of maturity? (j) How strong is the cluster based on its innovative capacity?� Figure 5.11 Cluster Strength by Initiator—Developing and Transition Countries a. Developing economies b. Transition economies 7 7 6 6 Average reply Average reply 5 5 4 4 3 3 2 2 1 1 sit ive ch lue pa ive nc c re ket en s fir of nm es rta mi n y e h n t po etit ca vat va ro in ar s cit r ai po no ac io be m vi us m of no p im Eco m m en g. B al ls In Nu Co ob ve c. e. Gl Le a. f. d. b. Business Government Donor Source: Ketels, Lindqvist, and Sölvell 2006. © Christian Ketels, Göran Lindqvist, and Örjan Sölvell. Used with permission; further permission required for reuse. Note: Cluster facilitators were asked about the strength of their clusters measured along many different dimensions. Results were compiled into three categories based on whether the cluster initiatives (CIs) came from the business community, the government, or the donor. The sets of bars show the responses from business, government, and donor CIs from developing countries (figure on left-hand side) and transition economies (figure on right-hand side) to survey questions, indicating ratings on a scale of 1 (weak) to 7 (strong). (a) How strong is the cluster based on the number of firms? (b) How strong is the cluster based on the number of levels of value chain? (c) How strong is the cluster based on its economic importance (to the nation)? (d) How strong is the cluster based on its global market reach? (e) How strong is the cluster based on its innovative capacity? (f ) How strong is the cluster based on its competitive position? (g) How strong is the cluster based on the business environment? Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Competitiveness and Clusters 123 Donors in developing countries target clusters that have fewer firms and fewer levels of the value chain and that are less economically important to the nation as a whole. In transition countries, donors target clusters that have less global market reach, less innovative capacity, weaker competitive position, and a less favorable business environment. The reasons for these findings are unclear; they may relate to the environment in which the donor operates, for example. The principle aim of CIs is to address barriers to competition. These barriers may arise from shortcomings in the business, government, or education sectors (Ketels, Lindqvist, and Sölvell 2006). These sectors are represented in CIs in vari- ous degrees. For ­ example, government often plays a dominant role in advanced economies; see ­ figure 5.12. This is less true of developing and transition econo- mies, where the business sector and donors take the lead. Part of the reason may well be that the capacity of government to launch CIs is weak in these econo- mies, so that donors step in to fill this void. Ketels, Lindqvist, and Sölvell (2006) further examine what occurs after the initiation of clusters—does government step back and allow the business ­ sector to take over or does it remain heavily involved? The scenario for all three types of economy—developing, transition, and advanced—is similar: “Government influence decreases over time while business becomes more important� (Ketels, Lindqvist, and Sölvell 2006, 36). Figure 5.13 shows which sector—­ government, business, or donor—was the most influential in determining which initial activities to undertake. Among advanced economies, government was the pri- mary initiator and remained dominant in selecting the initial participants in the CI; but it then transferred out when the time came to decide initial activities. This was deemed a good pattern by Ketels, Lindqvist, and Sölvell (2006). The decreasing role of government was also seen in developing and transition economies. The situation with donors was a little different; donors remained Figure 5.12 Entity Responsible for Initiating CI by Economy’s Underlying Level of Development 100 Share of respondents (%) 75 50 25 0 a. Developing b. Transition c. Advanced Business Government Donor Other Source: Ketels, Lindqvist, and Sölvell 2006. © Christian Ketels, Göran Lindqvist, and Örjan Sölvell. Used with permission; further permission required for reuse. Note: The sets of bars show the percentage of cluster initiatives (CIs) by initiator (business, government, donor, or other) classified by type of economy: (a) developing, (b) transition, and (c) advanced. Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 124 Competitiveness and Clusters Figure 5.13 Influence in First Stage of Cluster Initiatives’ Operation, by Sector and Economy Type 100 Share of respondents (%) 75 50 25 0 CI an ial iti al e an ial iti al e an ial iti al at at tiv iti tiv iti tiv iti cip it cip it cip it e ts es CI ts es CI ts es iti iti rti t in ac e in rti t in ac e in rti t in ac e in at In In iti pa lec pa lec pa lec cid cid cid a. a. In Se Se Se a. De De De b. b. b. c. c. c. Developing Transition Advanced Business Government Donor Source: Ketels, Lindqvist, and Sölvell 2006, 36. © Christian Ketels, Göran Lindqvist, and Örjan Sölvell. Used with permission; further permission required for reuse. Note: Percentages do not sum to 100 because data for “other actors� are not shown. The sets of bars show the percentage of cluster initiatives (CIs) by initiator (business, government, donor, or other) classified by the initiator’s decisions in the early stages to (a) initiate the CI, (b) select initial participants, and (c) decide initial activities and by each type of economy; developing, transition, and. advanced. heavily involved in the initial stage of CIs, though “in the longer run, . . . donor- initiated CIs appear to allow as much business sector influence as government, or even more� (Ketels, Lindqvist, and Sölvell 2006, 36). The challenge for donors in the short run is to also address the weaknesses in local and regional government institutions for which they are ­ compensating by becoming involved in CIs. Donors fulfill the role of government in transition and develop- ing economies—but there are limits to their involvement; they fail to address the underlying weaknesses in the business environment, and they are often influenced by their need to provide measurable results in a short time, often as short as 3 years. CIs should be used when long-term competitiveness is the goal and not short-term results such as increased employment or exports (Ketels, Lindqvist, and Sölvell 2006, 5). Policy Implications World Bank (2009) outlines how the existence of clusters helps to guide policy makers in forming policies for competitiveness. For example, government involvement in clusters can help identify barriers to competition and put in place policies to address these. Specialized inputs and skills are easier to access and are cheaper when firms are organized in a cluster setting, and public information and knowledge are more readily disseminated. Quasi-public goods such as infrastruc- ture, educational programs, and trade fairs are easier for government to oversee at the cluster level compared to the macro or regional level. Finally, clusters perform a useful search function in identifying those firms and industries that are not performing as well as others, thus providing valuable information for government and the business sector. ­ Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Competitiveness and Clusters 125 The many factors affecting competitiveness arise from the macroeconomy in general and the microeconomic environment of the firm specifically. Forces affecting these factors are constantly changing, rendering competitiveness a dynamic concept. As such, policy for competitiveness is multifaceted. Macroeconomic policy alone, while necessary, is not sufficient for improving competitiveness. Macro policy needs to be aware of its effect on how firms and markets operate. Furthermore, globalization has changed the way in which firms and markets operate, with greater technological absorption by firms and global integration by markets (World Bank 2009). On the micro- economic side, competitiveness is no longer primarily associated with price and cost but includes “connectivity, standards and certifications, quality and innovation, exploitation of cultural and geographic endowments, success of branding, etc.� (World Bank 2009, 67). Policy for competitiveness needs to be mindful of these changes and engage agents/institutions at many levels and from both public and private sector backgrounds. Being able to move forward with competitiveness policies to improve competitiveness requires a solid regime “to ensure that resources flow to the industrial clusters that have the best comparative advantage, and within those, to the firms that are economi- cally most efficient� (World Bank 2009, 68). Each country will have its own issues that impact negatively on the competi- tiveness of its firms and economy. The World Economic Forum’s Global Competitiveness Report (Porter et al. 2008) outlines the barriers to competitive- ness at the country level. The stage of development has implications for the 12 pillars of competitiveness outlined in the report. The Global Competitiveness Index highlights three stages of development—factor-driven economies, efficiency-driven economies, and innovation-driven economies. At the ­ high-income level, ­ ­ companies must compete by producing new and innovative products. At the middle-income level, firms tend to concentrate on fairly sophis- ticated interventions and the formation of firm-level and supply chain strategies. At the lower-income country level, efforts may concentrate on improving market and government imperfections in factor markets and demand conditions. ­ Despite this, there is no road