EUROPE AND CENTRAL ASIA STUDIES CRITICAL CONNECTIONS Promoting Economic Growth and Resilience in Europe and Central Asia OVERVIEW David Michael Gould Critical Connections Promoting Economic Growth and Resilience in Europe and Central Asia This booklet contains the overview, as well as a list of contents, from Critical Connections: Promoting Economic Growth and Resilience in Europe and Central Asia, doi: 10.1596/978-1-4648-1157-9. A PDF of the final, full-length book, once published, will be available at https://openknowledge.worldbank​ .org/, and print copies can be ordered at http://Amazon.com. Please use the final version of the book for citation, reproduction, and adaptation purposes. © 2018 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 This work is a product of the staff of The World Bank with external contributions. The findings, interpreta- tions, 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 IGO license (CC BY 3.0 IGO) http://creative​ commons.org/licenses/by/3.0/igo. 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: David Michael Gould. 2018. Critical Connections: Promoting Economic Growth and Resilience in Europe and Central Asia. Overview booklet. World Bank, Washington, DC. License: Creative Commons Attribution CC BY 3.0 IGO. Translations—If you create a translation of this work, please add the following disclaimer along with the attri- bution: 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. Adaptations—If you create an adaptation of this work, please add the following disclaimer along with the attribution: This is an adaptation of an original work by The World Bank. Views and opinions expressed in the adaptation are the sole responsibility of the author or authors of the adaptation and are not endorsed by The World Bank. Third-party content—The World Bank does not necessarily own each component of the content contained within the work. The World Bank therefore does not warrant that the use of any third-party-owned individual component or part contained in the work will not infringe on the rights of those third parties. The risk of claims resulting from such infringement rests solely with you. If you wish to re-use a component of the work, it is your responsibility to determine whether permission is needed for that re-use and to obtain permission from the copyright owner. Examples of components can include, but are not limited to, tables, figures, or images. All queries on rights and licenses should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; e-mail: pubrights@worldbank.org. Cover and interior design: Debra Naylor, Naylor Design. Cover images: inset: © sigdem/shutterstock; background image: © VTT Studio/shutterstock. All images used with permission. Further permission required for reuse. Contents Contents of the Full Book . . . . . . . . . . . . . . . . . . . . . . . . . . . . v About the Authors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix Foreword . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xv Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xvii Abbreviations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xix Regional Classifications Used in This Report . . . . . . . . . . . xxiii Critical Connections. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Main Findings of Critical Connections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Multidimensional Connectivity a Key to Europe and Central Asia’s Development and Growth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Europe and Central Asia Connectivity Is a Critical Source of Knowledge Transfers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 Foreign-Owned and -Managed Firms Tend to Perform Better and Contribute to Local Firms’ Productivity . . . . . . . . . . . . . . . . . . . . . . . . . . 15 Economic Migration Has Been Beneficial to Europe and Central Asia . . . . 22 Strong Infrastructure Transport Links Provide Important Support for Connectivity. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 The Growth of Supply Chains Reflects Greater Connectivity and Has Facilitated Increased International Knowledge Flows . . . . . . . . . . . . . . . 31 European and Central Asian Countries Have Moved toward More Open Policies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 Considerable Scope Remains for Improving Policies to Increase Connectivity in Europe and Central Asia. . . . . . . . . . . . . . . . . . 38 Annex A.  Selected Indicators. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46 iii Contents of the Full Book About the Authors Foreword Acknowledgments Abbreviations Regional Classifications Used in This Report Critical Connections Overview Multidimensional Connectivity: Pathways to Growth 1  and Shared Prosperity in Europe and Central Asia Main Messages Introduction Trends in Economic Connectivity Connectivity and Income Growth Trade-Offs and Resilience to Shocks Conclusion Annex 1A. Data Annex 1B. Network Graph Methodology Annex 1C. Multiplex PageRank Centrality Annex 1D. Centrality Indicator Notes References Spotlight 1: Trends in Foreign Direct Investment in Europe and Central Asia Notes Reference Knowledge Transfers from International Openness 2  in Trade and Investment: The European Case Main Messages Knowledge Creation in Europe Knowledge and Learning from Trade, Investment, and GVCs: Insights from the Economic Literature v vi  ●   Critical Connections: Promoting Economic Growth and Resilience in Europe and Central Asia Knowledge Diffusion in Europe: The Two-Stage Process of Technology Transfer Conclusion Notes References Spotlight 2: Attracting Foreign Direct Investment: The Role of Deep Preferential Trade Agreements Deep PTAs in ECA: A Snapshot Linking Deep Agreements with FDI: Empirical Strategy Linking Deep Agreements with FDI: Results Annex S2A. Definition of Country Groups Notes References 3 Connectivity and Firms Main Messages Firm Connectivity in ECA Annex 3A. Coverage of Orbis Data Annex 3B. Additional Tables Notes References Spotlight 3: Reaping Digital Dividends through Complementary Investments Notes References 4 Migration and Connectivity Main Messages Migration Patterns in Europe and Central Asia Migration Patterns in ECA Are Likely to Change Policies Should Aim to Improve the Integration of Migrants Emigration Generates Net Benefits in ECA Origin Countries Conclusion Annex 4A. Gravity Model Annex 4B. Additional Tables and Figures Notes References 5 Infrastructure Linkages: Cost, Time, and Networks Main Messages Connecting Cities and Neighbors: A Vicinity View of Transport Services in ECA From First Neighbors to Transport Networks: Connectivity as a Policy Objective Contents of the Full Book ● vii Connectivity as a Collective Challenge: Centrality and Criticality Conclusion Annex 5A. Methodology and Data Notes References Supply Chains in Europe and Central Asia: Connectivity 6  through Cross-Border Production Fragmentation Main Messages Factory Europe Are There Only Benefits from Increased Interdependence of Countries? Different Policies for GVC Upgrading Conclusion Annex 6A. Elasticities of Value Added in Exports, Gross Exports, and Fragmentation Intensity Annex 6B. Interdependence of Countries Annex 6C. Regression of Backward and Forward Participation Indexes over a Set of Policy Variables Notes References Trends in Connectivity-Related Policy Indicators 7  for Europe and Central Asia Main Messages Introduction MFN Tariffs Foreign Direct Investment Policies Preferential Trade Agreements Bilateral Investment Agreements Product Market Regulation EBRD Transition Indicators Policy Comovements—Are Policies Consistent? Conclusion Notes References About the Authors David M. Gould (World Bank) is currently Lead Economist in the World Bank’s Europe and Central Asia Region and the lead author of the ECA Critical Connections flagship. He is the author of several books and peer-reviewed journal articles on international trade and finance, migration, and economic policy. Currently, he is leading Europe and Central Asia regional studies on the development impact of disruptive technologies. During his 15 years at the World Bank, he has led teams to deliver country development strategies and analytical and lending operations in Europe, Latin America, and South Asia. Prior to joining the World Bank, he served as the Director of Global Economic Analysis at the Institute of International Finance and as Senior Economist and Policy Advisor at the US Federal Reserve. He has held visiting research positions at the Central Banks of Mexico and Chile. He holds a PhD in International Economics (with honors) from the University of California at Los Angeles and is a Chartered Financial Analyst charter holder. Megersa Abate (World Bank) is a Transport Economist in the World Bank’s Transport and Digital Development Global Practice. He has extensive expertise and interest in various topics, including freight demand modeling, air transport regulation, and transport connectivity. His research has been published in leading transportation economics journals. Before joining the Bank in 2016, he worked as a researcher at VTI, the Swedish National Road and Transport Research Institute, and at the VU University of Amsterdam. Earlier in his career, he worked at the Ethiopian Civil Aviation Authority as an air transport expert. He received his PhD in Transport Economics from the Technical University of Denmark in 2013; during his PhD studies, he was also a visiting student in the Institute of Transport Studies at the University of Leeds. Erhan Artuc (World Bank) is a Senior Economist in the World Bank’s Development Research Group. Prior to joining the World Bank in 2011, he was a faculty member at Koç University in Istanbul, Turkey. His most recent research focuses on interna- tional trade and migration policies and their effects on labor markets and development. His work has been published in leading academic and policy journals such as the Journal of International Economics, Economic Journal, and American Economic Review. He received his undergraduate degree from Bilkent University and a PhD in Economics from the University of Virginia. Omar Bamieh (University of Vienna) is an Assistant Professor of Economics at the University of Vienna. His research focuses on labor market institutions and the ix x  ●   Critical Connections: Promoting Economic Growth and Resilience in Europe and Central Asia interaction between legal frameworks and labor markets. He holds an MS in Economics and Social Sciences from Bocconi University in Milan and a PhD in Economics from the European University Institute in Florence, where he also worked as a Research Fellow in Global Economics at the Robert Schuman Centre for Advanced Studies. Cecilia Briceno-Garmendia (World Bank) is a Lead Economist in the World Bank’s Transport and ICT Global Practice, where she leads the economic research agenda in logistics and transport in the Latin America region and globally advises teams and governments in strategic issues pertaining to prioritization and planning of infrastructure investments, including aspects related to spending efficiency, green trucking and trucking sector performance, multimodal development corridors, and adaptation of transport networks to climate change. Previously, she led the economic team in the Office of the Director for Sustainable Development of the Latin America region, where she provided leadership for the analytical agenda on infrastructure, urban, and disaster risk management and climate change adapta- tion issues. She has worked extensively on issues pertaining to connectivity, logis- tics, and port performance and co-led the pathbreaking Africa Infrastructure Country Diagnostic. She has worked on projects and research in more than 70 countries. Before joining the World Bank, she worked in software engineering and the design of information and organizational systems for both private and public sector enterprises in República Bolivariana de Venezuela. She has an MBA from the Instituto de Estudios Superiores en Administración in Caracas, República Bolivariana de Venezuela, and a doctorate in Economics from Georgetown University. Matteo Fiorini (European University Institute) is a Research Fellow in Global Economics at the Robert Schuman Centre for Advanced Studies of the European University Institute in Florence, Italy. His research focuses on international trade, trade policy, migration, and development. Prior to joining the Schuman Centre, he worked as a researcher at the Institute’s Migration Policy Centre, the World Trade Organization, and Bocconi University. He holds an MS in Economics and Social Sciences from Bocconi University in Milan and a PhD in Economics from the European University Institute in Florence. Bernard Hoekman (European University Institute) is a Professor at the Robert Schuman Centre for Advanced Studies of the European University Institute. He is also a Research Fellow at the Centre for Economic Policy Research, a member of the World Economic Forum Council on Trade and Investment, and a Senior Fellow at the Centre for International Governance Innovation. His research focuses on trade and development, economic integration, and the multilateral trading system. Dror Y. Kenett (World Bank) is a multidisciplinary financial economist and an expert on financial networks, financial stability, and systemic risk. He is a consultant to the World Bank, Adjunct Professor at Johns Hopkins University, a research asso- ciate at the London School of Economics Systemic Risk Centre, and a visiting researcher at Boston University and at the Israel Securities Authority. He has also About the Authors ● xi held a researcher position in the US Department of the Treasury’s Office of Financial Research. He applies his scientific background to financial stability questions, focusing on network-based models, market structure, financial contagion and spill- overs, and correlation-based models. He has extensive policy experience and has contributed to the Office of Financial Research Financial Stability Report and par- ticipated in the development of the Office’s monitoring tools. He has published more than 40 papers in financial, physics, and engineering journals, including the Journal of Banking and Finance, Journal of Risk and Financial Management, Quantitative Finance, Nature Physics, and Scientific Reports. He has a PhD in Physics from Tel Aviv University in Israel. Mathilde Lebrand (World Bank) is an Economist in the World Bank’s Transport Global Practice. Currently she is working on the Belt and Road Initiative, economic corridor development, and connectivity. Previously she worked for the Europe and Central Asia Chief Economist office and contributed to several upcoming regional studies. Her research focuses on economic geography, international trade and global value chains, networks, and political economy. She has taught at the University of Montreal and has worked at the World Trade Organization in Geneva. She is a Research Fellow at the Center for Economic Studies ifo Institute (CESifo). She holds a PhD in economics from the European University Institute. Paloma López-Garcia (European Central Bank) has been a Senior Economist in the Directorate-General—Economics at the European Central Bank since 2015. Before that she was Coordinator of the Competitiveness Research Network (CompNet) in the Directorate-General—Research, and she has also worked at the Instituto de Empresa Business School and in the Research and Economics Department of the Central Bank of Spain. She has published articles in the European Economic Review, Small Business Economics, and Economics of Innovation and New Technology, among other peer-reviewed journals. Her research topics are microanalysis of productivity and employment growth, innova- tion, and trade and competitiveness. She earned her PhD at the London School of Economics in 2003. Çag ˘ lar Özden (World Bank) is a Lead Economist in the World Bank’s Research Department. His research explores the nexus of globalization of product and labor markets, government policies, and economic development. His current research projects explore the determinants and patterns of global labor mobil- ity; impacts of migrants on destination labor market outcomes; linkages between migration, trade, and foreign direct investment flows; medical brain drain; and linkages between aging and global economic integration. He has edited three books and published numerous papers in leading academic journals such as the American Economic Review and the Economic Journal. He is a Fellow of the Institute of Labor Economics (IZA) and of the Centre for Research and Analysis of Migration (CreAM) and serves on the advisory board of the Economic Research Forum. He received his undergraduate degrees in Economics and Industrial Engineering from Cornell University and his PhD in Economics from Stanford University. xii  ●   Critical Connections: Promoting Economic Growth and Resilience in Europe and Central Asia Georgi Panterov (World Bank) is a Research Analyst in the World Bank’s Office of the Chief Economist for Europe and Central Asia. His research interests are focused on machine learning, econometrics, blockchain, and cryptocurrencies. During his time at the World Bank, he has contributed to the Golden Aging flag- ship report, the Critical Connections flagship report, and the Europe and Central Asia economic update reports. Before joining the World Bank, he worked at Google, the US Department of Agriculture, and American University. He is cur- rently a PhD candidate in Economics at American University in Washington, DC. Nadia Rocha (World Bank) is a Senior Economist in the World Bank’s Macroeconomics, Trade and Investment Global Practice. Prior to joining the Bank in 2016, she worked for five years in the World Trade Organization’s Economic Research and Statistics Division. She was seconded to the Colombian Ministry of Trade to serve as a Senior Advisor on Trade during 2015. Her current work focuses on regionalism, trade costs, global value chains, and trade and gender. She holds a BA in economics from Bocconi University in Milan, an MA in Economics from Pompeu Fabra University of Barcelona, and a PhD in International Economics from the Graduate Institute, Geneva. Daria Taglioni (World Bank) is the Principal Economist for the Europe and Central Asia and East Asia and Pacific Regions in the Economics and Private Sector Development Vice Presidency of the World Bank Group’s International Finance Corporation. Prior to joining the World Bank Group, Daria worked at the European Central Bank and at the Organisation for Economic Co-operation and Development. Her research focuses on trade and competitiveness. She has published articles in the American Economic Review and Journal of International Economics, among other peer-reviewed journals. She holds a PhD in International Economics from the Graduate Institute of Geneva. Shawn Tan (World Bank) is an Economist in the World Bank’s Finance, Competitiveness and Innovation Global Practice and is currently working on pri- vate sector development and trade issues in the countries of Eastern Europe and the Western Balkans. He has worked on reports such as the World Development Report 2016: Digital Dividends, Reaping Digital Dividends: Leveraging the Internet for Development in Europe and Central Asia, and the high-growth enterpreneur- ship report and has written papers on international trade, firm productivity, and high-growth firms. Before joining the World Bank, he worked at the Singapore Economic Development Board, where he was a negotiator for Singapore’s free trade agreements and bilateral investment treaties and worked on trade facilita- tion