Duc Minh Pham, Claire Honore Hollweg, Brian Mtonya, Deborah Elisabeth Winkler, and Thuy Nguyen Vietnam: Connecting value chains for trade competitiveness December 2019 Duc Minh Pham, Claire Honore Hollweg, Brian Mtonya, Deborah Elisabeth Winkler, and Thuy Nguyen Vietnam: Connecting value chains for trade competitiveness December 2019 © 2020 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW, Washington DC 20433 Telephone: 202-473-1000; Internet: www.worldbank.org This work is a product of the staff of The World Bank with external contributions. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endor­ sement 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. All queries on rights and licenses should be addressed to the Publishing and Knowledge Division, The World Bank, 1818 H Street NW, Washington DC 20433, USA; fax: 202-522-2625; email: pubrights@ worldbank.org. Cover and layout: hoanghaivuong ii Vietnam: Connecting value chains for trade competitiveness Contents Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix Abbreviations .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x Executive summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi Chapter 1. Toward trade-oriented connectivity policy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Chapter 2. A new approach for value-chain-based connectivity. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 2.1. Selecting key value chains���������������������������������������������������������������������������������� 9 2.2. Identifying value-chain links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 2.3. Defining the spatial structure of value chains . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 2.4. Value-chain-based connective propensity and key corridors .. . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 2.5. Critical consolidated transport network for ten selected value chains .. . . . . . . . . . . . . . . 24 Chapter 3. Efficient international trade gateways. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 3.1. Overview of trade gateways. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 3.2. Trade by gateway type . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 3.3. Analysis of top trade gateways. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 Chapter 4. Regional specialization and coordination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 4.1. Provincial specialization .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 4.2. Aligning trade and growth strategies with regional specialization. . . . . . . . . . . . . . . . . . . . . 42 4.3. Core versus lagging regions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 Chapter 5. Economic zones and value chains . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46 5.1. Industrial agglomeration/concentration through economic zone development and value-chain development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 5.2. Economic zones and clusters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50 5.3. How economic and industrial zones can contribute to value-chain creation and development .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 Chapter 6. Implementing trade-oriented connectivity and competitiveness policies . . . . . . . . . . 56 6.1. Making connectivity policy and transport investment more robustly trade-oriented by integrating the comprehensive value-chain connectivity assessment and trade gateways analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 6.2. Establishing an efficient mechanism for coordinating trade and transport connectivity and gvc policies, and implementing policy recommendation one��� 58 6.3. Securing firm-level data for qualified multisectoral policy analyses on trade, transport, and GVCs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 Annex 1. Analysis of the textile and garment value chain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 A1.1. Industry overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 A1.2. Value-chain links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 A1.3. Spatial structure and value-chain mapping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 A1.4. Value-chain-based connectivity and key corridors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72 Contents iii Annex 2. Analysis of the leather and footwear value chain .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 A2.1. Industry overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 A2.2. Value-chain links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 A2.3. Spatial structure and value-chain mapping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 A2.4. Value-chain-based connectivity and key corridors .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 Annex 3. Analysis of the electronics value chain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82 A3.1. Industry overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82 A3.2. Value-chain links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83 A3.3. Spatial structure and value-chain mapping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84 A3.4. Value-chain-based connectivity and key corridors .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88 Annex 4. Analysis of the automotive value chain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90 A4.1. Industry overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90 A4.2. Value-chain links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 A4.3. Spatial structure and value-chain mapping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93 A4.4. Value-chain-based connectivity and key corridors .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96 Annex 5. Analysis of the wood products value chain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97 A5.1. Industry overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97 A5.2. Value-chain links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98 A5.3. Spatial structure and value-chain mapping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99 A5.4. Value-chain-based connectivity and key corridors .