map, and the respective roles of the private and public sector in formulating policy for competitiveness are unclear. Imperfect information characterizes both the government and the business sector. The formation of CIs provides a valuable resource for policy makers in helping to identify the barriers to competitiveness. The involvement of both public and private sector actors renders the CI a fertile place in which to identify policy issues, which can then be brought to the attention of policy makers. Porter’s diamond analysis provides a framework for identifying policy reforms, while technical tools such as “value-chain analysis, market-trend analysis, and competi- tiveness positioning analyses can ascertain operational efficiency of such reforms� (World Bank 2009, 77). Finally, the CI provides a forum for the development of a detailed policy map with suggestions on the most effective way to implement reforms. Table 5.6 shows the possible policy and strategic recommendations that may emanate from a CI. Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 126 Competitiveness and Clusters Table 5.6 Possible Policy and Strategic Recommendations from a Cluster Initiative Private sector business strategy Public policy recommendations recommendations Cluster-specific • Remove entry/exit barriers in industries related to the • Identify new product and market cluster segments and develop business • Remove regulatory burdens that prevent firms from strategies for increased outreach functioning efficiently • Shop floor enhancements of technology • Develop institutions that cater to the collective R&D and management for higher productivity needs of firms in the cluster • Improve the capacity of specialized input • Develop institutions that offer specialized skills for and service providers competitiveness • Market research • One-stop shop for dissemination of public information • Promotion of specific products in the local, on products and markets regional, and international markets • Facilitate export promotion and FDI attraction • Develop semiprivate institutions such • Develop provisions for basic provisions such as land, as business associations, research and labor, and capital as well as advanced factors such advisory centers, knowledge transfer as skilled labor, technology and equipment, faster/ centers, etc. cheaper transportation, etc. Economy-wide • Restructure the incentive regime and set up • Increase private sector investments in performance measurement systems as necessary infrastructure and services • Develop basic infrastructure necessary for industries to • Strengthen private sector capacity to function smooth and sophisticate the overall • Develop sound institutions that contribute to the supply chain capitalization of natural and socioeconomic endowments • Develop strong, competitive institutions • Develop strong human capital for training and R&D • Expedite overall regulatory reform Source: World Bank 2009. Note: FDI = Foreign Direct Investment; R&D = research and development. Conclusion The chapter discussed clusters and competitiveness by focusing initially on the definition of and background to the cluster concept, highlighting its application to competitiveness, and then examining the role for government in promoting and developing CIs. The chapter concluded with a discussion of the policy impli- cations of clusters for competitiveness. Clusters are a means of stimulating economic development at the local, regional, and global level. They play an important role in the modern economy and its search for competitiveness. Clusters arise at many different levels and for many different areas of economic activity. The chapter focused on industrial clusters and looked at the advantages of these for competitiveness; it also ­ presented examples of clusters. It looked at the implications of industrial clusters for both the sector and the geographical area. The former analysis focused on the links between firm strategy, structure, and rivalry; input factor conditions; demand conditions; and the presence of related supportive industries. The cluster concept spans the local, regional, national, and international arena, and the chap- ter looked at these links across geographic space. A key factor in industrial ­ clusters was innovation. The second part of the chapter examined CIs, which are organized efforts to increase growth and competitiveness within a region and are also a tool for government in pursuing policy reform. Following an example of an approach to ­ Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Competitiveness and Clusters 127 developing a CI, the chapter presented the characteristics of successful CIs. CIs were compared based on the level of economic development of the underly- ing economy. Differences were seen in objectives, target industries, cluster strength, the types of initiator (donor or government), and their influence. Finally, the chapter noted that the existence of clusters helps to guide policy makers in forming policies for competitiveness. The chapter concluded with a description of public policy and private sector implications of CIs. Notes ­ nowledge 1. See Nallari, Griffith, and Yusuf (2012) for a discussion of creative cities and k cities. 