issues for multinational companies in Singapore. His research interests are broadly in international trade, economic geography, and firm productivity and performance. He holds a PhD in Economics from the University of Melbourne. Gonzalo Varela (World Bank) is a Senior Economist in the Global Trade and Regional Integration Unit of the World Bank’s Macroeconomics, Trade and Investment Global Practice. Prior to joining the World Bank, he was a Lecturer at the University of Sussex and at Uruguay’s Ministry of Industry, Energy, and Mining. About the Authors ● xiii His work agenda focuses on global integration and economic performance and on the analysis of trade policy and competitiveness. He holds a BSc in Economics from the Universidad de la República in Uruguay and both an MA in International Economics and a PhD in Economics from the University of Sussex. Hernan Winkler (World Bank) is a Senior Economist in the World Bank’s Jobs Group. He specializes in applied microeconomics, with a particular focus on issues related to labor markets, technological change, and the sources and conse- quences of poverty and inequality. His research has been published in peer- reviewed economics journals, including the Review of Economics and Statistics and the Journal of Development Economics. He was a lead author of the World Bank regional report Reaping Digital Dividends: Leveraging the Internet for Development in Europe and Central Asia. He has been part of the core teams of several regional reports, including Diversified Development, Golden Aging, and Risk and Returns. He was previously a researcher at the Center for Distributive, Labor and Social Studies (CEDLAS) at the National University of La Plata in Argentina, where he conducted research on poverty and distributional issues affecting countries in Latin America and the Caribbean. He holds a master’s degree in Economics from the National University of La Plata and a PhD in Economics from the University of California at Los Angeles. Thea Yde-Jensen (World Bank) is a Researcher in the World Bank Group’s Poverty and Equity Global Practice, where she conducts research on issues related to liveli- hoods, labor market outcomes, and displacement. Her expertise and research interests particularly focus on examining the interlinkages of labor markets and inequality and poverty. Previously she worked as a Researcher in the Bank’s Office of the Chief Economist for Europe and Central Asia, focusing on issues related to employment and firms’ access to finance and international networks. Prior to join- ing the World Bank, she worked in the International Monetary Fund’s Statistics Department and in the Department of Economics at Copenhagen Business School. She has a BS and an MS in Economics from the University of Copenhagen. Foreword In mid-2014 when Critical Connections was first contemplated, the Europe and Central Asia (ECA) region was still emerging from the global financial crisis, growth was uncertain and tepid, and policy makers were largely focused on mitigating further financial and macroeconomic risks from ongoing weakness in the banking sector and large fiscal deficits. Appropriately, the policy discourse was largely targeted to shoring up near-term challenges, rather than on assembling the building blocks that would provide the foundation for restoring the promise of long-term resilient growth. Critical Connections was born out of the desire to help policy makers focus their attention on their long-term goals of regional and global integration to capture the benefits of connectivity, from which ECA countries had advanced so far during the early years of market expansion in the 1990s and early 2000s. What started simply as an exploration into policies to capture the gains of specialization and knowledge transfers has taken on much greater meaning in recent times. The trend toward regional and global integration is under serious threat as many voters, particularly in high-income countries, see nationalism and protection as a remedy to greater economic uncertainty. But as former UK Prime Minister Gordon Brown noted in a 2015 speech, “the problems that give rise to nationalism can’t be solved by nationalism and in an interdependent world the problems that give rise to isolationism and protectionism cannot be solved by isolationism and protectionism.” While Critical Connections does not provide answers to assuage all the concerns about our changing global economy, it does provide an invaluable insight into understanding—at the firm and country level—the interdependence of our world and how it has historically operated, and currently operates, to advance economic growth and shared prosperity. A key insight of this report is that ECA’s international connectivity through trade, foreign direct investment, migration, telecommunications, transportation, and other avenues facilitates the transfers of knowledge and technology that are critical to long-term growth and shared prosperity. These connections complement one another because of the tacit (learning by doing), rather than explicit (contained in books or blueprints), nature of knowledge transfers. Migration, for example, enhances knowledge spillovers through trade and foreign investment by migrants transferring information on foreign markets and supporting connections to them. Similarly, the internet and efficient transport links are both necessary for successful e-commerce. xv xvi  ●   Critical Connections: Promoting Economic Growth and Resilience in Europe and Central Asia Moreover, the depth of ECA’s connections and the geographic composition of the connections both matter. Knowledge transfers are greater from countries that themselves have strong links to third countries. These transfers also emerge from linkages between firms in global value chains as well as foreign ownership and management practices that generate local spillovers. While these connections are important for prosperity, however, one should not be naive about their impact. Despite its overall benefits, increased connectivity exposes ECA countries to shocks, particularly those emanating from countries at the center of international economic transactions, which may have contributed to economic insecurity. However, by providing alternative sources of external demand and financing, a broad range of connections can reduce those risks and help countries cope with both domestic and external shocks. The European side of the ECA region is an ideal laboratory for observing the role of multidimensional connectivity in action. Regional supply chains are strong, and links between countries across the various forms of connectivity allow observations on how connectivity opens doors for the knowledge transfers that support resilient growth. Nonetheless, in many European countries, progress on deepening connectivity has stalled since the global financial crisis, and productivity growth attributed to connectivity has suffered. In Central Asia, despite recent moves toward building greater interconnectedness, the region remains among the least connected globally. Because of both its geographical position and its limited infrastructure, many Central Asian countries are only weakly connected to other ECA countries and the global economy. The vast distances between Central Asia, Europe, and East Asia will remain an obstacle to connectivity. However, infrastructure investments and policies to improve integration through freer trade, infrastructure, and investment policies are likely to provide large growth benefits in Central Asia. Many ECA countries can be proud of what they have achieved in building greater connectivity and advancing development during the past 25 years. But because the economic benefits of connectivity through knowledge and technology transfers are not obvious, while the challenge of economic uncertainty is, building the case for deepening connections requires solid and clear evidence. By recognizing the challenges as well as making explicit the potential opportunities of greater connectivity through various channels, Critical Connections can assist ECA’s policy makers in building the foundations for deepening important connections in the coming decades. Cyril Muller Vice President Europe and Central Asia Region World Bank Group Acknowledgments This report was written by a team led by David Michael Gould, Lead Economist in the World Bank’s Office of the Chief Economist for Europe and Central Asia. The core team members were Erhan Artuc, Cecilia Briceno-Garmendia, Bernard Hoekman (European University Institute), Mathilde Lebrand, Çag ˘ lar Özden, Georgi Panterov, Nadia Rocha, William Shaw, Daria Taglioni, Shawn Tan, Ekaterina Ushakova, Gonzalo Varela, Hernan Winkler, and Thea Yde-Jensen. The work was carried out under the overall supervision and guidance of Hans Timmer, Chief Economist for the Europe and Central Asia Region. The Macroeconomics, Trade, and Investment team benefited from the guid- ance of Jose Guilherme Reis and comments and discussions with Jean Francois Arvis, Cordula Rastogi, and Daniel Saslavsky. The Transport team appreciates the general guidance of Juan Gaviria and comments and discussions with Baher El-Hifnawi, Carolina Monsalve, and other members of the Transport team. Many thanks go to Rashmi Shankar for her extremely helpful contributions during the early stages of the report and Moritz Meyer for his generous time and discussions on applied network analysis. Peer reviewers Luis-Felipe Lopez-Calva, Caroline Freund, Bill Maloney, Aaditya Mattoo, and Russell Hillberry provided very helpful advice and comments on the report. This report would not have been timely or relevant without the insights and inputs of European Central Bank staff members, who provided data, analysis, and technical support for the work on “Knowledge Transfers from International Openness in Trade and Investment: The European Case,” and staff members of the International Civil Aviation Organization, particularly Dr. Ananthanarayan Sainarayan and colleagues, who generously provided data on origin-to-­destination airline connectivity. Private and public sector organizations and experts in Romania and Moldova also provided significant inputs and insights; in Romania: Startnet, Softelligence, Fondul Proprietatea, the Foreign Investors Council, Oracle România, the Ministry of Transport, the Ministry of External Commerce, the Ministry of Communications and Information Society, UPC Romania, Dr. Adrian Curaj (UNESCO), AmCham, H. Essers, Robin Martens (International Project Management), Kuijken Logistics Group, and Autonom Rent-A-Car; in Moldova: Ionel, Andragrup SRL, the European Business Association, GIZ (German Cooperation for International Development), the Ministry of Transport and Roads Infrastructure, APIUS, Moldova Investment and Export Promotion Organization, the Ministry of Finance, Star Legal Consulting, AmCham, Danube Logistics, and the Bureau for Relations with the Diaspora. xvii xviii  ●   Critical Connections: Promoting Economic Growth and Resilience in Europe and Central Asia Many people participated in the writing of the report. The main authors and contributors were • Overview: David Michael Gould • Chapter 1: David Michael Gould, Dror Kenett, and Georgi Panterov (with con- tributions from Angel Bogoev [American University], Michael Danziger [Center for Complex Network Research and Department of Physics, Northeastern University], Dobrina Gogova, Xin Yuan [Boston University], and Tlek Zeinullayev [Harvard University]) • Chapter 2: Paloma López-Garcia (European Central Bank) and Daria Taglioni (with contributions from Francesco Chiacchio [European Central Bank], Alvaro Espitia, Katerina Gradeva [European Central Bank], Laura Gomez-Mera, Asier Mariscal [Carlos III University of Madrid], Nadia Rocha, and Gonzalo Varela) • Chapter 3: Shawn Tan, Hernan Winkler, and Thea Yde-Jensen • Chapter 4: Erhan Artuc and Çag ˘ lar Özden (with contributions from Gnanaraj Chellaraj, Julio Elias, David Michael Gould, Bingjie Hu, Zovanga Kone, Tu Chi Nguyen, and Michael Packard) • Chapter 5: Cecilia Briceno-Garmendia, Mathilde Lebrand, and Megersa Abate (with contributions from Rodrigo Archondo, Gözde Isik, and Tetyana Kuchma) • Chapter 6: Mathilde Lebrand • Chapter 7: Omar Bamieh (University of Vienna), Matteo Fiorini (European University Institute), and Bernard Hoekman (European University Institute) • Spotlights 1 and 2: Gonzalo Varela and Nadia Rocha • Spotlight 3: Hernan Winkler • Content and Technical Editing: William Shaw • Overview Editing: Richard Alm Ekaterina Ushakova oversaw the production and support of the report. Many thanks go to all the commentators and reviewers in the initial stages of the report, particularly Europe and Central Asia Country Directors and Managers and Cyril Muller, Vice President of the World Bank’s Europe and Central Asia Region. Abbreviations AMI average management index BIT bilateral investment treaty BRI Belt and Road Initiative BvD Bureau van Dijk CIS Commonwealth of Independent States CMEA Council for Mutual Economic Assistance DCFTA Deep and Comprehensive Free Trade Agreements EBRD European Bank for Reconstruction and Development ECA Europe and Central Asia EEA European Economic Association EEC European Economic Community EU European Union twenty-eight member countries EU13 The thirteen EU members that have joined the union since 2004 (Bulgaria, Croatia, Cyprus, the Czech Republic, Estonia, Hungary, Latvia, Lithuania, Malta, Poland, Romania, the Slovak Republic, and Slovenia). EU15 The fifteen original EU members (Austria, Belgium, Denmark, Finland, France, Germany, Greece, Ireland, Italy, Luxembourg, the Netherlands, Portugal, Spain, Sweden, and the United Kingdom). FDI foreign direct investment FDIRRI FDI Regulatory Restrictiveness Indicators GATS General Agreement on Trade in Services GDP gross domestic product GVC global value chain HIC high-income country ICT information and communication technologies LMIC lower-middle-income country MBI Mobility Barriers Index MDC multidimensional connectivity MFN most favored nation MIPEX Migrant Integration Policy Index MNE multinational enterprise NACE European Classification of Economic Activities NAFTA North American Free Trade Agreement NTM nontariff measures xix xx  ●   Critical Connections: Promoting Economic Growth and Resilience in Europe and Central Asia NUTS-3 nomenclature of territorial units for statistics (Nomenclature des Unités territoriales statistiques), level 3 OECD Organisation for Economic Co-operation and Development OLS ordinary least squares PMR product market regulation PPML Poisson pseudo–maximum likelihood PTA preferential trade agreement RTA revealed technology advantage SPS sanitary and phytosanitary TBT Technical Barriers to Trade TFP total factor productivity TiVA Trade in Value Added TRIMS Trade Related Investment Measures UMIC upper-middle-income country WDI World Development Indicators WEF World Economic Forum WITS World Integrated Trade Solution WMS World Management Survey WTO World Trade Organization Countries and Economies International Organization for Standardization three-letter country codes; italics designate countries in the Europe and Central Asia region AFG Afghanistan ALB Albania ARE United Arab Emirates ARG Argentina ARM Armenia ATG Antigua and Barbuda AUS Australia AUT Austria AZE Azerbaijan BEL Belgium BEN Benin BFA Burkina Faso BGD Bangladesh BGR Bulgaria BHS Bahamas, The BIH Bosnia and Herzegovina BLR Belarus BLZ Belize BRA Brazil BRB Barbados BWA Botswana CAN Canada Abbreviations ● xxi CHE Switzerland CHL Chile CHN China COL Colombia CMR Cameroon CRI Costa Rica CYP Cyprus CZE Czech Republic DEU Germany DNK Denmark DOM Dominican Republic DZA Algeria ECU Ecuador EGY Egypt, Arab Rep. ESP Spain EST Estonia ETH Ethiopia FIN Finland FRA France GAB Gabon GBR United Kingdom GEO Georgia GHA Ghana GRC Greece GUY Guyana HKG Hong Kong SAR, China HRV Croatia HUN Hungary IDN Indonesia IND India IRL Ireland ISL Iceland ISR Israel ITA Italy JAM Jamaica JOR Jordan JPN Japan KAZ Kazakhstan KEN Kenya KGZ Kyrgyz Republic KWT Kuwait LBN Lebanon LTU Lithuania LUX Luxembourg LVA Latvia MDA Moldova MEX Mexico xxii  ●   Critical Connections: Promoting Economic Growth and Resilience in Europe and Central Asia MKD Macedonia, FYR MLT Malta MNE Montenegro MOZ Mozambique MUS Mauritius MYS Malaysia NAM Namibia NGA Nigeria NIC Nicaragua NLD Netherlands NOR Norway NZL New Zealand OMN Oman PAK Pakistan POL Poland PRT Portugal PRY Paraguay ROU Romania RUS Russian Federation SAU Saudi Arabia SGP Singapore SRB Serbia SWE Sweden SVK Slovak Republic SVN Slovenia SWZ Swaziland THA Thailand TJK Tajikistan TKM Turkmenistan TTO Trinidad and Tobago TUR Turkey TZA Tanzania UKR Ukraine UZB Uzbekistan XKX Kosovo (not listed as an ISO standard country; the unofficial two- and three-digit codes are used by the European Commission and others until an ISO code is assigned) YUG Serbia and Montenegro (former Yugoslavia) ZAF South Africa ZMB Zambia Regional Classifications Used in This Report Europe and Central Asia Northern Europe Southern Europe Central Europe Western Europe Western Balkans Denmark Greece Bulgaria Austria Albania Estonia Italy Croatia Belgium Bosnia and Herzegovina Finland Portugal Czech Republic France Kosovo Latvia Spain Hungary Germany Macedonia, FYR Lithuania Cyprus Poland Ireland Montenegro Sweden Malta Romania Luxembourg Serbia Slovak Republic Netherlands Slovenia United Kingdom South Caucasus Central Asia Russian Federation Turkey Other Eastern Europe Armenia Kazakhstan Belarus Azerbaijan Kyrgyz Republic Moldova Georgia Tajikistan Ukraine Turkmenistan Uzbekistan xxiii Critical Connections The countries of the Europe and Central Asia (ECA) region, along with much of the rest of the world, find themselves engaged in a revival of one of the fundamental questions of economic policy: how much to open to the rest of the world. At the turn of the century, the issue seemed largely settled, and most nations viewed greater openness as a key component of the path to prosperity. In these heady days, the European Union (EU) deepened with a drive toward greater integration and expanded by incorporating nations transitioning to market-based economies. More recent events—most notably, the global financial crisis and the tough times that followed—sowed the seeds of doubts about the benefits of globalization, leading to a rise of protectionist and nationalist economic sentiments, exemplified by Britain’s referendum to withdraw from the EU. In 2018, how much to open to the rest of the world now dominates the political economy of the ECA region, not just within the advanced EU economies, but also among the emerging economies of the region. Deciding where to draw the line between openness and protection- ism has become a pivotal and divisive issue, often tinged with emotion. With this publication, the World Bank offers new research on the process of economic ­ integration, showing its potential benefits without ignoring the downsides. Main Findings of Critical Connections • The ECA region’s international connectivity through trade, foreign direct invest- ment (FDI), migration, telecommunications, transportation, and other avenues 1 2  ●   Critical Connections: Promoting Economic Growth and Resilience in Europe and Central Asia facilitates the transfers of knowledge and technology that are critical to long- run growth and shared prosperity. These connections complement each other. For example, migration encourages trade and foreign investment by providing knowledge spillovers between host and home country markets and supporting connections to them. Similarly, the internet and efficient transport links