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103 Annex 6. Analysis of the rubber value chain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105 A6.1. Industry overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105 A6.3. Spatial structure and value-chain mapping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106 A6.4. Value-chain-based connectivity and key corridors .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108 Annex 7. Analysis of the rice value chain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111 A7.1. Industry overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111 A7.2. Value-chain links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112 A7.3. Spatial structure and value-chain mapping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112 A7.4. Value-chain-based connectivity and key corridors .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115 Annex 8. Analysis of the coffee value chain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117 A8.1. Industry overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117 A8.2. Value-chain links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118 A8.3. Spatial structure and value-chain mapping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118 A8.4. Value-chain-based connectivity and key corridors .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120 Annex 9. Analysis of the fruit and vegetable value chain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123 A9.1. Industry overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123 A9.2. Value-chain links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124 A9.3. Spatial structure and value-chain mapping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125 A9.4. Value-chain-based connectivity and key corridors .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126 Annex 10. Selected value chains, their segments, and industrial codes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129 References ������������������������������������������������������������������������������������������������������������������������������� 133 iv Vietnam: Connecting value chains for trade competitiveness BOXES, FIGURES, MAPS AND TABLES BOXES Box 2.1. Distinction between value chain, supply chain, and industrial cluster .. . . . . . . . . . . . . . . . . . . . . . . . 13 Box 2.2. Industrial classification and harmonized commodity coding system. . . . . . . . . . . . . . . . . . . . . . . . . . . 15 Box 3.1. Electronics value chain and samsung Vietnam. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 Box 4.1. Policies for different types of lagging regions.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 Box 5.1. China’s experience with economic zones and clusters: “top-down” versus “bottom-up”�������������������������������������������������������������������������������� 51 Box 5.2. Policies supporting links and the role of SEZs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 Box 5.3. Employment and skills training in uganda (E4D/SOGA). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 Box 6.1. Visualization of cluster mapping in the United States. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63 FIGURES Figure 1 (Box 2.1). Apparel value chain versus supply chain������������������������������������������������������������ 13 Figure 1.1. Export-led growth and poverty reduction (1992-2017). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  3 Figure 1.2. Structural change in technology embodied in export (1997-2017).. . . . . . . . . . . . . . . . . . . . . . . . . . . . .  5 Figure 1.3. Structural change in total export by value chain (1997-2017).. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  5 Figure 1.4. Quality of trade-related infrastructure versus trade per capita. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  7 Figure 2.1. Methodological overview.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  9 Figure 2.2. Value chain selection based on revealed comparative advantage (RCA) and trade performance. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 Figure 2.3. Aquaculture value-chain links, inputs-outputs table 2016.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 Figure 2.4. Refined links: aquaculture value chain. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 Figure 2.5. Locational distribution of the aqua-culturing segment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 Figure 2.6. Locational distribution of the fishing segment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 Figure 2.7. Locational distribution of the aqua-processing segment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 Figure 2.8. Connective model for a value chain. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 Figure 3.1. Trade value by gateway type (2011-2016). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 Figure 3.2. Share of trade value by gateway type (2011-2016). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 Figure 3.3. Products imported through sea gateways (2011 and 2016). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 Figure 3.4. Top 15 products exported through sea gateways (2011 and 2016). . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 Figure 3.5. Products exported through air gateways. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 Figure 3.6. Products imported through air gateways. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 Figure 3.7. Top 12 gateways by trade value. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 Figure 3.8. Six most significant gateways by trade value.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 Figure 4.1. Regional specialization of Ha Nam province.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 Figure 4.2. Manufacturing agglomeration versus provincial income. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 Figure 4.3. Manufacturing agglomeration versus provincial trade. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 Figure 4.4. Manufacturing agglomeration versus poverty. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 Figure 6.1. Integrated four-pillar framework for trade facilitation and logistics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 Contents v Figure A1.1. Employment in T&G . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 Figure A1.2. T&G exports. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 Figure A1.3. Decomposition of the T&G export. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 Figure A1.4. Trade balance of up- and mid-stream T&G segments. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 Figure A1.5. T&G value-chain links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66 Figure A1.6. T&G value-chain segments. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 Figure A1.7. Locational distribution of the yarn segment.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68 Figure A1.8. Locational distribution of the fabric segment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68 Figure A1.9. Locational distribution of the clothing segment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69 Figure A1.10. Locational distribution of the other garment segment.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 Figure A2.1. Employment in leather and footwear . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 Figure A2.2. Leather and footwear exports.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 Figure A2.3. Leather and footwear value-chain links.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 Figure A2.4. Leather and footwear value-chain segments.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 Figure A2.5. Locational distribution of the leather segment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 Figure A2.6. Locational distribution of the handbags and other leather segments.. . . . . . . . . . . . . . . . . . . . . . . . . 77 Figure A2.7. Locational distribution of the footwear segment.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78 Figure A3.1. Employment in the electronics sector. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82 Figure A3.2. Electronics exports compared to other sectors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82 Figure A3.3. Decomposition of the electronics gross export. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83 Figure A3.4. Electronics value-chain links. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83 Figure A3.5. Electronics value-chain segments. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84 Figure A3.6. Locational distribution of the electronics components segment.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85 Figure A3.7. Locational distribution of the final 3C products segment.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85 Figure A4.1. Employment in the automotive industry.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90 Figure A4.2. Automotive sector trade value. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 Figure A4.3. Decomposition of the automotive gross export. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 Figure A4.4. Automotive value-chain links. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92 Figure A4.5. Automotive value-chain segments. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92 Figure A4.6. Locational distribution of the auto parts and components segment. . . . . . . . . . . . . . . . . . . . . . . . . . . 93 Figure A4.7. Locational distribution of the modules and final assembly segment. . . . . . . . . . . . . . . . . . . . . . . . . . . 94 Figure A5.1. Employment in wood products manufacturing.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97 Figure A5.2. Rapid export growth in wood products. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97 Figure A5.3. Decomposition of wood products’ gross export.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98 Figure A5.4. Wood products value-chain links. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98 Figure A5.5. Wood products value-chain segments. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99 Figure A5.6. Locational distribution of the planting and foresting segment.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99 Figure A5.7. Locational distribution of the sawmilling segment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100 Figure A5.8. Locational distribution of the wood products and furniture segment. . . . . . . . . . . . . . . . . . . . . . . 101 Figure A6.1. Employment in the rubber industry. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105 Figure A6.2. Rubber sector exports. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105 Figure A6.3. Rubber value chain links.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106 Figure A6.4. Rubber value-chain segments. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106 vi Vietnam: Connecting value chains for trade competitiveness Figure A6.5. Locational distribution of the rubber planting segment .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107 Figure A6.6. Locational distribution of the rubber processing segment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107 Figure A6.7. Locational distribution of the final rubber products segment .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108 Figure A7.1. Employment in the rice sector . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111 Figure A7.2. Rice exports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111 Figure A7.3. Rice value-chain links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112 Figure A7.4. Rice processing value-chain segments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112 Figure A7.5. Locational distribution of the rice planting segment .