2. Alfred Marshall (1920) suggested a threefold classification of the reasons for industrial concentration nearly a century ago. In Nallari, Griffith, and Yusuf’s (2012, 8) paraphrase, he suggested that concentration arises because of “(a) knowledge spill- ­ overs, (b) the advantages of thick markets for specialized skills, and (c) the backward and forward linkages associated with large local markets.� Note also that “what Porter called ‘clusters’ have been labeled by economic geogra- phers variously as: ‘industrial districts’, ‘new industrial spaces’, ‘regional industrial complexes’, or, ‘innovative milieux’—to name but a few. The exact terminology depends on particular theoretical perspectives or research interests.� Local Government Association, “Industrial Clusters and Their Implications for Local Economic Policy.� http://www.idea.gov.uk/idk/core/page.do?pageId=8507296#contents-1. 3. Local Government Association, “Industrial Clusters and Their Implications for Local Economic Policy.� http://www.idea.gov.uk/idk/core/page.do?pageId=8507296#​ contents-1. 4. Firms of a similar type might support trade or professional associations that may help in disseminating best practice and lead to an upgrading of skills. 5. The Global Cluster Initiative Survey (GCIS) 2003 “identified more than 500 cluster initiatives around the world, primarily in Europe, North America, New Zealand and Australia. 238 completed the on-line survey, representing a broad range of technology areas� (Sölvell, Lindqvist, and Ketels 2003, 10). The survey covered the (1) setting, (2) objectives, (3) process, and (4) performance of the cluster initiatives. 6. The 10 tools are (1) cluster mapping; (2) product and market segmentation; (3) SWOT (strengths, weaknesses, opportunities, and threats); (4) Gap analysis comparing actual performance with potential performance); (5) Porter’s five forces (­ analysis; (6) value chain analysis; (7) market trends analysis; (8) competitive position- ing analysis; (9) old and new institutions for collaboration; and (10) monitoring and evaluation (World Bank 2009). 7. In 2008, Mills, Reynolds, and Reamer said that the number of cluster initiatives (CIs) had “expanded significantly in the last five years� and referred to the “several hundred distinct cluster initiatives� in the United States (14). They identified the following specific initiatives: Cleveland’s WIRE-net; the St. Louis BioBelt; Florida’s Technology Coast Manufacturing and Engineering Network; Southeast Michigan’s Automation Alley; Oregon Metals Initiative; and the Massachusetts Life Sciences Collaborative. 8. “Key factors include a high level of company trust in government initiatives and hav- ing influential local government decision makers, which are both clearly related to good Cluster Initiative performance� (Ketels, Lindqvist, and Sölvell 2008, 7). Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 128 Competitiveness and Clusters 9. “CIs for clusters in areas designated by government as attractive perform significantly better in attracting new firms� (Ketels, Lindqvist, and Sölvell 2008, 7). 10. “For promoting cluster growth, establishing an exchange with other clusters in the same industry is beneficial� (Ketels, Lindqvist, and Sölvell 2008, 7). 11. “In developing and transition economies, there is usually less trust among companies and between companies and government than in advanced economies� (Ketels, Lindqvist, and Sölvell 2006, 6). 12. Advanced economies are all countries that fall outside the developing and transition classifications, or as Ketels, Lindqvist, and Sölvell (2006, 10) suggest: they are “high- income economies (OECD or non-OECD) which are not transition economies.� References Chinitz, B. 1961. “Contrasts in Agglomeration: New York and Pittsburgh.� American Economic Review 51: 279–89. Cortright, J. 2006. “Making Sense of Clusters: Regional Competitiveness and Economic Development.� Discussion paper, Metropolitan Policy Program, Brookings Institution. http://www.brookings.edu/~/media/Files/rc/reports/2006/03cities_­c ortright/​ 20060313_Clusters.pdf. Fujita, M., P. Krugman, and A. Venables. 1999. The Spatial Economy: Cities, Regions and International Trade. Cambridge, MA: MIT Press. Glaeser, E. L., H. D. Kallal, J. A. Scheinkman, and A. Shleifer. 1992. “Growth in Cities.� Journal of Political Economy 100: 1126–52. Henderson, J. V. 1997. “Externalities and Industrial Development.