are both necessary for successful e-commerce. Therefore, a balanced approach to increasing all dimensions of connectivity is desirable. • The depth of overall connections and the geographic composition of the connections both matter. Knowledge transfers are greater from countries that themselves have strong links to third countries. These transfers also emerge from firm linkages in global value chains as well as foreign ownership and management that generate local spillovers. • Deep integration of countries into the EU along many dimensions has gener- ated important benefits to growth through knowledge transfers. Central Asia, the South Caucasus, and the Western Balkans have benefited from regional connections as well, but the gains have been less pronounced. Much of the difference is due to the lack of direct and indirect connectivity to the wider global economy in the eastern part of ECA. • Despite its overall benefits, increased connectivity has encountered opposition—most notably, Britain’s June 2016 vote to exit the EU. National challenges often contribute to the backlash, but increased connectivity can expose ECA countries to external shocks, particularly those emanating from countries at the center of international economic transactions. By providing alternative sources of external demand and financing, however, a broad range of connections can reduce those risks and help countries cope with both domestic and external shocks. Introduction The ECA region has a rich history of regional integration and connectivity to the broader world economy, which has stimulated the growth of knowledge and tech- nological innovation. Indeed, through migration, trade, investments, and other interactions, ECA countries have depended on, and benefited from, connectivity with other countries for centuries. The Silk Road, formally established during China’s Han Dynasty in the second century BCE, facilitated more than the exchange of commercial goods. It was also a conduit for art, religion, philosophy, technology, language, science, and architecture (Starr 2015). Similarly, the Age of Discovery (1453–1660 CE) led to the deepening of a global community that was associated with profound advances in commerce and culture. As new navigation technology made sailing long distances possible, Europeans took to the seas to forge direct trading relationships with China, Indonesia, and Japan. Historians contend that it was the spice trade that fueled the development of faster boats, encouraged the discovery of new lands, and fostered new diplomatic relationships between East and West (Parthesius 2010; Bernstein 2013). In recent times, the most prominent feature of ECA connectivity has been regional integration through the gradual expansion of what is now the EU. The 1951 Critical Connections ● 3 European Coal and Steel Community, a sectoral integration initiative among six European states, led to a much more ambitious agreement to form a European Economic Community in 1957. Over the next half century, the Community grew incrementally in geographic reach, issue coverage, and depth of policy cooperation. Within the EU, economic connections have progressively deepened from the initial lowering of trade barriers through the Single Market’s convergence of regulation and finally the adoption of the euro as a common currency by 19 member states. Today, the 28-country EU incorporates the free movement of goods, services, capi- tal, and people, with associated supranational common institutions—all the hard- won results of a multigenerational push toward greater connectivity. A major feature of European integration in the past 20 years has been the process of EU accession—most notably by 10 Baltic and Central European coun- tries (the Czech Republic, Estonia, Hungary, Latvia, Lithuania, Poland, the Slovak Republic, and Slovenia in 2004, followed by Bulgaria and Romania in 2007). Until the dissolution of the Soviet Union in 1991, the 10 nations that joined the EU had in one form or another been part of the ECA region’s second major regional bloc: the Council for Mutual Economic Assistance. The perceived advantages of con- nectivity led to a looser form of economic integration and cooperation between Russia and the former Soviet republics—the Commonwealth of Independent States (CIS). In the past decade, Russia has sought to deepen the CIS into a com- mon market and economic union and pursued a process of deepening economic integration with a subset of its neighbors through the creation of a Eurasian Economic Union. However, while progress has been made, the strength of global connectivity in the CIS remains much lower than in the EU. The ECA region’s growing participation in global and regional supply chains has greatly increased the importance and variety of international economic con- nections across the region. These forces have expanded ECA countries’ regional connections more rapidly than their connections outside the region. Nevertheless, as shown in the example of trade connectivity (figure 1), many ECA countries have achieved substantial increases in global connectivity through their links to other ECA countries, such as Germany (DEU in the figure), France (FRA), or the United Kingdom (GBR), that have strong global connections. The ECA region’s persistent efforts to integrate reflect at the very least an intui- tive appreciation of the potential benefits from greater connectivity. More formally, economists have recognized the superiority of openness over autarky. In studying linkages between nations, they have focused on how knowledge transfers through international connectivity boost long-term growth, rather than one-time jumps in output due to gains from specialization (Romer 1990). Much of the knowledge gain from connectivity comes from “tacit” knowledge—the kind that comes through learning by doing and face-to-face interactions. Unlike “explicit” knowl- edge, it cannot be transferred in texts and blueprints. When looking at connectivity and knowledge transfers, analysts typically con- sider one channel at a time—such as trade, FDI, migration, telecommunications, or transport links. While many cross-country studies find, for example, that the level of trade or FDI relative to gross domestic product (GDP) is positively associ- ated with growth, they generally do not consider how many forms of connectivity work together. For example, it is hard to imagine trade taking place on the historic 4  ●   Critical Connections: Promoting Economic Growth and Resilience in Europe and Central Asia FIGURE 1  Exports of manufactured goods a. 2000 b. 2014 Source: Calculations based on data from the United Nations Conference on Trade and Development. Note: The size of each country node reflects the total volume of trade. Each node has two outgoing links, which point to the country’s two top ­ export partners. Countries in the Europe and Central Asia region are shown in shades of blue. The methodology for plotting the countries attempts to show clearly the connections between countries in the global network of countries. The largest country nodes are pulled to the outer boundaries of the figure, but the pull is counterbalanced by the number and strength of connections with partner countries. Consequently, country nodes will tend to be grouped ­ ­ together if they share common connections. Silk Road without migration and transportation networks, or the recent develop- ment of e-commerce without high-speed internet and an efficient means of trans- ferring goods from one country to another. The importance of each connectivity channel for growth is likely to be affected by the strength of other channels—particularly when technology transfers depend on both tacit and explicit knowledge. For example, FDI by higher-income ECA economies in those with lower incomes can be an impor- tant source of knowledge transfers through exposure to sophisticated produc- tion techniques and management styles that are learned “on the job.” Migrants often learn important skills working abroad, and workers and managers from the investing country typically accompany the FDI. Thus, FDI and migration can work together to accelerate technology transfers within ECA. In Moldova, for example, connections developed through migration to Northern, Western, and Southern Europe in the 1990s subsequently generated Italian investment in the garment industry as well as German investment in factories for the assembly of electronic car components. Because of these initial connections and foreign investments, Moldova is now developing a service and Critical Connections ● 5 manufacturing industry for the local market, creating its own brands, and exporting to other ECA countries. In addition to being mutually reinforcing, connectivity channels vary in depth and geographic composition. Being well connected to highly connected countries can provide benefits beyond being well connected to comparatively isolated countries. The advanced economies in Europe have provided a gateway for knowledge trans- fers from outside of ECA. Poland, for example, leveraged its growing ties to Germany to develop connections with that country’s trading partners and expand trade to broader markets within Europe and beyond. In the ECA region and other parts of the world, greater connectivity has delivered overall economic benefits for growth and development. Regional and global connectivity have been a tremen- dous “convergence machine,” raising living standards in lower-income countries to those of wealthier middle- to high-income countries (see World Bank 2012). The gains, however, are not evenly distributed or universally recognized. The 2007 global economic crisis and various commodity price shocks underscored the importance of understanding the potential risks of increased connectivity transmit- ting shocks from one country to another.1 Voters, both in Europe and elsewhere, are now questioning whether the benefits of greater connectivity are worth the costs. In addition to the United Kingdom’s 2016 vote to exit the EU (Brexit), recent elections in several European countries reflect an underlying skepticism regarding the benefits of deepening cooperation, with voters increasingly favoring parties seeking greater national autonomy instead of greater regional integration. Some analysts have attributed the lack of enthusiasm to concerns over the large migration flows and recent influx of refugees. Certainly, large sudden shifts in migrant flows, due to natu- ral disasters or wars, bring critical societal issues into play for domestic policy consid- eration. But larger questions have been raised about the downsides of regional integration and globalization in general and the role that deeper integration initia- tives have played as a driver in the rise of populism (see, for example, Rodrik 2018). Thus far, the skepticism has not led to a widespread retreat from integration among ECA countries. The institutions and policies that promote regional and global connectivity remain largely intact, with most countries continuing to benefit. However, ECA integration has been slowing, and the chal- lenges and questions call for a better understanding of ECA connectiv- While recognizing ity and its economic impacts. Analyzing the evolution of ECA’s regional the benefits of greater and global connectivity calls for paying particular attention to how connectivity, it is the various types of connections have interacted with one another important to acknowl- edge the potential and why connectivity in the region and in the larger global network downside risks through has played a key role in boosting growth and living standards. While the transmission of recognizing the benefits of greater connectivity, it is important to economic shocks. acknowledge the potential downside risks through the transmission of economic shocks as well as the choices countries face regarding which types of connections to strengthen with various partner countries. This short publication summarizes the main findings of the World Bank flagship study Critical Connections.2 Its primary purpose is to offer a deep analysis of ECA connectivity and how it has evolved over the past two decades. In a key innovation of the study, a network analysis measure of multidimensional connectivity captures the relationship between different forms of ­ connectivity and their joint impacts on 6  ●   Critical Connections: Promoting Economic Growth and Resilience in Europe and Central Asia growth and the transmission of shocks. The next step is to examine how knowl- edge flows through trade and investment channels from the ECA’s frontier firms to less technically advanced companies, improving productivity at the firm level. In the ECA region, firm connectivity does not just exert its influence through foreign investment; it also works through enhancing management practices. Ties between firms are associated with better outcomes in foreign-owned or -managed firms as well as with spillover effects that improve outcomes in locally owned and managed firms. Another complementary channel of ECA connectivity is migration. A new methodology for filling in large gaps in our knowledge of ECA migration, particu- larly regarding skills and gender, provides insights into the trends and determi- nants of migration and migration’s economic impact on the region. Facilitating the movement of people and goods across the ECA region is the focus of the next layer of connectivity: infrastructure linkages. Another key innova- tion looks at the time and cost involved in moving goods and people across the region, rather than the kilometers and density of roads and rail links. This network analysis yields a richer perspective on the ECA’s transport links. The development of supply chains has been a key organizational outcome of the depth of ECA infor- mational channels and conduits for connectivity. The development of Europe’s supply chains (“Factory Europe”), and the efficiency gains they provide, reflects the successes of narrowing policy barriers to trade, investment, migration, infor- mation and communication technology (ICT), and transport. Finally, ECA coun- tries’ policy progress in supporting international connectivity over time and relative to other countries is evaluated to guide future policy actions. Multidimensional Connectivity a Key to Europe and Central Asia’s Development and Growth International connections include trade, FDI, migration, ICT, and transport links. Most studies measure the impact of each of these channels individually. This study takes a different approach, creating an indicator that combines all channels (networks) in a functional form that recognizes their complementarity—​ the multidimensional connectivity index (represented in figure 2 as the MDC network). The measure reflects both the depth of each channel between each pair of countries (e.g., the size of bilateral trade relative to each country’s GDP) and the benefits a country may reap from being connected to another well-­ connected country (e.g., Croatia’s trade with Germany is likely to boost ­knowledge spillovers more than Croatia’s trade with Albania owing to Germany’s wider global connections in addition to its higher level of technology). Compared with traditional approaches, this method more accurately mea- sures a country’s exposure to knowledge flows via direct and indirect interna- tional connections. The analysis presented in this study emphasizes the importance of complementary and balanced connectivity across the various channels. The impact on growth of any single connectivity channel is expected to decline as additional knowledge gains from the channel diminish—unless other channels of connectivity grow as well. In other words, policies to improve trade without complementary policies to improve investment and transport will Critical Connections ● 7 FIGURE 2  Multidimensional connectivity combines many channels of connectivity Trade network FDI network N-network MDC network Note: This figure presents an indicative representation of the multidimensional connectivity (MDC) network that incorporates the relationship between all networks—trade, FDI, and other measured ­ global networks (N)—into a single collapsed network. A modified form of PageRank centrality for each country (node) is developed based on this collapsed network and used as an indicator of how overall connectivity influences growth overall and growth of the bottom 40 percent of the income distribution. FDI = foreign direct investment. have diminishing returns. Thus, promoting balanced connectivity across trade, transport, foreign investment, and other channels is likely to be more beneficial than focusing on enhancing only one channel. The various channels exhibit some degree of substitutability, but complemen- tarity dominates. In some contexts, for example, trade may substitute for FDI because firms can either export a product to a foreign market or invest in the for- eign market to produce there. However, the information flows from trade tend to complement those from FDI. Firms may discover opportunities to export to a for- eign market because of their exposure through investing there. Thus, improving connectivity in one dimension improves connectivity through other channels. In terms of per capita levels in ECA subregional multidimensional connectivity (table 1), Western Europe has the highest global ranking, followed by Northern, Central, and Southern Europe, while Russia, Turkey, and Eastern Europe are in the middle range, and the Western Balkans, Central Asia, and the South Caucasus have the lowest levels of overall connectivity. Not surprisingly, higher per capita levels of connectivity are associated with higher levels of development, 8  ●   Critical Connections: Promoting Economic Growth and Resilience in Europe and Central Asia TABLE 1  Multidimensional Connectivity Varies by ECA Subregion, with the Highest Connectivity in the Western Part of the Region and the Lowest in the Eastern Part Multidimensional ECA subregions connectivity Trade FDI Migration ICT Airline Portfolio flows High connectivity Western Europe 6 6 6 9 9 15 19 Northern Europe 12 12 17 26 21 23 22   of which Baltics 30 28 36 38 50 28 21 Southern Europe 25 24 26 21 28 23 22 Central Europe 31 27 34 36 41 46 46 Medium connectivity Russian Federation 55 53 61 28 63 64 83 Turkey 57 51 67 33 73 79 40 Eastern Europe 62 59 60 81 54 57 76 Low connectivity Western Balkans 88 75 97 45 88 86 99 Central Asia 94 99 93 101 101 103 101 South Caucasus 104 104 102 64 104 104 93 Note: The table shows global rankings, from best to worst, in combined per capita connectivity, with lower values indicating better connectivity. Subregion indicators are median values of the subregion’s countries ECA = Europe and Central Asia; FDI = foreign direct investment; ICT = information and communication technology. reflecting both the number and depth of connections a country has. Tables A.1 and A.2 show individual country rankings of multidimensional connectivity on an absolute and a per capita basis, respectively. Central Asia and the South Caucasus rank low on overall connectivity, but because they started from a low base, they also saw the greatest improvement from 2000 to 2014 (figure 3). The South Caucasus saw connectivity increase by nearly 75 percent, while Central Asia saw connectivity increase by more than 40 percent. Eastern Europe and the Western Balkans, although also starting from relatively low levels, have not seen increases as rapid, with connectivity increasing only 20 percent and 10 percent, respec- tively. The key challenge for these regions is to find ways to improve balanced