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113 Figure A7.6. Locational distribution of the rice processing segment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113 Figure A8.1. Employment in the coffee industry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117 Figure A8.2. Coffee exports .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117 Figure A8.3. Coffee value-chain links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118 Figure A8.4. Coffee processing value-chain segments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118 Figure A8.5. Locational distribution of the coffee planting segment .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119 Figure A8.6. Locational distribution of the coffee processing segment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119 Figure A9.1. Employment in the fruit and vegetable sector . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123 Figure A9.2. Fruit and vegetable exports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123 Figure A9.3. Fruit and vegetable value-chain links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124 Figure A9.4. Fruit and vegetable processing value-chain segments .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125 Figure A9.5. Locational distribution of the fruit and vegetable planting segment . . . . . . . . . . . . . . . . . . . . . . . . 125 Figure A9.6. Locational distribution of the fruit and vegetable processing segment . . . . . . . . . . . . . . . . . . . . 126 MAPS Map 2.1. Spatial structure of the aquaculture value chain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 Map 2.2. Connective propensity of the aquaculture value chain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 Map 2.3. Connective propensity of ten export-oriented value chains . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 Map 3.1. Main trade gateways . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 Map 4.1. Locational distribution of provincial specialization in the north .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 Map 4.2. Locational distribution of provincial specialization in the south . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 Map 5.1. Spatial structure of industrial zones versus the textile and garment value chain . . . . . . . . . 48 Map 5.2. The aquaculture value chain and economic zones .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50 Map A1.1. Geographic distribution of the T&G value chain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71 Map A1.2. Connective propensity of the T&G value chain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 Map A2.1. Geographic distribution of the leather and footwear value chain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 Map A2.2. Connective propensity of the leather and footwear value chain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 Map A3.1. Geographic distribution of the electronics value chain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87 Map A3.2. Connective propensity of the electronics value chain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89 Map A4.1. Geographic distribution of the automotive value chain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 Map A4.2. Connective propensity of the automotive value chain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96 Map A5.1. Geographic distribution of the wood processing value chain .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102 Map A5.2. Connective propensity of the wood products value chain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104 Map A6.1. Geographic distribution of the rubber value chain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109 Contents vii Map A6.2. Connective propensity of the rubber value chain. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110 Map A7.1. Geographic distribution of the rice value chain. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114 Map A7.2. Connective propensity of the rice value chain. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116 Map A8.1. Geographic distribution of the coffee value chain. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121 Map A8.2. Connective propensity of the coffee value chain. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122 Map A9.1. Geographic distribution of the fruit and vegetable value chain.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127 Map A9.2. Connective propensity of the fruit and vegetable value chain.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128 TABLES Table ES.1. Policy priorities for value chain connectivity and trade competitiveness ������������������ xxii Table 2.1. Industrial priorities through 2025 with a vision toward 2035 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 Table 2.2. Spatial structure of the aquaculture value chain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 Table 2.3. Key corridors of the aquaculture value chain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 Table 3.1. Vietnam’s main gateways . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 Table 4.1. Regional specialization of ca mau province . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 Table 5.1. Share of establishments, employment, and revenue of firms in zones . . . . . . . . . . . . . . . . . . . . . . . . 48 Table 5.2. The aquaculture value chain and related industrial and economic zones .. . . . . . . . . . . . . . . . . . . 