� Journal of Urban Economics 42: 449–79. Hoover, E. M., and F. Giarratani. 1948. The Location of Economic Activity. New York: McGraw-Hill. Isard, W. 1956. Location and Space Economy. Cambridge, MA: MIT Press. Jacobs, J. 1969. The Economy of Cities. London: Penguin Books. Ketels, C., G. Lindqvist, and Ö. Sölvell. 2006. “Cluster Initiatives in Developing and Transition Economies.� Center for Strategy and Competitiveness, Stockholm School of Economics, Stockholm, Sweden. ———. 2008. “Clusters and Cluster Initiatives.� Center for Strategy and Competitiveness, Stockholm School of Economics, Stockholm, Sweden. http://www.europe-innova​ .eu/c/document_library/get_file?folderId=148901&name=DLFE-9310.pdf. Marshall, A. 1920. Principles of Economics. London: Macmillan. Mills, K. G., E. B. Reynolds, and A. Reamer. 2008. “Clusters and Competitiveness: A New Federal Role for Stimulating Regional Economies.� Metropolitan Policy Program, Brookings Institution, Washington, DC. Nallari, R., B. Griffith, and S. Yusuf. 2012. Geography of Growth: Spatial Economics and Competitiveness. Washington, DC: World Bank. Piore, M., and C. Sabel. 1984. The Second Industrial Divide. New York: Basic Books. Porter, M. E. 1990. The Competitive Advantage of Nations. New York: Free Press. ———. 2001. “The Microeconomics of Development.� Paper presented at the conference Competitiveness and Development: Vision and Priorities for Action, Caracas, Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Competitiveness and Clusters 129 June 20–21. http://www.cid.harvard.edu/archive/andes/documents/presentations/ caracas_0601/porter_competitivenessforum_062101.pdf. ———. 2008. “Clusters, Innovation, and Competitiveness: Findings and Implications for Policy.� Paper prepared for European Presidency Conference on Innovation and Clusters, Stockholm, Sweden, January 23. Porter, M. E., M. Delgado, C. Ketels, and S. Stern. 2008. “Moving to a New Global Competitiveness Index.� In The Global Competitiveness Report 2008–2009, 43–63. Geneva: World Economic Forum; Stockholm, Sweden: Ivory Tower AB. Rosenfeld, S. 1997. “Bringing Clusters into the Mainstream of Economic Development.� European Planning Studies 5 (1): 3–23. Sölvell, Ö., G. Lindqvist, and C. Ketels. 2003. The Cluster Initiative Greenbook. Stockholm, Sweden: Ivory Tower AB. World Bank. 2009. Clusters for Competitiveness: A Practical Guide and Policy Implications for Developing Cluster Initiatives. Washington, DC: World Bank. Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Environmental Benefits Statement The World Bank is committed to reducing its environmental footprint. In sup- port of this commitment, the Office of the Publisher leverages electronic pub- lishing options and print-on-demand technology, which is located in regional hubs worldwide. Together, these initiatives enable print runs to be lowered and shipping distances decreased, resulting in reduced paper consumption, chemical use, greenhouse gas emissions, and waste. The Office of the Publisher follows the recommended standards for paper use set by the Green Press Initiative. Whenever possible, books are printed on 50% to 100% postconsumer recycled paper, and at least 50% of the fiber in our book paper is either unbleached or bleached using Totally Chlorine Free (TCF), Processed Chlorine Free (PCF), or Enhanced Elemental Chlorine Free (EECF) processes. More information about the Bank’s environmental philosophy can be found at http://crinfo.worldbank.org/crinfo/environmental_responsibility/index.html. Clusters of Competitiveness  •  http://dx.doi.org/10.1596/978-1-4648-0049-8 Why do some developing countries grow faster than others? One answer lies in competition, which encourages innovation to spur growth. Competition occurs at diverse levels within a country and among countries—firms, industries, regions, nations, and the global economy. The recent global financial crisis has sharpened the focus on the need for stronger innovation and competitiveness in a growth-challenged world. Countries that develop and implement policies to foster increasingly competitive environments will facilitate growth and improved standards of living for their citizens. Clusters of Competitiveness presents the ­ concepts underlying competitiveness across its many dimensions and the price and nonprice measures of competitiveness. The primer also discusses experiences of—and lessons learned for—developing economies. Finally, the book focuses on clusters as a means of stimulating economic development at the local, regional, and global levels. As systems of interconnection between private and public sector entities, clusters play an important role in the modern economy and its search for competitiveness. This primer examines in detail the essential role that clusters play in guiding the formation of competition policies that promote growth. ISBN 978-1-4648-0049-8 SKU 210049