connectivity, particularly through easing domestic constraints on doing business and facilitating trade, FDI, airline, and ICT connectivity. For the ECA region as a whole, while improvements in connectivity have slowed since 2008, it still has grown faster than global connectivity since 2000, reflecting the EU integration process as well as strides taken in transition economies. Using this study’s new MDC measure, we find a closer association with growth than when considering individual connectivity indices separately. In the ECA region, the depth of a country’s international connections in 2000 contributed to growth over the subsequent 16 years, after accounting for other fundamental determinants of growth typically used in cross-country studies (such as initial GDP, education, size of government, inflation, investment rate, and quality of governance). This is because a deepening of each channel tends to increase the Critical Connections ● 9 FIGURE 3  Europe and Central Asia’s connectivity has grown, but there are wide variations across subregions Growth in connectivity, percent, 2000–14 80 70 60 50 40 30 20 10 0 Central Central Eastern Northern Russian South Southern Turkey Western Western Global Asia Europe Europe Europe Federation Caucasus Europe Balkans Europe Note: Subregional and global indicators are median growth rates. growth impact of other channels. The association between MDC and growth is shown in figure 4, along with each individual component connectivity channel. The level of trade connectivity has the most significant individual impact on growth, followed by measures of connectivity through FDI, migration, and airline flights. Trade connectivity is statistically significant and associated with more rapid income growth of the bottom 40 percent of the income distribution, but the other connectivity indicators are not, perhaps because the bottom 40 per- cent benefit more directly from trade and less so from other forms of connectivity. The increase in international connectivity over the past decade has occurred in tandem with severe disruptions to the international economy, most notably the global financial crisis. Greater connectivity may have increased ECA countries’ exposure to such shocks, but it may also have increased countries’ ability to cope with them. Least vulnerable to shocks are countries with very high levels of connec- tivity and countries with very low levels of connectivity. The former can more easily find alternative export markets or sources of finance, and the latter are more insu- lated from the global economy. Countries in the middle of the range tend to be the most vulnerable to shocks for lack of easy alternatives to compensate for declines in connectivity. One example is countries highly dependent on a well-connected country—as shown in figure 5 in the case of a shock originating in Germany. Using multidimensional connectivity to better understand the transmission of shocks also indicates that ripples radiate across countries indirectly con- nected to places experiencing hard times. The impacts are not always obvious. Take, for example, the potential impact of Brexit. In this example, a 20 percent drop in all connections between the United Kingdom and the 27 remaining EU countries (EU27) is simulated. As expected, the United Kingdom suffers the greatest harm, followed by small countries of the EU27, such as Malta and Ireland, that get most of their global connectivity through the UK. However, other countries outside the EU that are not directly affected by Brexit, such as 10  ●   Critical Connections: Promoting Economic Growth and Resilience in Europe and Central Asia FIGURE 4  Connectivity’s a. Single-dimension connectivity effects on overall and bottom-40 growth effects on overall and 1.6 *** bottom-40 growth 1.4 1.2 1.0 0.8 0.6 *** *** 0.4 * 0.2 * 0 –0.2 –0.4 Trade FDI connectivity Migration ICT connectivity Portfolio flows Airline connectivity connectivity connectivity Overall growth Bottom-40 growth b. Multidimensional connectivity effects on overall and bottom-40 growth 1.6 *** 1.4 1.2 1.0 0.8 *** 0.6 0.4 0.2 0 Multidimensional connectivity Overall growth Bottom-40 growth Note: All coefficients (except those on multidimensional connectivity) are estimated with ordinary least squares regression; multidimensional connectivity is estimated using a maximum likelihood procedure. The connectivity variables, including multidimensional connectivity, are normalized using the standard normal distribution; therefore, the size of the coefficient represents the annual growth impact of a one- standard-deviation change. FDI = foreign direct investment; ICT = information and communication technology. Significance level: * = 10 percent, ** = 5 percent, *** = 1 percent. Norway, Senegal, Libya, and Fiji, are nonetheless indirectly affected through the connections of connections (table 2). Interestingly, these countries are more affected by Brexit than some EU countries, such as Austria, Estonia, Lithuania, Romania, and Bulgaria. How connectivity transmits shocks is relevant to the past decade’s shift in public sentiment away from openness and toward a more inward-looking stance. At the same time, the new mood increases the need to better understand how connectivity works to improve economic performance. The next few sections Critical Connections ● 11 FIGURE 5  A shock originating in Germany has the largest impact on countries that gain their global connectivity through Germany Shock: 10 percent fall in all types of connectivity 0 SGP URY NZL HKG CAN PAN AUS BLZ PRY TTO ARG JAM SLV MYS IRL KGZ GUY KWT ECU GTM MEX OMN BLR THA –1 ATG BRN SWZ JOR QAT CRI SAU NOR SWE BWA FIN GBR BEN GAB ALB MUS ARE LUX JPN AZE GEO MOZ BHS DOM PER ISR IDN BRA CHN ARM CYP NGA MAR CHL FRA AFG CMR KEN COL USA TGO ETH GHA ESP BEL Change in level of connectivity (percent) BFA TJK NAM DZA EST EGY DNK LBN LVA ZAF BRB PRT IND CHE –2 TUN LTU ITA PAK BHR BGR NLD BIH RUS MKD KAZ GRC –3 SVK CZE Most affected by shock HUN due to high reliance –4 MLT on connectivity HRV AUT with Germany SYR TUR MDA BGD –5 LKA POL UKR –6 0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 Initial level of connectivity: Multidimensional index TABLE 2  EU and Non-EU Countries Most Affected by Brexit 1. United Kingdom 15. Italy 2. Malta 16. Poland 3. Ireland 17. Germany 4. Cyprus 18. Latvia 5. Netherlands 19. Finland 6. Denmark 20. Hungary 7. Luxembourg 21. Czech Republic 8. Sweden 22. Senegal 9. France 23. Libya 10. Spain 24. Suriname 11. Norway 25. Slovenia 12. Greece 26. Fiji 13. Portugal 27. Iceland 14. Belgium 28. Austria Note: The table ranks countries according to the impact on the countries from Brexit, with a ranking of 1 indicating the greatest impact. EU = European Union. 12  ●   Critical Connections: Promoting Economic Growth and Resilience in Europe and Central Asia describe the ECA region’s recent experiences with the key channels of greater connectivity: knowledge transfers, foreign ownership and management, migra- tion, infrastructure, and supply chains. Europe and Central Asia Connectivity Is a Critical Source of Knowledge Transfers ECA connections with other countries through trade, investment, and production sharing are important because they increase access to technology and ideas critical to growth. Importing firms learn from exposure to more diverse and sophisticated inputs to their production, and exporters learn through opportunities to achieve economies of scale, upgrade workers’ skills, and improve products to compete in foreign markets. Local firms involved in FDI learn through technology transfers and exposure to high- skilled workers. Moreover, local firms not involved in trade or FDI also may learn through exposure to, or competition from, more internationally connected firms. All of this emphasizes the critical importance of openness to international transactions for gaining access to the knowledge essential for growth and productivity enhancements. Romania’s greater openness after it joined the EU led a Belgian logistics provider and an American software company to extend global value chains (GVCs) into the country, creating spillovers that benefited the local economy (box 1). BOX 1 Global Value Chain Spillovers in Romania H. Essers and Oracle are two examples of foreign Oracle is a major multinational company head- companies investing in Romania that illustrate quartered in the United States, specializing in benefits from foreign investments. developing and marketing database software and H. Essers is a leading European logistics firm technology, cloud-engineered systems, and with headquarters in Belgium, focusing on chemi- enterprise software products. In the mid-2000s, it cals, pharmaceuticals, health care, and high-quality opened a branch in Bucharest and began to hire products. After its integration with a Dutch com- local engineers for routine software develop- pany already doing business in Romania, the ment. In addition to short-term spillovers, Belgian firm increased its presence in Romania, Oracle’s entry has spurred a new generation of with an eye on Eastern Europe and Central Asia. entrepreneurs who got their start at the compa- Knowledge and know-how coming from traditional ny’s operations in Bucharest and went on to logistics hubs like the Netherlands and Belgium create their own businesses. One of them is subsequently improved Romania’s logistics perfor- Softelligence, a Romanian software company that mance. Logistics is the backbone of supply chains, designs tailored mobile applications for financial making production fragmentation and the smooth institutions. The low cost of entry for new entre- coordination of its stages possible. Knowledge preneurs in this industry—coupled with competi- spillovers occur through clients’ learning good tive wages, a qualified workforce, and excellent practices in quality norms, information technology, internet connectivity—has boosted this sector and cold chains. and diversified the economy. Critical Connections ●  13 Europe compares well to the global frontier in manufacturing productivity, but it lags behind the global frontier in services and in some innovation-based growth industries. Technology creation in European manufacturing is very simi- lar to that in other advanced economies, as measured by the gap in labor productivity growth between frontier firms in Europe and the Organisation for Economic Co-operation and Development (OECD) countries (figure 6).3 However, labor productivity growth is lower in European services firms than in firms at the OECD frontier. The numbers suggest that the continent could be served well by pursuing better connectivity to the global frontier firms in the services sector. Similar sectoral differences between the most advanced European and global firms can be seen in the intensity of investment in research and development (R&D). FDI inflows, FDI outflows, and imports are all important to productivity growth across the world; as such, they are conduits for the transmission of both tacit and explicit knowledge. Empirical evidence suggests that when a pair of countries are linked through FDI or trade, an increase in R&D investment in one is associated with an increase in total factor productivity (TFP) in the other. In other words, FDI and trade help international R&D spillovers materialize.4 When the focus is trade as a conduit of R&D spillovers, evidence reveals that the quality of domestic insti- tutions is an important factor that facilitates spillovers. Better business environ- ments, the quality of tertiary education systems, and stronger patent protection are associated with stronger R&D spillovers. FIGURE 6  Europe lags 3.0 behind the frontier in services 2.5 Average annual labor productivity growth, percent, 2010–13 2.0 1.5 1.0 0.5 0 Manufacturing Services European frontier OECD frontier Source: Calculations based on data from the Organisation for Economic Co-operation and Development (OECD) and Amadeus. Note: Sample is based on firms with more than 20 employees. The European frontier is among the EU15 (that is, the original core countries of the European Union). The technology gap is proxied by the differ- ence in productivity growth between frontier firms and other firms (laggards) in the same sector and year. 14  ●   Critical Connections: Promoting Economic Growth and Resilience in Europe and Central Asia Productivity increases in most firms are generated by the absorption of knowl- edge from other sources, rather than through their own investment in creating new knowledge. A firm’s potential to learn from existing knowledge can be measured by the difference between the firm’s TFP (or its TFP growth rate) and that of the most advanced firms in the sector. In Europe, advanced firms tend to be larger, have higher levels of capital relative to labor, invest more in intangibles (such as marketing practices), and have more-educated workers than other firms, although some of these differences vary by sector. The transfer of knowledge from international sources tends to follow a two- stage process (figure 7). First, the most advanced domestic firms absorb knowledge from the most advanced firms globally, often through participation in GVCs that involve production sharing through trade, investment ties, and contractual agree- ments. Second, less advanced domestic firms absorb this knowledge through their exposure to the most advanced domestic firms. By contrast, the direct technology transfer between the most advanced global firms and the less advanced domestic firms tends to be limited. Econometric evidence for the ECA region confirms that a rise in TFP growth among advanced domestic firms (defined as the top 20 percent of domestic firms by TFP in each sector) leads to a similar increase in TFP among other domestic firms, but an increase in TFP among the most advanced global firms has little direct impact on the less advanced domestic firms. This analysis sheds light on the productivity slowdown in many ECA coun- tries after the global economic crisis. Productivity growth in Central and Eastern European EU members fell by 8.2 percentage points in 2008–14, compared to 2000–07 ­ (figure 8). The crisis was transmitted through global supply chains and sharply reduced domestic firms’ engagement in these chains, which serve advanced markets in Europe and the United States. This decreased opportuni- ties for firms in Central and Eastern Europe to learn, led to a fall in their R&D spending as a share of GDP, and lowered their propensity to introduce new FIGURE 7  How technology flows from European frontier firms (global value chain lead firms) to the remaining European firms STAGE 1: National Remaining GVC Integration of frontier: firms, • Predominantly through frontier: national firms in core STAGE 2: Top-producivity exposed to domestic networks Global lead production processes Outsourcing of firms technology firms in of GVC leads noncore • Some direct access to participating from non-CEE capabilities foreign multinationals Exposure to directly in national countries learning GVCs frontier Basic capabilities Irregular engagement of FDI with nonnational frontier firms on noncore GVC activities requiring noncore activities Note: CEE = Central and Eastern Europe; FDI = foreign direct investment; GVC = global value chain. Critical Connections ● 15 FIGURE 8  Productivity growth was lower in Central and Eastern Europe during the crisis Difference between labor productivity growth in Central European EU countries and that in Eastern European EU countries, 2000–07 versus 2008–14 2 0 –2 –4 –6 –8 –10 Latvia Estonia Lithuania Slovenia Romania Czech Slovak Bulgaria Croatia Hungary Non-CEE Poland Republic Republic EU Capital deepening Labor quality Total factor productivity Labor productivity Source: Calculations based on Conference Board data. Overall macroeconomic data may reflect both sectoral changes and within-firm productiv- ity growth. Note: “Non-CEE EU” refers to the unweighted average for Austria, Belgium, Cyprus, Denmark, Finland, France, Germany, Ireland, Italy, Luxembourg, Malta, the Netherlands, Portugal, Spain, Sweden, and the United Kingdom. CEE = Central and Eastern Europe; EU = European Union. products or processes—as shown in the World Bank’s Enterprise Surveys. In this instance, exposure to international volatility was a major driver of slower growth following the crisis. However, the “cure” of reducing firms’ exposure to international volatility through restrictions on trade or FDI would be worse than the disease, because such policies would diminish firms’ opportunities to learn through participation in global supply chains and other international transac- tions. This would particularly depress growth in less advanced countries, where firms are further from the productivity frontier and thus have greater opportuni- ties to learn through connections with foreign markets. International knowledge flows and their productivity impacts take place within companies—so their internal operating characteristics are likely to be important in determining whether connectivity gains are large or small. A critical factor is management. Looking at micro data, the next section focuses on how foreign management, regardless of ownership, can influence firm outcomes. Foreign-Owned and -Managed Firms Tend to Perform Better and Contribute to Local Firms’ Productivity The share of ECA firms owned by foreigners (excluding firms owned by parent companies located in tax havens) ranges from negligible, in countries such as Belgium, Bulgaria, Hungary, Russia, Ukraine, and most Southern European 16  ●   Critical Connections: Promoting Economic Growth and Resilience in Europe and Central Asia ­ ountries, to 5 ­ c percent or more in most of Central Europe, Latvia, Lithuania, and the Western Balkans, to more than 32 percent in Ireland and the United Kingdom. More than half of foreign-owned firms in ECA also have predominantly foreign management (figure 9). Across the ECA, foreign-owned firms tend to be larger than domestic firms, although the age of foreign-owned firms is not, on average, significantly different from that of local firms (figures 10 and 11). Many ECA firms are owned by people or firms in large, rich economies, such as Germany or the United States. However, geographic proximity, common language, cultural heri- tage, trade ties, and immigration from the source country are also important determinants of foreign ownership (table 3). Firms that are foreign owned or foreign managed tend to achieve higher growth in operating revenues, employment, and average wages than other firms (figure 12). Foreign-owned firms with foreign management have 28.3 percent higher growth in operating revenue, 19.6 percent higher job growth, and 16.8 ­percent higher wage growth than local firms. Foreign affiliates with local managers also perform better than local firms, although less so than foreign firms with foreign management. However, it is unclear whether the foreign firms’ better performance reflects the impact of foreign own- ership or management or foreign companies’ tendency to invest in the most productive regions, sectors, or firms. FIGURE 9  Foreign-owned and foreign-managed firms in ECA, 2013 35 30 25 Share of all firms (percent) 20 15 10 5 0 SVK CZE ROU SVN POL HRV BGR HUN DNK LVA EST FIN SWE LTU ISL UKR RUS GRC ITA ESP PRT BIH SRB GBR IRL AUT DEU FRA NLD BEL Central Europe Northern Europe Southern Europe Western Western Europe Other Eastern Federation Russian Balkans Europe Foreign-owned firms Foreign-managed firms Foreign-owned and foreign-managed firms Source: Calculations using Orbis data. Note: Sample excludes firms with owners in tax haven countries. ECA = Europe and Central Asia. Critical Connections ● 17 FIGURE 10  Large firms are more likely to be foreign owned in ECA Share of foreign-owned firms by number of employees, 2013 20 18 18.0 17.7 17.2 16 14.8 14 12 11.7 11.2 Percent 10 8.6 8.5 8 6.7 6.9 6 4.6 3.8 4.1 3.8 4 2 2.0 1.6 1.8 2.0 2.3 0.6 0 0.0 0.1 0.2 0.3 0.0 0.1 0.1 0.5 Central Europe Northern Europe Ukraine Russian Federation Southern Europe Western Balkans Western Europe 1–9 employees 10–49 employees 50–249 employees 250+ employees Source: Calculations using Orbis data. Note: Sample excludes firms with owners in tax haven countries. ECA = Europe and Central Asia. FIGURE 11  There is no clear relationship between a firm’s age and the likelihood of its being foreign owned Share of foreign-owned firms by age of firm, 2013 8 7.0 7 6 5.9 5.6 5.2 5 4.9 Share or firms (percent) 4.2 4 3.5 3.4 3.5 3.0 3 2.9 2.8 2.6 2.7 2.1 2.1 2 1.6 1.4 1.2 1.3 1 0.7 0.1 0.1 0.0 0.1 0.0 0.1 0.1 0 Central Europe Northern Europe Ukraine Russian Federation Southern Europe Western Balkans Western Europe 1–4 years 5–9 years 10–29 years 30+ years Source: Calculations using Orbis data. Note: Data for 2013. Sample excludes firms with owners in tax haven countries. 18  ●   TABLE 3  Most Foreign Firms in ECA Are Owned by German and US Firms Most common global ultimate owner Others (from left to right, top to bottom): Denmark, Norway, Russian Federation, United Belgium, Croatia, Slovenia, Region Germany United States Kingdom Netherlands Austria France Italy Finland Sweden and Japan Central Europe Northern Europe Ukraine Russian Federation Southern Europe Western Balkans Western Europe Note: Sample excludes firms owned by tax haven countries. Each row in the table shows the five (or more, if there is a tie) most common countries of ownership, among the top ten countries of ownership, for each country or region at left. For the Russian Federation and Ukraine, the rows show the five countries with the largest ownership shares. ECA = Europe and Central Asia. Critical Connections: Promoting Economic Growth and Resilience in Europe and Central Asia Critical Connections ● 19 FIGURE 12  Foreign-owned 30 and -managed firms perform 25 better than local firms 20 Percent 15 10 5 0 Growth in operating revenues Growth in jobs Growth in average wages Locally owned, foreign managed Foreign owned, locally managed Foreign owned and managed Note: Each bar in the figure represents the difference in growth (of the type labeled) between the type of firm depicted in that bar and that of firms that are both locally managed and locally owned. underlying coefficients are statistically significant. All ­ Transfer of management practices from the source country is likely an important reason for the better performance of foreign firms. For example, US-owned firms in Europe have management practices that place more emphasis on merit in determining career success, which is associated with greater use of ICT, than do domestic firms or firms owned by other countries (Bloom et al. 2018). On average in the ECA region, but not in the most advanced ECA economies, foreign-owned firms tend to have better management practices than local firms (figure 13). The source country’s management quality is significantly related to the performance of its foreign affiliates: foreign affiliates from countries with better management practices perform better than other foreign affiliates, even after differences ­ between the source countries’ income levels, financial development, population, and stock of immigrants are taken into account. Local firms without foreign ownership or management can also benefit from the presence of foreign-owned firms (figure 14). Local firms can learn from observing management practices and technology in foreign affiliates, or through hiring workers trained in foreign affiliates. However, evidence of such effects across ­ industries is mixed. Better-performing foreign affiliates also may affect local firms through competition—by forcing them either to improve or to exit the market. Local firms tend to achieve significantly higher growth in operating revenues and wages in regions or sectors in ECA countries with higher shares of foreign firms than in sectors in which foreign affiliates are less prevalent. For small and young firms, there is no statistically significant relationship between the share of foreign firms in a sector and local firms’ employment growth. A possible interpretation of this result is that the presence of foreign firms forces some successful small and young local firms to become more efficient by increasing capital relative to labor, slowing job creation. In addition, ­ other firms that cannot compete shed labor (to more efficient firms) or close. Again, these relationships may in part reflect foreign owners’ decisions to invest in better-­performing sectors or regions. 20  ●   Critical Connections: Promoting Economic Growth and Resilience in Europe and Central Asia FIGURE 13  Foreign affiliates 6 tend to have better management practices IRL FIN SWE than local firms DEU BEL DNK ITA LTU PRT GBR HUN AUT AMI of foreign-owned firms GRC POL FRA NLD CZE ESP HRV SVK 5 BGR RUS SVN ROU BIH 4 3 3 4 5 6 AMI of locally owned firms Western Europe Western Balkans Southern Europe Russian Federation Central Europe Other Eastern Europe Northern Europe 45-degree line Source: Calculations using data from the World Economic Forum (WEF) Global Competitiveness ­ Survey and Orbis. Note: A country’s Average Management Index (AMI) is based on the WEF Global Competitiveness survey that measures the quality of national business schools and the reliance of professional ­ ­ management. This index is also highly correlated with the World Management Survey, which is more comprehensive, but not as widely available. Because of competition from better-managed companies, larger and older firms are more likely to upgrade and adjust compared with younger and smaller firms. Increases in the quality of management in foreign affiliates are associated with faster growth in operating revenues, wages, and employment in local firms more than four years old and having more than 50 employees, but lower growth in these performance indicators in younger and smaller firms. Foreign ownership of firms tends to reduce the level of employment volatility in a country’s domestic economy. Interestingly, once a number of variables that influ- ence firm performance are controlled for, the performance of an average foreign firm in the ECA region is not statistically correlated with local economic growth. Foreign firms are less responsive to local economic conditions than local firms, regardless of whether foreign and local economies have the same business cycle. This could reflect better access to finance during an economic upswing in the parent country or a search for opportunities abroad when the parent ­ company’s profits at home are limited relative to the destination country. Regardless of the economic conditions in the parent country, foreign companies’ employment decisions seem to be less procyclical with respect to the domestic economy than domestic compa- nies: the former tend to create fewer jobs when the local (host) economy expands (figure 15). Likewise, this also means that foreign companies tend to destroy fewer jobs than domestic firms when the local economy experiences a recession, possibly reflecting access to external factors that allow foreign companies to buffer the Critical Connections ● 21 FIGURE 14  The positive spillovers of well-managed foreign firms seem weaker for small and young firms a. Regional spillover effects 60 2 Changes due to a one-unit increase in the Average Management Index All-firm impact (right axis) score of foreign firms in the region (percentage points) 1.5 40 1 20 0.5 0 0 –0.5 –20 –1 –40 –1.5 –60 –2 Mature and Small and Mature and Small and Mature and Small and larger firms young firms larger firms young firms larger firms young firms Change in operating revenue Change in number of employees Change in average wage b. Sectoral spillover effects 20 1.5 Changes due to a one-unit increase in the Average Management Index All-firm impact (right axis) 15 1.0 score of foreign firms in the sector (percentage points) 10 0.5 5 0 0.0 –5 -0.5 –10 –15 -1.0 –20 -1.5 –25 -2.0 –30 –35 -2.5 Mature and Small and Mature and Small and Mature and Small and larger firms young firms larger firms young firms larger firms young firms Change in operating revenue Change in number of employees Change in average wage Note: Small firms are those with 49 employees or fewer; young firms are those four years old or younger. Each bar represents the effect of increasing by one point the Average Management Index scores of foreign firms. The bars in each panel show the baseline effect (mature and larger firms), the baseline effect plus the interaction term associated with size, and the baseline effect plus the interaction terms associated with size and age simultaneously. 22  ●   Critical Connections: Promoting Economic Growth and Resilience in Europe and Central Asia FIGURE 15  Foreign firms’ 10 employment decisions are less procyclical than those of 8 Foreign firms’ employment growth relative to local firms' GDP growth and difference between foreign-firm and their domestic peers 6 Local GDP growth local-firm employment growth (percent) 4 2 0 –2 –4 –6 –8 –10 t1 t4 t7 t10 t13 t16 t19 t22 t25 t28 t31 t34 t37 t40 t43 Time period impact of the decline in economic activity. In other words, while foreign firms seem to contribute less to job creation than their local counterparts when the local econ- omy is growing, they seem to bring more stability to the labor market during a downturn in economic activity because they lay off workers to a lesser extent than local companies do. It is just a short step from foreign owners and managers to the broader topic of migration, a hot-button issue in recent years. The next section focuses on how the ECA region’s increasing migration facilitates trade, knowledge transfers, and other benefits associated with greater connectivity. Economic Migration Has Been Beneficial to Europe and Central Asia In general, openness to migration, including that by foreign managers, helps many countries gain the skills, technology, and resources required to improve efficiency and compete in an increasingly complex, globalized world. In destina- tion countries, workers who are close substitutes for immigrants (e.g., they have similar skills) may lose as a result of lower wages or diminishing job opportuni- ties. At the same time, workers with skills complementary to those of immigrants may benefit. While the net economic effect for the country overall is positive, income distribution impacts may be positive or negative depending on the skill mix of the native and immigrant populations. Outside of economic consider- ations, large sudden shifts in migrant flows as a result of natural disasters or wars, such as the recent Syrian refugee crisis, bring humanitarian and local social impacts into play for the host country. These are critical societal issues for domes- tic policy consideration but are outside the purview of this analysis. Critical Connections ● 23 Both emigration and immigration rates in many ECA countries are higher than the global average (map 1), mostly driven by the removal of barriers to mobility within the EU and large migration flows following the opening up of Eastern-bloc countries. High levels of ECA migration have encouraged greater cross-border investment and trade (for example, by helping firms learn about foreign markets) and have facilitated the sharing of technology and knowledge between countries (for example, through schooling and language skills attained abroad). A large diaspora can generate substantial economic benefits for many origin countries in the ECA region. Remittances are an important source of income, have a positive impact on long-term economic growth and poverty reduction, and can improve access to capital markets. Diasporas are also a significant source of invest- ment, export demand, and knowledge transfers for ECA economies, particularly given the disproportionately high flows of skilled emigrants from regional coun- tries (see box 2). Finally, the increasing share of migrants going to the United States and Northern, Western, and Southern Europe may have contributed to improving institutions in ECA transition economies by increasing their populations’ exposure to the norms of competitive democratic countries. What determines migration connectivity? To answer this question, this report develops a global bilateral migration matrix showing the number of migrants between all country pairs. Constructing the matrix requires estimating migration flows for the many countries missing such data and then estimating the global relationship between population flows and various drivers of migration (figure 16). Most migrants move to countries with similar or higher levels of per capita income. Migration tends to increase as the distance between countries decreases. A large share of low-skilled migrants move to neighboring countries, but high-skilled migrants are more likely to move to nonneighboring countries, reflecting the tendency for high-skilled emigrants from developing countries to move to high- income OECD countries (especially English-speaking countries). Sharing a similar language also has a positive effect on migration flows, particularly for skilled migrants, whose jobs often require strong language skills. Finally, the existence of a diaspora tends to increase the flow of migrants (particularly for unskilled workers) by reducing the costs of information, financing movements, and perhaps reducing the risks involved in migration. Some characteristics of ECA migration differ from these global patterns. Migrants from ECA countries (other than the advanced European economies), regardless of education level or gender, tend to move to other countries within the region. By contrast, migrants in other regions are no more likely to move within the region, once distance between countries and common borders are taken into account. High-skilled (but not low-skilled) migrants from former Soviet Union countries tend to go to other former Soviet Union countries, where they find similar institutions (a legacy of the Soviet Union) and close economic inte- gration. Differences in the size of the working-age population (those 25–65 years of age) are also important determinants of migration flows. Countries with larger and younger working-age populations tend to have larger emigration to coun- tries with smaller and older working-age populations. Because of aging popula- tions in Central and Eastern European countries, a smaller working-age population emigrates from those countries. Central Asia, however, remains a 24  ●   Critical Connections: Promoting Economic Growth and Resilience in Europe and Central Asia MAP 1  Emigration and immigration shares have seen the highest increase in Europe and Central Asia a. Emigration shares, 2010 IBRD 43819 | JULY 2018 Emigrants as a percentage of the population, 2010 20–100 10–20 5–10 1–5 0–1 No data b. Immigration shares, 2010 IBRD 43820 | JULY 2018 Immigrants as a percentage of the population, 2010 20–100 10–20 5–10 1–5 0–1 No data Source: World Bank 2018. Note: Reported data for Central Asia for 2010 are particularly spotty; therefore this figure relies heavily on an estimation methodology developed by the World Bank staff. Critical Connections ● 25 BOX 2 Marius Stefan of Autonom Romania: Knowledge transfers through travel and studies abroad Marius Stefan, a Romanian national, graduated with than three million Romanians emigrated—most an MBA from the University of Maryland in 2004.a moving from the economically depressed region His proud parents flew in from Romania for the cer- around Piatra Neamt to Spain and Italy. Many of the emony, and Marius decided to show them around migrants returned to Romania frequently, and they the US East Coast on a short road trip. Marius did it were glad to have the opportunity to rent a car. the way he learned from friends: he rented a car. When Marius realized the potential, he decided to His father was amazed that a private i­ndividual— act on a lesson from business school: after a pilot from a foreign country, no less—could rent a car project is successful, focus all your efforts on scaling so easily. Such businesses were unknown in up as rapidly as possible. Romania. Marius had a good understanding of the Marius’s brother had been educated in France car rental market from one of the case studies from and had been working in Paris as an international his MBA program, and he explained the business consultant. After a year of successful business devel- model to his father. opment, Marius added his brother’s experience and After seeing how it was done and listening to knowledge to the business, and they have been his son, Marius’s father made an unexpected pro- working together as the enterprise has expanded posal: that they would open a car rental business in beyond anyone’s dreams. Romania. Marius’s family was from Piatra Neamt, a Autonom now operates in 46 locations, employs small town in one of the poorest regions in north- 300 workers, and offers more than 5,000 cars for rent. east Romania. It had almost no tourism and little The company is going through a stage of accelerated economic activity at the time. While starting a busi- growth. It is developing a division for long-term rent- ness was exciting, Marius thought there was no als (operating leases) both organically and through potential and quickly forgot the conversation. acquisitions. It recently acquired the operational leas- Marius returned to Romania and worked in sev- ing division of Banca Transilvania, the largest bank in eral small businesses in Bucharest, the country’s capi- Romania, and the plan is to double the company’s tal. But his father kept reminding him about his turnover and assets in 2018. While developing the proposal to start a car rental business in his home business across Romania, the company started to town. Marius finally relented and agreed to ship three expand abroad. It started Autonom Hungary three vehicles to Piatra Neamt to gauge the potential for years ago and Autonom Serbia in May 2018—in car rentals. To his surprise, the rent-a-car business effect, transitioning from a national champion to a became an overnight success. Within a few weeks, regional player. employees at the business were calling Marius in None of this would have happened had the Bucharest, asking why the company had just three Stefan brothers not migrated abroad. An idea cars. There were always more people wanting to rent, sparked by a routine car rental in another country, but too few cars to satisfy the demand. combined with the knowledge gained by studying Marius traveled from Bucharest to Piatra Neamt abroad, helped Marius and his brother build a suc- to investigate and found a simple explanation. cessful company that testifies to the power of con- When the country joined the European Union, more nectivity and its knowledge transfers. a. This box was based on discussions with Marius Stefan in 2018. The author thanks Mr. Stefan for being so generous with his time. 26  ●   Critical Connections: Promoting Economic Growth and Resilience in Europe and Central Asia FIGURE 16  ECA migration 0.8 is driven by geography, language, historical ties, and 0.6 past migration 0.4 0.2 0 –0.2 –0.4 –0.6 –0.8 –1.0 Diaspora Distance Colony Contiguity Language similarity Male unskilled Male skilled Female unskilled Female skilled Note: Calculations are for 2010. “Skilled” migrants are those completing tertiary education. The size of the bars in the figure represents the coefficient of a regression equation and the percentage point impact on migration from a percentage point change in the migrant and home/host country attributes. Regressions also include regional dummies, such as migration within Europe and Central Asia (ECA) subregions and within the rest of the world. subregion with a relatively young working-age population and has relatively large emigration. In addition, while women make up slightly less than half of global emigrants, they are the majority of ECA emigrants, possibly because of their higher skill levels relative to the global pool of emigrants. Migration and the other connectivity components discussed in previous sec- tions require highways, railroads, air links, and cargo transport to reach their full potential. The next section examines the role infrastructure plays in enhancing the ECA region’s connectivity. While all forms of transportation are important in this regard, this research project focuses on the differences in cost and time needed to move goods and people among various countries. Strong Infrastructure Transport Links Provide Important Support for Connectivity Transport infrastructure forms the bedrock for international (and domestic) connectivity. This is obvious in the cases of trade in goods and migration, but ser- vices trade and investment flows also require supportive logistics, travel-related ser- vices, and infrastructure and may be linked to goods trade. Measuring the extent to which a country’s transport infrastructure facilitates connectivity through the chan- nels mentioned is complicated. Many traditional infrastructure indicators—such as kilometers of roads and rail, their condition, or the density of connections—provide only limited information on the economic value of transport connections. To get an alternative view, this study looks at new data on transport services and network analysis tools to measure the economic value of ECA countries’ connections through roads, railroads, and to a lesser extent maritime transport. Critical Connections ● 27 The economic value of connections is measured here by the cost of trans- port, the time required, and the importance of the destination country in the overall transport network. Time is a particularly important consideration when the type of product (for example, perishable goods or parts and components traded within supply chains) or the nature of the passenger trip (for example, urgent business) requires rapid and reliable transport. Information on these dimensions is therefore essential for evaluating infrastructure investments’ real impact on growth and welfare. With some exceptions, domestic travel in the most advanced economies tends to take less time and be more affordable for residents (given the high incomes in these countries) than travel in other countries (figure 17). By contrast, the time required to deliver a container from one main city to another within a country dif- fers little across ECA subregions, except in the case of Russia, for which distance between major cities plays a large role. However, time required for cargo ship- ments between countries is quite high in a few countries that serve as gateways to their neighbors (e.g., Belgium, Luxembourg, and the Netherlands), likely owing to the high level of congestion in their highway systems. The cost and time required for passengers to connect with neighboring coun- tries varies greatly within ECA, from the highest levels in Central Asia to the lowest in the Western Balkans. Advanced Europe’s high-level road and rail infra- structure also delivers fast travel times at relatively low cost for residents, given their high incomes. Transport takes up a larger share of income in poorer ECA FIGURE 17  Transport connectivity (cost and time) between and within ECA countries Regional averages Passengers a. Within-country cost and travel time b. Between-neighbor-country cost and travel time South Caucasus Western Balkans Western Balkans Western Balkans Western Balkans Advanced Europe Central Europe Central Europe Central Asia South Caucasus Eastern Europe Advanced Europe Eastern Europe Central Europe South Caucasus Eastern Europe Central Europe Eastern Europe Advanced Europe Turkey Advanced Europe Turkey Turkey South Caucasus Russian Turkey Central Asia Central Asia Federation Russian Russian Russian Federation Federation Federation Central Asia 0 10 20 30 40 50 0 5 10 15 20 25 0 50 100 150 0 10 20 30 40 50 Average passenger Average travel Average passenger Average travel cost (euros) time (hours) cost (euros) time (hours) continued 28  ●   Critical Connections: Promoting Economic Growth and Resilience in Europe and Central Asia FIGURE 17  continued Freight c. Within-country cost and delivery time d. Between-neighbor-country cost and delivery time Western Balkans Advanced Europe Western Balkans Central Europe South Caucasus Central Asia Central Europe Eastern Europe Central Europe Central Europe Eastern Europe Western Balkans Eastern Europe Eastern Europe South Caucasus Advanced Europe Advanced Europe South Caucasus Advanced Europe South Caucasus Turkey Turkey Central Asia Turkey Central Asia Western Balkans Turkey Central Asia Russian Russian Russian Russian Federation Federation Federation Federation 0 500 1,000 1,500 2,000 0 0.5 1.0 1.5 2.0 0 1,000 2,000 3,000 0 0.5 1.0 1.5 2.0 2.5 Average container Average container Average container Average container sending cost (euros) delivery time (days) sending cost (euros) delivery time (days) Note: Within-country transport connectivity as measured here is multimodal, averaging across road, rail, and bus modes the price that must be paid to travel to a representative main city in the country. Transport connectivity between neighboring countries as measured here is multimodal between a country’s capital city and main cities of neighboring countries. Only countries with complete data for time and cost for all modes (road, bus, and rail) are included. Within-country freight transport connectivity for a given city is measured as the average price to send a container from that city to the other main cities within a country. Between-neighbor-country freight transport connectivity is measured as the average price to send a container from a country’s capital city to the main cities of neighboring countries. “Advanced Europe” includes countries in Western, Southern, and Northern Europe that signed the Maastricht Treaty or joined the European Union before 1995. ECA = Europe and Central Asia. countries, with most Central European and Baltic countries in the middle of the pack. The average cost and time required to ship a container from a country’s capital city to the main city of neighboring countries varies little among most subregions, except for higher costs and time in Central Asia, Russia, and Turkey. Central Asian countries have much higher travel costs for both road and container transport and much longer travel times than other ECA regions. These countries might get help in integrating and improving their connectivity through recent or expected infrastructure projects gathered under the Belt and Road Initiative, a long-term project designed to reduce the cost of transport from China to Europe. Along with Portugal and Spain, the island countries Cyprus, Ireland, and Malta are also among the countries with the highest costs and longest time to reach the rest of the ECA network. South Caucasus per- forms better in terms of costs compared to time, whereas in Western Europe the opposite is the case. Central and Eastern European countries have rela- tively cheaper and faster connections to the rest of the network. The similarity in these rankings largely reflects road transport costs, which are determined in part by infrastructure quality and its impact on average speeds. The cost of Critical Connections ●  29 moving containers is another matter. Unlike passenger road costs, this cost reflects other parameters, such as logistics costs, the presence of rent seekers, and the degree of competitiveness among service providers. As a result, coun- tries’ cost versus time performance is more diverse when looking at the cost of moving containers rather than people. Countries like Armenia, Greece, Kosovo, the former Yugoslav Republic of Macedonia, and Turkey have relatively better connectivity in terms of container costs than they do in terms of time. Montenegro, Norway, Slovenia, and others have relatively better time connec- tivity than cost connectivity. Understanding country specificities requires a deeper look into institutional factors, the quality of logistics, and the competi- tiveness of the transport sector. In addition to cost and time, determining the economic value of connections requires considering destination countries’ place in the network. The availability of close connections varies greatly within the ECA region: the total number of a country’s neighbors (or neighbors of neighbors) ranges from 2 to 22, and the level of aggregate GDP in the countries adjacent to a country varies significantly from country to country. Thus, some connections have a higher value than others in terms of access to a larger market. So in planning transport investments, a country should consider whether a particular investment improves access to a relatively isolated country or to a country with connections to a wider network. Transportation infrastructure channels the movement of goods and people along major cross-country networks and, within networks, corridors. The com- prehensive nature of the economic benefits that accrue to countries from being on a particular corridor, or at a specific crossroads of a network, remains an open question. A key question is whether roads or rail that pass though coun- tries provide economic benefits if ancillary businesses associated with the cor- ridor fail to materialize. If a country’s economic and business environment is sufficiently attractive for investment, however, transit flows may increase export and import opportunities for firms along these routes (or corridors), develop new sectors such as logistics services, and generate nonmaterial benefits (flows of ideas and knowledge) to boost productivity. Firms located in transport hub countries may benefit from lower production costs and an improved ability to deliver on time. Higher transport network connectivity might be desirable for increasing a country’s participation in regional and GVCs, attracting FDI, or increasing its participation in development corridors. It is important to identify the most critical countries in transport networks. Doing so reveals which countries have more control over transportation net- works’ operability and what shocks in these countries imply for other con- nected countries. Measuring critical transport networks can help countries target investments to reduce their vulnerability to specific country shocks that might impede access to markets or other areas of the network. More generally, critical countries in the transport network are those where disruption would have a major impact on subnetworks or countries that can be de facto discon- nected. For container costs, Russia is the most critical country in the network in Eurasia (figure 18). A country’s cost-driven criticality reflects the increase in costs that a shipper (in the case of containers) would incur if it had to avoid that 30  ●   Critical Connections: Promoting Economic Growth and Resilience in Europe and Central Asia FIGURE 18  Cost-driven CYP criticality in container network for Europe and ALB KSV Central Asia MNE MKD BIH GRC SRB HRV BGR TUR MLT ROM SVN HUN MDA ITA AUT PRT ARM SVK ESP GEO UKR CHE CZE FRA POL LUX GBR AZE DEU BLR BEL IRL RUS LTU LVA NLD EST DNK NORFIN KAZ SWE KGZ ISL TJK UZB TKM Note: The results shown in the figure capture only the Europe and Central Asia (ECA) transport net- work and do not include connectivity to countries outside ECA. Consequently, large ports (e.g., Rotterdam) that are connected to the United States or China will not appear as critical, although they are in the larger global context. Circle size indicates level of criticality (larger diameter = greater criti- cality). For illustrative purposes, the circles representing the top five countries in criticality are colored in green. Lines between nodes indicate the presence in the physical network of an optimal corridor connecting countries. Locations of ­ circles and countries are not linked to geography in any way. Results for time-driven criticality are not presented, as the results are very similar to those presented. country in shipping cargo. Germany, Ukraine, Hungary, and Poland are among the five most network-critical countries. As expected, islands or isolated coun- tries have a very low criticality. While France is not a top-five country in terms of network criticality, disruptions in the French transport network would affect connectivity to the rest of the ECA region for Spain, Portugal, the United Kingdom, and Ireland: Portugal’s connection to the European network is con- tingent upon Spain’s, and so forth. Different goals for transport networks may imply quite different investment priorities: • Countries may choose to strengthen partnerships—for example, to reach large markets, participate in supply chains, or connect to countries with high levels of technology (so the potential for learning is greater). As revealed by the cost and time of freight transport, ECA countries can be grouped into three categories in terms of partnerships: (a) the Western Balkans and Central Europe incur lower costs to reach the largest economies of Europe; (b) countries in Central Critical Connections ● 31 Asia and South Caucasus together with Russia incur lower costs to reach coun- tries with similar technology levels; and (c) countries in Advanced and Eastern Europe, as well as Turkey, incur lower costs to reach either the largest ECA economies or countries with more sophisticated technology. • Another possible goal among countries is to maximize the size of the market made accessible by their transport systems. Advanced Europe captures the largest amount of GDP per unit cost of transport (a container, private car, or railway ticket) among ECA subregions, followed by Eastern and Central Europe (85–90 percent of Advanced Europe’s market potential), South Caucasus and Turkey (50 percent), and Central Asia (40 percent). While the size of investments and the quality of services are important, many countries’ ability to increase their market connectivity by improving transport is limited by long distances from markets and difficult terrain. • Countries also may choose their investments to maximize their integration within the ECA transport network. Some connections contribute more to a country’s overall connectivity with the region than others. The Czech Republic, the Slovak Republic, and Austria are the three most integrated countries in the ECA network, while Central Asia remains poorly integrated. Factors over which a country has no control, such as the number of neighboring countries it has, are also key elements in determining its degree of integration, but its transportation network is more important. More cooperation among countries, especially along corridors, could increase the global benefits of transport investments. When it comes to the Belt and Road Initiative, for example, the benefit of investments for the network as a whole varies by individual segment. Reducing the cost of shipping a container in the Kazakhstan- China segment would have the largest impact on Kazakhstan’s ability to reach foreign markets, but Russia and Germany also would benefit significantly. Improving the Belarus-Russia segment would mostly benefit Belarus. A cost reduc- tion on the Poland-Belarus segment would have the smallest impact on the network as a whole. Although it would provide broad support for all aspects of ECA connectivity, bet- ter infrastructure has particular relevance for cross-border supply chains. Today’s busi- nesses, rather than being concentrated in a single country, find their production of goods is now divided among plants in different countries, with each assigned pro- duction of particular components or the assembly of components from other plants. The final consumer product may thus reflect inputs from a number of countries. The Growth of Supply Chains Reflects Greater Connectivity and Has Facilitated Increased International Knowledge Flows Regional supply chains are deepening around the world and are focused in three clusters—Asia, Europe, and North America. Europe’s supply chain is largely focused on Germany, particularly in motor vehicles, retail, and machinery and equipment. Despite enormous reductions in the prices of transport and communication, 32  ●   Critical Connections: Promoting Economic Growth and Resilience in Europe and Central Asia geographic proximity continues to be important in coordinating production through- out supply chains. In fact, the importance of regional supply chains increased some- what faster than that of global supply chains from 1995 to 2011. Proximity remains important for several reasons. Suppliers often need com- plex information (tacit knowledge) that cannot easily be codified in blueprints or instruction manuals (explicit knowledge), requiring frequent, face-to-face communications with lead firms that are difficult and costly to achieve when plants are separated by thousands of kilometers. Proximity is associated with similarities in culture and l ­anguage as well as migration networks—both of which facilitate these detailed information transfers. On-time delivery and reli- able quality are critical in supply chains, where the lack of an intermediate input can slow production all along the chain. Therefore, lead firms may place greater emphasis on allocating production to close-by firms they know, rather than seeking cheaper locations at greater d ­ istance. Locating plants in proximity to one another can help improve the allocation of workers and machines across firms, facilitate the transfer of knowledge across plants, and enable more effi- cient use of infrastructure and other public services. Finally, regional integra- tion agreements such as the EU encourage supply chain production through regional partners by lowering costs at the border, establishing similar legal and regulatory frameworks, and providing confidence in the stability of integration frameworks over time. The growth of supply chains has generated substantial benefits. Participation in supply chains can expand the range of goods produced in developing countries. For example, a poor country that finds it difficult to compete with more sophisti- cated firms in the production of complex electronic products may be able to exploit its advantage of low-cost labor to assemble such products from compo- nents produced elsewhere. The transfer of knowledge is heightened in supply chains, which often involve intensive contacts with more sophisticated firms through trade, investment, and the movement of technicians and managers. Exposure to such knowledge can improve productivity. The growth of supply chains can also increase productivity through more intense competition, greater specialization (which can improve worker performance through learning by doing), and access to increased diversity of inputs. In OECD countries, growth in a coun- associated with growth in its real labor produc- try’s participation in supply chains is ­ tivity (figure 19), although this association may partly reflect the fact that more productive countries are more likely to participate in supply chains. Increased participation in European (and Asian) supply chains has been associated with rising revenues from exports of both goods and services, even after subtracting the value of associated imports. To be more precise, partici- pation in supply chains boosts the gross exports recorded in the balance of payments statistics. However, a substantial share of these export revenues must be devoted to paying for imported inputs used in the production of the goods, so the funds channeled to domestic profits and wages (i.e., domestic value added) may be a small fraction of gross export revenues. ECA countries that have increased their participation in supply chains tend to achieve a more rapid increase than other countries in the domestic value added generated by exports. From 2000 to 2011, for example, Turkey and Poland experienced Critical Connections ● 33 FIGURE 19  Participation in 0.10 global value chains is correlated with higher labor productivity Growth in labor productivity (percent) 0.05 0 –0.05 –0.10 –0.20 –0.10 0 0.10 Growth in GVC participation (percent) Source: World Bank labor productivity data and country global value chain participation index for member countries of the Organisation for Economic Co-operation and Development over the period 2009–11. Note: Each dot in the figure represents one country for one year. GVC = global value chain. some of the largest percentage increases in the share of exports through sup- ply chains among transition countries (figure 20, panel a) as well as the highest growth rate of exports of value added (figure 20, panel b). By contrast, Slovenia, Russia, and Hungary saw decreases in the share of exports through supply chains over the same period and only modest growth of value-added exports. Greater participation in supply chains also tends to increase a country’s depen- dence on other countries, potentially raising economic volatility in some segments of the country’s economy. For example, a natural disaster in Indonesia that inter- rupts the production of an intermediate good may idle workers in the Czech Republic and reduce the profits of German retailers. Finding the central sectors and the major cross-border links is important to understanding how positive or adverse shocks spread through production net- works in the ECA region. A country or a sector that is central might be able to spread ideas to the rest of the network, but it might also more frequently receive shocks from the rest of the network. The ECA production network is organized around several clusters that include sectors (for example, retail trade and motor vehicles) from different parts of the region. Having sectors from different regions in the same produc- tion cluster illustrates the interdependence of country-sectors across most ECA countries through input-output linkages. By evaluating which countries and sec- tor clusters are most critical for production in ECA, it becomes clear that motor vehicles in Germany are the most central sector in the ECA production network. This sector largely relies on wide-reaching regional value chains to organize its production. The retail sectors in Italy, France, Germany, and Russia are all 34  ●   Critical Connections: Promoting Economic Growth and Resilience in Europe and Central Asia FIGURE 20  Among the a. Production fragmentation, 2000–11 transition countries, greater production fragmentation is HUN associated with a more rapid RUS increase in the flows of value SVN added in exports ROU SVK DEU BGR POL CZE TUR –20 –10 0 10 20 30 Change in the ratio of exports to value added (percentage points) b. Exports of value added, 2000–11 SVN HUN DEU RUS CZE SVK BGR ROU POL TUR 0 5 10 15 20 Average growth per year (percent) Source: Calculations based on Organisation for Economic Co-Operation and Development Trade in Value Added database. important as well, but Germany’s machinery and equipment sector is among the most critical value chain sectors. Outside of the mature EU countries, manufac- turing clusters in Poland, Russia, and Turkey play a secondary role. By far, France, Germany, and Italy are the most important centers for the ECA trade production network, followed by Russia and Turkey. The least central countries are the Baltic countries, the Eastern European countries, and Portugal. To this point, Critical Connections has focused on important aspects of con- nectivity in the ECA region. The phenomenon is multidimensional, with its vari- ous components working together to increase productivity. The gains, not shared equally among or within countries, are spurred by foreign ownership and management, migration, vital infrastructure, and supply chains. Critical Connections ●  35 Connectivity, however, carries risks and ignites changes in national economies. The remaining task for the study centers on ECA policies: first, the ECA region’s recent record on promoting connectivity, and second, the challenges and opportunities that remain. European and Central Asian Countries Have Moved toward More Open Policies While most economic policies can affect international connectivity in some way, this study’s evaluation of ECA policy progress focuses on a set of areas that have important implications for openness to international connections. These include policies governing import tariffs, preferential trade agreements (PTAs), inward FDI, bilateral investment treaties (BITs) that protect investors from expropriation and adverse changes in investment policies, product market regulations, sectoral domestic regulations, visa regimes, and integration of migrants. ECA countries have made important progress on policies to boost international connectivity. In part, this has reflected individual countries’ efforts to open them- selves up to the global economy, particularly following the collapse of the Soviet Union. However, regional agreements have also played a critical role, including the increasing sectoral coverage and depth of agreements within the EU, the expan- sion of the EU membership, and the formation of the Eurasian Economic Union, composed of Armenia, Belarus, Kazakhstan, and Russia. The EU also has entered into more than 50 PTAs with countries in and around Europe. Over time, these agreements have shifted their primary focus away from reducing trade barriers, expanding to include liberalizing trade in services, pub- lic procurement markets, and cross-border investment. In addition, they have added provisions governing how the agreements are implemented through national regulatory regimes (figure 21). This broader agenda requires more com- plex decisions than negotiations over tariff levels. For example, regulations designed to achieve such objectives as safeguarding the health and safety of consumers (which are consistent with a trade agreement) have a greater scope than regulatory measures that serve only to protect domestic producers (which are inconsistent with the market integration goals of a trade agreement). Studies have found that participation in deeper and more comprehensive trade agree- ments is related to a country’s ability to attract FDI. Each provision added to an agreement between a pair of countries (particularly in the areas of competition policies, investment, movement of capital, and intellectual property rights) is associated with an average 3 percent increase in FDI flows between the agree- ment partners. ECA countries have, on average, made considerable progress in reducing barriers to trade and investment: • Tariff levels applying to countries outside of preferential agreements have fallen steadily in ECA countries—from an average of 7 percent in 1988–96 to 5 percent in 2006–15. ECA’s average tariff in the latter period was lower than that in all other regions except North America (Canada and the United States). 36  ●   Critical Connections: Promoting Economic Growth and Resilience in Europe and Central Asia FIGURE 21  ECA ranks 50 among the top regions in regard to the number of trade agreements and 40 investment treaties 30 20 10 0 Total number of Total number of Percentage of bilateral investment treaties preferential trade agreements most-favored-nation tariffs (country average) (country average) (country average) East Asia and Pacific Latin America and the Caribbean North America Sub-Saharan Africa Europe and Central Asia Middle East and North Africa South Asia Note: ECA = Europe and Central Asia. Although tariff rates were reduced in all ECA subregions over this period, they were particularly low in high-income countries. • ECA countries are among the least restrictive globally toward inward FDI, according to the OECD’s FDI Regulatory Restrictiveness Index (although data are not available for the Western Balkans and the South Caucasus subregions). EU countries are the least restrictive, but all ECA subregions achieved some reduction in restrictions from 1997 to 2015. • ECA’s average score on an index of the intensity of use of PTAs dwarfs those of all other regions, driven by regional integration among EU member states and—to a lesser extent—an expansion of PTAs by Turkey and Russia. • The average ECA country has entered into more BITs—almost 50 from 2000 to 2016—than the average country in other global regions. Almost 60 percent of all BITs signed by the average ECA country involve a partner in the ECA region. ECA countries have also made progress in reducing product market restrictions that hamper international connectivity, but the advances have slowed in recent years. A measurement of product market restrictions that includes barriers to entrepreneurship, trade, and investment, as well as the impact of the scope and nature of state control of the economy, shows that the restrictiveness of the aver- age European country has declined since 1998—but progress has stalled since 2008. Similarly, an index of the degree of policy restrictiveness in the ECA region implied by domestic regulatory regimes for energy, transport, and communica- tions has improved since the mid-1980s (though the country coverage of this index is limited)—but progress has been negligible over the past 10 years. According to both indicators, the ECA region has more restrictive policies, on average, than Canada and the United States. Critical Connections ● 37 The transition countries (mostly not covered in the two regulation indicators discussed in the previous paragraph) have made significant progress in improving market-friendly regulations, as measured by the Transition Indicator Database developed and managed by the European Bank for Reconstruction and Development. The market openness indicators in formerly centrally planned econ- omies in the Baltic countries and Central Europe improved markedly from 1989 to 2000, although the pace of reform slowed in the subsequent decade. Other sub- regions also made significant progress, but scores on these indicators of open markets vary considerably. For example, the gap between scores for the average country in Central Asia and the OECD benchmark is four times larger than the gap between the average country in the Baltics and Central Europe and the OECD standard. Barriers to cross-border movement of people in high-income ECA countries—as measured by visa requirements, visa-issuing Barriers to cross-border practices, and consular services—declined moderately from movement of people in 2006 to 2012, and they remain substantially less than those high-income ECA imposed by the United States. However, obtaining a formal ­countries—as measured sector job is more difficult for immigrants in high-income ECA by visa requirements, countries than in the United States. The decline in visa restric- visa-issuing practices, tions in both the ECA and the United States has applied mostly and consular services— to nationals of high- and upper-middle-income countries, while declined moderately from 2006 to 2012. nationals from poorer countries have seen little decline in restric- tions on moving to high-income countries. In many of the region’s countries, limited integration of immigrants has impaired their contributions to host economies. For exam- ple, unemployment rates in the region tend to be higher among the foreign- born than the native-born population. Support for integration is even more important for refugees, who tend to take longer after entry to participate in the labor force. Although scores vary within the region, the ECA fares poorly on average when compared with other regions regarding policies supporting migrant integration (as measured by the Migrant Integration Policy Index). The average ECA country has had less success in integrating migrants than three of the four comparator countries selected for this exercise: Canada, the Republic of Korea, and New Zealand. (The ECA region has had more success than the fourth comparator country, Japan). The Central Europe subregion and Turkey have the worst performance, and while average scores in Western, Southern, and Northern Europe are closer to those of the best performers outside ECA, no ECA subregion has a score above the best-performing regions. In summary, ECA countries have been global leaders in cooperation through PTAs and BITs and are comparable to high-income countries in terms of poli- cies toward immigration. However, the average ECA country is more restrictive than countries in non-ECA regions regarding domestic regulations and migrant integration policies. The trend toward more open policies has slowed signifi- cantly, however, particularly since the first decade of this century. Little prog- ress has been made in tariff liberalization (as of the beginning of the 2000s), the use of BITs (as of the end of the 2000s), or reductions in FDI regulatory restrictions and product market liberalization (as of 2010). 38  ●   Critical Connections: Promoting Economic Growth and Resilience in Europe and Central Asia Considerable Scope Remains for Improving Policies to Increase Connectivity in Europe and Central Asia Over the past few decades, steps toward increasing connectivity have brought economic growth and greater productivity to most countries in the ECA region. It would be beneficial to develop these connections more broadly and deeply to support broad-based growth. This conclusion rests firmly on Critical Connections’ primary innovation: a multidimensional approach that examines the depth and breadth of ECA countries’ connections, both within the region and globally. It explicitly recognizes the complementarity of the individual channels of ­ connectivity: trade, FDI, migration, ICT, portfolio flows, and transport. The princi- pal message is that diversity in country connections and balance in all channels of international connectivity are critical for achieving the highest impact on growth and economic resilience. It is not enough to focus on a few countries for connectiv- ity or on one or two channels. The deepening of one channel can boost the impact of the other channels on growth. In addition to recognizing that channels are mutually supportive, multidimen- sional connectivity provides the following lessons: • ECA policies that build regional and global connectivity will stimulate robust growth. Being well connected in the global network of countries is important for long-run inclusive economic growth. • Multidimensional connectivity puts a spotlight on cross-border transfers of knowledge and ideas as the wellspring of sustained growth. An ongoing improvement in the stock of knowledge leads to increases in TFP that allow countries to get more output from the same amount of inputs. Knowledge flows that occur though connectivity are mostly “tacit”—that is, “learning by doing”—and not transmitted via books or blueprints. • As a result of the complementarity of the channels that link countries together, a balanced connectivity profile may be more important for knowledge spill- overs and growth than being well connected in a single connectivity dimension. Deep and comprehensive FTAs are a way to achieve this. • The number of connections a country’s economic partner has might be just as important as the type of connections. Not all partner countries are the same in this regard. Some partner countries have more connections than others, which makes them potentially better conduits for knowledge transfers. • In some cases, a country will be better off completing all channels of connectiv- ity to a poorly connected partner than building a single channel of connectivity to a well-connected country. • Connectivity’s being multidimensional implies that shocks in one dimension (say, migration) can have adverse effects in other dimensions (say, FDI and trade) as well. However, countries with the greatest connectivity are among those with the most resilience to shocks. The ECA region is a great laboratory for observing the role of multidimensional connectivity in action. Regional supply chains are strong, and links between coun- tries across the various forms of connectivity allow observation of how connectivity Critical Connections ●  39 opens doors for the knowledge transfers that support sustained growth. The varia- tion in connectivity between countries can be exploited empirically to explore which forms of connectivity matter the most. Although the ECA region as a whole has moved toward greater connectivity, progress has been uneven across countries. Lower trade barriers have not always been associated with fewer restrictions on immigration or product markets, and some countries still rely heavily on other ECA partners for global connectivity. Most higher-income countries have pursued complementary policies in most areas of connectivity, but complementary policies have been less prevalent in lower-­ middle-income countries (e.g., lower tariffs are not uniform across partner coun- tries and lose effectiveness because of more restrictive domestic regulatory regimes). Moreover, infrastructure linkages remain quite poor in some parts of ECA, particularly Central Asia. ECA is a diverse region, and the appropriate policy mix to promote multidi- mensional connectivity will vary from one country to the next. This study supports some general observations about the direction of policy, how- ECA is a diverse ever. Most obviously, countries can maximize their exposure to interna- region, and the tional knowledge flows and their ability to exploit their comparative appropriate policy advantages by maintaining low barriers to international transactions, mix to promote including low tariff rates, minimal constraints on inward or outward multidimensional FDI, and efficient procedures for border transactions. Multidimensional connectivity will vary connectivity suggests countries should not try to rely on one or two from one country types of connections; rather, they should develop a wide variety of to the next. mutually supportive outside links, including migration. In addition, partici- pation in deep multilateral trade agreements that support integration of services markets and the reduction of differences in rules governing product mar- kets would increase the impact of low barriers to international transactions on connectivity. Broad improvements to the domestic business climate can make opening up an economy to the outside world more beneficial. A host of policies, desirable in themselves because they increase domestic economic efficiency, offer improved connectivity as a bonus: strengthening institutions, boosting financial sector devel- opment, and ensuring flexible labor markets. Adopting international best practice for standards governing product markets, worker protections, and the environ- ment also tends to encourage international connectivity, particularly by making it easier to participate in global and regional supply chains. Infrastructure investments that focus on improving the efficiency of logistics ser- vices are critical for most forms of international connectivity. Recognizing how some connections are more meaningful can give countries a framework to help evaluate the costs and benefits of infrastructure projects. In most economies, trade in final goods relative to trade in services is not as dominant as it once was. So improve- ments in telecommunications, including reducing the cost and increasing the effi- ciency of internet connections, would support international commercial transactions and improve contacts with diasporas and other foreign sources of knowledge. Greater connectivity has two sides: the sending and the receiving countries. A lack of pertinent market information could keep the two from building busi- ness connections. Countries seeking to improve connectivity—in particular, 40  ●   Critical Connections: Promoting Economic Growth and Resilience in Europe and Central Asia those without a history of involvement with trade and FDI—may benefit from proactive policies that can help increase foreign investment and contacts with foreign firms. Investment promotion activities may be useful to encourage investors who lack sufficient information on domestic business opportunities or the policy regime and to reduce unnecessary or burdensome procedures for investment approvals. Policies should aim to boost the positive effects of connectivity at the firm level, which are associated with better performance. Encouraging skilled immigration may facilitate the introduction of the foreign management practices that increase productivity. Promoting linkages with foreign affiliates, both within ECA and out- side it, increases the kind of learning that improves the operation of local firms. Policies to help local firms acquire and absorb efficient global best practices could include increasing access to finance, management training, and supporting labor force mobility. Migration is a key element of multidimensional connectivity. Improving its benefits within the ECA will require policy reforms to better integrate migrants into the labor force and increased investment in education for all workers to cope with the ongoing transformation of work driven by technological change. From 2000 to 2010, the share of temporary migration rose in more than two-thirds of high-income ECA host countries. Temporary work has also been a more preva- lent feature of the labor market, reflecting greater connectivity and faster tech- nological change. Helping all workers benefit from a rapidly changing global economy involves moving away from social safety nets tied to long-term employ- ment toward general safety nets, allowing for more flexible contracts, invest- ments in education, and the removal of constraints on workers’ ability to move to find employment. Improvements in various types of connectivity are perhaps most critical for the lower-middle-income countries, particularly those in the South Caucasus and Central Asia. Because of both their geographic position and limited infrastructure, many of these countries are only weakly connected to other ECA countries and the global economy. The vast distances between Central Asia and Europe will remain an obstacle to connectivity. However, infrastructure investments and policies to improve integration through freer trade, infrastructure, and investment policies are likely to provide large growth benefits in Central Asia. The more diversified upper-middle-income countries, on the other hand, have a broader set of opportunities to improve connectivity. Participation in supply chains is strong in most of these countries, but balancing the currently uneven supply chain linkages, particularly low levels of imports of intermediate goods, versus already-high levels of exports of these goods, would support greater ben- efits from connectivity. A large set of policies—from removing barriers to trade and FDI to strengthening intellectual property protection and competitiveness reforms—are needed to improve participation in value chains and make the most of cross-border production opportunities. The high-income countries that recently entered the EU have established relatively open economies and strong domestic business climates in the context of deep integration within the EU. Nevertheless, further efforts are required to bring their domestic business climates to the level of the most advanced European countries. Most of the transition EU economies would benefit from Critical Connections ●  41 reducing the excessive economy-wide restrictiveness in product markets that are relevant from a ­connectivity dimension. This calls for a review of the state’s role in the economy and the extent to which regulatory regimes impede new firms’ entry into sectors. If a country has relatively closed markets internally, its external connectivity will be affected by the reduction in its attractiveness for inward FDI and participation in global supply chains. The ECA region’s once-confident march toward greater connectivity has for the most part stalled in the past decade. Voices are currently casting doubt on the wisdom of opening to the outside world. The economic benefits of deeper and more diverse connectivity, however, are strong—most notably, the knowl- edge transfers from trade, FDI, and migration that deepen participation in mul- tinational supply chains and lead to faster growth. By exploring multidimensional ­ connectivity and its impact, Critical Connections provides a framework for under- standing the benefits of and concerns about globalization and helps provide information for policy discussions and actions that recognize how the various aspects of connectivity might work together to deliver resilient and faster growth. 42  ●   Critical Connections: Promoting Economic Growth and Resilience in Europe and Central Asia Annex A.  Selected Indicators TABLE A.1  Multidimensional Connectivity Indexes (on an Absolute Basis) Global ranking, from best to worst, in combined connectivity (lower rankings indicate better connectivity) Multidimensional Country connectivity Trade FDI Migration ICT Airlines Portfolio flows ECA Germany 2 1 5 3 4 3 3 United Kingdom 4 6 2 4 2 1 4 Netherlands 5 10 3 14 12 8 14 France 6 5 6 5 5 4 5 Belgium 7 7 7 18 9 13 18 Italy 8 8 13 7 6 7 6 Spain 10 12 12 10 7 6 9 Switzerland 13 15 10 17 8 10 17 Ireland 14 16 11 29 14 14 31 Sweden 15 17 14 19 13 12 19 Poland 17 28 25 20 27 28 23 Austria 18 18 22 21 18 15 20 Russian Federation 22 19 37 13 23 30 13 Czech Republic 23 21 28 39 35 25 38 Hungary 25 25 26 43 37 33 42 Denmark 28 26 23 25 20 11 24 Luxembourg 29 38 4 61 39 37 57 Finland 30 22 24 31 30 21 30 Turkey 31 32 39 16 33 27 16 Portugal 32 31 29 28 25 20 29 Norway 33 33 21 22 22 16 21 Slovak Republic 35 34 41 52 53 68 51 Ukraine 40 42 64 48 67 66 47 Greece 42 39 40 26 26 22 27 Bulgaria 48 47 59 62 58 62 62 Lithuania 50 49 58 68 80 49 65 Croatia 51 53 54 56 49 57 54 Estonia 52 50 57 84 59 61 78 Belarus 54 65 101 67 100 100 63 Latvia 58 55 63 79 73 54 76 Cyprus 60 71 52 74 45 51 70 Malta 62 64 60 94 70 63 95 Kazakhstan 63 62 83 53 95 88 53 Bosnia and Herzegovina 68 68 76 65 64 80 82 Macedonia, FYR 77 69 74 95 79 79 96 Albania 78 80 87 76 72 75 90 Moldova 86 89 100 102 89 92 102 Georgia 97 103 102 99 102 104 97 Armenia 98 105 103 103 103 103 101 Kyrgyz Republic 99 96 95 106 101 101 104 Tajikistan 100 106 106 107 105 106 105 Azerbaijan 101 102 107 87 104 102 84 Other countries United States 1 2 1 1 1 2 1 China 3 3 8 6 15 19 7 Canada 9 9 9 9 3 5 8 Mexico 11 11 20 8 10 9 11 Japan 12 4 17 2 19 18 2 Singapore 16 14 19 42 29 29 40 Brazil 19 29 18 11 24 32 10 Malaysia 20 13 31 36 31 38 37 continued Critical Connections ● 43 TABLE A.1  continued Multidimensional Country connectivity Trade FDI Migration ICT Airlines Portfolio flows India 21 24 38 12 17 24 12 Indonesia 26 23 35 24 48 36 25 Thailand 27 20 34 32 43 31 32 Hong Kong SAR, China 34 27 16 33 16 26 33 South Africa 36 35 30 27 38 50 26 Argentina 37 37 27 30 42 48 28 Chile 38 41 33 40 50 59 39 Israel 39 36 46 37 28 56 35 New Zealand 41 45 32 41 21 35 41 Morocco 43 46 51 45 46 40 52 Peru 44 52 45 49 52 60 48 United Arab Emirates 45 43 53 34 47 42 34 Saudi Arabia 46 40 43 23 36 53 22 Egypt, Arab Rep. 47 57 49 47 55 52 46 Colombia 49 58 44 35 41 45 36 Nigeria 53 67 48 46 32 73 45 Tunisia 55 44 71 59 51 44 60 Trinidad and Tobago 56 59 61 70 62 69 77 Costa Rica 57 54 65 72 63 43 68 Pakistan 59 48 66 44 66 76 43 Dominican Republic 61 51 67 54 40 34 58 Algeria 64 56 81 38 56 55 44 El Salvador 65 60 73 57 34 46 72 Panama 66 77 50 75 77 64 74 Guatemala 67 61 77 58 44 58 64 Qatar 69 70 56 60 57 47 56 Bangladesh 70 63 92 51 97 94 50 Uruguay 71 73 55 77 71 77 71 Bahamas, The 72 84 42 92 65 17 89 Jordan 73 66 68 82 61 67 80 Mauritius 74 74 47 93 83 83 94 Ecuador 75 85 70 55 60 65 55 Ghana 76 78 82 83 78 85 83 Jamaica 79 87 62 63 68 39 79 Sri Lanka 80 76 91 69 93 93 66 Oman 81 72 84 64 54 41 59 Kenya 82 93 72 71 84 71 69 Bahrain 83 91 85 80 99 82 75 Lebanon 84 79 78 73 76 78 67 Cameroon 85 90 79 78 92 87 73 Mozambique 87 97 69 89 90 98 91 Kuwait 88 81 89 50 69 95 49 Gabon 89 94 75 88 88 97 86 Barbados 90 99 36 97 75 70 100 Syrian Arab Republic 91 75 88 66 74 81 61 Namibia 92 86 90 96 86 89 92 Paraguay 93 95 80 86 85 86 88 Guyana 94 92 86 85 81 84 108 Botswana 95 83 94 90 91 96 85 Swaziland 96 82 93 104 94 99 103 Ethiopia 102 100 99 81 98 90 81 Brunei Darussalam 103 88 96 91 96 91 87 Benin 104 98 104 101 107 105 99 Belize 105 101 97 105 87 74 106 Antigua and Barbuda 106 104 98 108 82 72 107 Burkina Faso 107 108 108 100 108 108 98 Afghanistan 108 107 105 98 106 107 93 Note: ECA = Europe and Central Asia; FDI = foreign direct investment; ICT = information and communication technology. 44  ●   Critical Connections: Promoting Economic Growth and Resilience in Europe and Central Asia TABLE A.2  Multidimensional Connectivity Indexes (on a Per Capita Basis) Global ranking, from best to worst, in combined connectivity (lower rankings indicate better connectivity) Multidimensional Country connectivity Trade FDI Migration ICT Airlines Portfolio flows ECA Luxembourg 1 2 1 1 1 3 32 Ireland 2 3 6 4 3 5 28 Netherlands 3 6 5 21 15 13 9 Belgium 4 4 4 6 7 19 79 Switzerland 6 5 8 2 2 7 7 Malta 7 23 7 35 23 6 53 Sweden 9 8 13 8 10 12 84 Norway 11 13 12 10 13 8 15 Cyprus 12 37 3 16 6 11 14 United Kingdom 13 19 15 11 9 17 5 Denmark 14 11 17 15 12 4 11 Finland 15 9 16 31 25 16 16 Austria 16 7 19 9 17 14 70 Germany 18 10 20 12 19 26 19 France 19 17 18 13 22 30 18 Czech Republic 21 12 28 29 34 34 30 Hungary 22 18 25 51 37 36 26 Spain 23 25 24 17 29 23 48 Estonia 24 20 30 23 27 22 1 Portugal 26 27 29 24 30 24 21 Slovak Republic 27 14 32 41 43 56 66 Italy 31 22 31 32 32 33 12 Lithuania 35 28 40 45 77 35 2 Latvia 37 32 41 20 55 21 60 Poland 38 34 36 47 45 49 34 Croatia 41 44 37 59 38 39 56 Greece 42 45 42 26 31 31 20 Bulgaria 45 43 45 71 51 50 62 Turkey 56 55 57 39 76 62 85 Belarus 58 48 73 30 83 100 98 Bosnia and Herzegovina 60 60 63 88 49 63 82 Macedonia, FYR 61 52 58 83 65 58 69 Russian Federation 63 49 67 43 73 80 40 Ukraine 71 62 76 27 87 81 104 Albania 75 79 75 37 60 48 75 Kazakhstan 76 68 83 25 99 90 99 Georgia 84 102 65 63 98 99 90 Moldova 86 81 94 48 62 68 80 Armenia 87 90 70 77 89 79 81 Kyrgyz Republic 95 99 97 50 100 95 94 Tajikistan 102 109 101 67 107 98 103 Azerbaijan 106 106 107 87 103 93 106 Other countries Singapore 5 1 11 14 11 20 10 Hong Kong SAR China 8 15 10 22 5 29 4 Bahamas, The 10 41 9 36 14 1 8 Canada 17 16 22 7 4 18 22 Barbados 20 85 2 49 28 10 63 Australia 25 35 23 3 21 44 17 Qatar 28 33 21 42 18 9 23 New Zealand 29 39 26 5 8 27 13 United States 30 30 27 18 26 46 64 Mauritius 32 57 14 52 66 52 6 Trinidad and Tobago 33 31 33 66 33 28 72 continued Critical Connections ● 45 TABLE A.2  continued Multidimensional Country connectivity Trade FDI Migration ICT Airlines Portfolio flows United Arab Emirates 34 29 34 55 39 37 77 Malaysia 36 21 43 54 40 60 65 Israel 39 24 46 19 24 51 61 Chile 40 46 35 58 63 66 29 Japan 43 26 50 75 57 71 24 Panama 44 59 38 53 52 40 73 Bahrain 46 42 49 72 70 42 37 Uruguay 47 73 39 40 56 59 3 Mexico 48 40 51 56 42 53 107 Thailand 49 38 55 78 82 69 36 Costa Rica 50 36 60 70 54 38 83 Argentina 51 66 44 34 61 74 27 Saudi Arabia 52 51 54 44 41 65 42 Oman 53 56 56 61 35 32 41 South Africa 54 53 53 64 68 82 39 Brazil 55 69 48 69 85 87 49 Gabon 57 80 47 46 81 75 33 Brunei Darussalam 59 50 71 60 67 43 78 China 62 54 62 113 88 96 89 Peru 64 74 59 85 72 77 71 El Salvador 65 63 61 80 16 45 86 Jamaica 66 83 52 68 46 25 88 Morocco 67 75 66 106 79 61 44 Dominican Republic 68 61 64 62 36 41 91 Tunisia 69 47 79 90 59 55 25 Jordan 70 65 68 33 44 54 51 Indonesia 72 71 77 99 97 89 43 Guyana 73 64 80 81 53 47 76 Philippines 74 58 84 79 80 86 45 Colombia 77 84 72 86 74 78 50 Swaziland 78 67 89 82 91 83 93 Antigua and Barbuda 79 92 69 28 20 2 58 Lebanon 80 77 74 103 64 57 31 Namibia 81 70 93 74 75 67 92 Kuwait 82 76 86 104 48 73 55 Botswana 83 72 92 109 92 76 87 Egypt, Arab Rep. 85 86 78 97 95 85 67 Guatemala 88 78 88 84 58 64 96 Belize 89 93 87 38 47 15 74 Ecuador 90 96 85 65 69 70 101 Algeria 91 82 98 107 93 72 52 Ghana 92 87 95 100 50 94 105 Paraguay 93 100 96 57 94 84 46 Nigeria 94 101 90 101 71 104 68 Cameroon 96 105 91 91 102 97 35 India 97 88 99 95 84 101 47 Zimbabwe 98 89 100 76 90 91 54 Sri Lanka 99 91 102 92 105 102 59 Mozambique 100 107 82 73 104 103 108 Kenya 101 103 81 94 86 88 95 Pakistan 103 97 103 111 78 106 57 Sudan 104 98 105 105 112 113 113 Syrian Arab Republic 105 94 104 108 96 92 102 Bangladesh 107 95 108 96 101 109 38 Benin 108 108 109 89 111 111 100 Togo 109 104 113 93 108 105 97 Ethiopia 110 111 110 112 109 107 112 Afghanistan 111 113 106 110 106 110 110 Niger 112 112 111 98 113 112 111 46  ●   Critical Connections: Promoting Economic Growth and Resilience in Europe and Central Asia Notes 1. Transmission of shocks is not new. The bubonic plague of 542 CE, which decimated the Byzantine Empire, is thought to have arrived in Constantinople (today’s Istanbul) by way of the Silk Road. The spice trade was also accompanied by struggles for economic dominance as wars were fought, lands were colonized, and fortunes were made and lost. 2. The full study is available at http://www.worldbank.org/en/region/eca/publication /­critical-connections. 3. Note that the OECD average includes the European Union countries. 4. In a seminal contribution, Coe, Helpman, and Hoffmaister (1997) identify that by trad- ing with industrial countries with a large “stock of knowledge” accumulated through R&D activities, developing countries boosted their productivity by importing intermedi- ates and capital goods that embodied knowledge and information. Van Pottelsberghe de la Potterie and Lichtenberg (2001) identify that FDI, and in particular outward FDI, has also been a conduit for R&D spillovers for 13 industrial countries (including 11 EU member countries). References Bernstein, William J. 2008. A Splendid Exchange: How Trade Shaped the World. New York: Grove. Bloom, Nicholas, Kalina Manova, John Van Reenen, Stephen Teng Sun, and Zhihong Yu. 2018. “Managing Trade: Evidence from China and the US.” NBER Working Paper 24718, National Bureau of Economic Research, Cambridge, MA. http://www.nber.org​ /­papers​/­w24718. Coe, David T., Elhanan Helpman, and Alexander W. Hoffmaister. 1997. “North–South R&D Spillovers.” Economic Journal 107: 134–49. Parthesius, Robert. 2010. Dutch Ships in Tropical Waters: The Development of the Dutch East India Company (VOC) Shipping Network in Asia 1595–1660 . Amsterdam: Amsterdam University Press. Rodrik, Dani. 2018. “Populism and the Economics of Globalization.” Journal of International Business Policy 1 (1–2): 12–33. https://link.springer.com/article/10.1057/s42214​ -018-0001-4. Romer, Paul. M. 1990. “Endogenous Technological Change.” Journal of Political Economy 98: S71–S102. Starr, S. Frederick. 2015. Lost Enlightenment: Central Asia’s Golden Age from the Arab Conquest to Tamerlane. Princeton, NJ: Princeton University Press. van Pottelsberghe de la Potterie, Bruno, and Frank Lichtenberg, 2001. “Does Foreign Direct Investment Transfer Technology across Borders?” Review of Economics and Statistics 83 (3): 490–97. World Bank. 2012. Golden Growth: Restoring the Lustre of the European Economic Model. Washington, DC: World Bank. ———. 2018. Moving for Prosperity: Global Migration and Labor Markets. Policy Research Report. Washington, DC: World Bank. T he countries of the Europe and Central Asia region, along with much of the rest of the world, find themselves engaged in a revival of one of the fundamental questions of economic policy: how much to open to the rest of the world. This question now dominates the political economy of the region, not just within the advanced economies of the European Union but also among the region’s emerging market economies. In Critical Connections, the World Bank offers new research on the process of economic integration, showing its potential benefits without ignoring the downsides. The report examines how trade, investment, migration, and other linkages among countries drive economic growth in the Europe and Central Asia region. It breaks new ground by using a multidimensional approach that recognizes how each connectivity channel is likely to be affected by the strength of other channels. The multidimensional view offered by this approach makes it clear that diversity in country connections and balance in all channels of connectivity are critical for achieving the greatest impact on growth and economic resilience. Europe and Central Asia provides a great laboratory for observing the role of multidimensional connectivity in action. The region’s 47 countries vary widely in the degree of openness of their economies. Its collective experience shows how the various elements of cross-border connectivity work together to accelerate progrowth knowledge transfers, which in turn boost productivity through participation in today’s global value chains. Which countries a country has as its economic partners might be just as important as the type of connection it has with them, because being well connected to highly connected countries can provide benefits beyond being well connected to comparatively isolated countries. Although greater connectivity can expose countries to external shocks, the report presents a fact-based argument for policies that seek to build deeper and more diverse connections within the Europe and Central Asia region and globally. The message is timely. Europe’s once-confident march toward economic integration has slowed over the past decade, with voices in many countries questioning the wisdom of opening to the global economy. Critical Connections serves as a reminder to citizens and policy makers that greater regional and global connectivity has been a tremendous “convergence machine,” raising living standards in lower-income countries toward those of wealthier middle- to high-income countries. By exploring multidimensional connectivity and its impact, the report provides a framework for understanding the many benefits and challenges of globalization and helps provide information for policy discussions and actions that recognize how the various aspects of connectivity might work together to deliver resilient and faster growth. Europe and Central Asia Studies Europe and Central Asia Studies feature analytical reports on the main challenges and opportuni- ties faced by countries in the region, with the aim of informing a broad policy debate. Titles in this regional flagship series undergo extensive internal and external review prior to publication. SKU 33229