50 viii Vietnam: Connecting value chains for trade competitiveness Acknowledgments This report was prepared by a World Bank’s team led by Duc Minh Pham (Senior Economist) and consisted of Claire Honore Hollweg (Senior Economist), Brian Mtonya (Senior Economist), Deborah Elisabeth Winkler (Consultant), and Thuy Nguyen (Consultant) with contributions from Sebastian Eckardt (Lead Economist), Jung Eun Oh (Senior Transport Specialist), Douglas Zhihua Zeng (Senior Economist), and Charles Kunaka (Lead Private Sector Specialist). Research assistance was provided by Phan Cong Duc (Consultant), Hoang Hong Diep (Consultant), and Nguyen Cuong (Consultant). This report is a product of the Second Australia - World Bank Partnership Program and serves as a key analytical input for the Vietnam Development Report 2019, “Connecting Vietnam for Growth and Shared Prosperity”. The team is grateful for the guidance of Hassan Zaman (Regional Director; Equitable Growth, Finance and Institutions Practice; East Asia and the Pacific), Ousmane Dione (Country Director for Vietnam), Deepak Mishra (Practice Manager; Macro Economics, Trade and Investment Practice), Irina Astrakhan (Practice Manager; Finance, Competitiveness and Innovation Practice), Jacques Morisset (Program Leader and Lead Economist), and Achim Fock (former Operations Manager for Vietnam). Richard Record (Lead Economist), Gerald Paul Ollivier (Lead Transport Specialist), Nguyen Thang (Director, Centre for Analysis and Forecast, Vietnam’s Academy of Social Sciences), Vu Thanh Tu Anh (Director of Research, Fulbright Economics Teaching Program in Ho Chi Minh City), and Mombert Hoppe (Senior Economist) kindly acted as peer-reviewers and we thank them for their comments. The team would wish to acknowledge the strong support and contribution from Vietnam Institute of Development Strategy (VIDS), especially comments and advice provided by Mr. Nguyen Van Vinh, Vice President, Mr. Nguyen Huu Khanh, Ms. Nguyen Quynh Trang and other experts of the VIDS. The team would like to express its gratitude to the Australian Embassy’s team of the Second Australia – World Bank Group Strategic Partnership in Vietnam, especially Justin Baguley, Cain Roberts and Nguyen Linh Huong for financing this work and for their strong support and helpful guidance. The report benefited from editorial support provided by Jessica Wholey, administrative assistance from by Le Thi Khanh Linh and Dinh Thi Hang Anh, and communications support from Nguyen Hong Ngan and Le Thi Quynh Anh of the World Bank Vietnam Country Office. Acknowledgments ix Abbreviations ACIC ASEAN Common Industrial Classification AFTA ASEAN Free Trade Area ASEAN Association of Southest Asia Nations CP-TPP Comprehensive Progressive Trans-Pacific Partnership DDC Direct value-added contribution DOIT Department of Industry and Trade DPI Department of Planning and Investment DVA Domestic value addition E&E Electronic and Electric EAP East Asia and Pacific FDI Foreign Direct Investment FVA Foreign value addition GDP Gross Domestic Product GDVC General Department of Vietnam Customs GSO General Statistics Office GVCs Global Value Chains HCMC Ho Chi Minh City HS Harmonized System I/O Input-output data ICDs Inland Clearance Depots ICT Information and communication technology IDC Indirect value-added contribution ISIC International Standard Industrial Classification IWT Inland waterways LPI Logistics Performance Index LQ Location quotient MNEs Multinational enterprises MOC Ministry of Construction MOF Ministry of Finance MOIT Ministry of Industry and Trade MOST Ministry of Science and Technology MOT Ministry of Transportation MPI Ministry of Planning and Investment NR National Road NTFC National Trade Facilitation Committee OD model Origin-destination model OECD Organization for Economic Cooperation and Development PC People’s Committee RCA Revealed Comparative Advantage Index RIM Reimported intermediates SDPs Small development projects SEDS Socio-Economic Development Strategies SEZs Specialized Economic Zones SI Sourcing intensity SMEs Small and medium enterprises T&G Textile and Garment TDSI Transport Development and Strategy Institute US-BTA United States and Vietnam Bilateral Trade Agreement VDR Vietnam Development Report VIA Vietnam Industry Agency VIDS Vietnam Institute of Development Strategy VSIC Vietnam Standard Industrial Classification WB World Bank WCO World Customs Organization WTO World Trade Organization x Vietnam: Connecting value chains for trade competitiveness Executive summary Vietnam’s export-led growth strategy and global integration are among the key factors behind the country’s remarkable achievements in growth and poverty reduction over the last two and a half decades. During this period, Vietnam’s per capita income increased nearly fourfold and poverty was reduced from around 53 percent in 1992 to 2 percent in 2016. Vietnam has become one of the most open economies in the world with a trade-to-GDP ratio of 187.52 percent in 2018. Merchandise export growth averaged more than 15 percent per annum in the last ten years; nearly five times the global export growth. The country’s export basket has improved in its technological content and has diversified in both its geographic destination and its product mix. There are nevertheless challenges that continue to confront Vietnam’s export performance. Many of Vietnam’s manufacturing exports have low domestic value addition, where Vietnam performs primarily assembly functions. Trade costs remain high compared to the average regional level. Domestic firms’ participation in key global value chains (GVCs) is limited, and instead, export performance is largely driven by the foreign direct investment (FDI) sector, accounting for more than 70 percent of total exports. Vietnam will likely be able to maintain its high export performance even if these challenges are not addressed, but there is scope for Vietnam to benefit even more from trade. Trade competitiveness – one key element of a productive economy – can be enhanced in three key ways, among others: (i) lowering trade costs associated with policy barriers to trade, (ii) improving the efficiency and reliability of transport infrastructure, and (iii) enhancing the integration of domestic production into GVCs as indicated in the three-pillar policy framework on trade competitiveness in Pham, Mishra, Chong et al. (2013). This report looks at the last two pillars, recommending policies to support trade competitiveness by improving the efficiency of transport infrastructure to enhance Vietnam’s ability to integrate into GVCs1. To facilitate GVC integration, Hollweg, Smith, and Taglioni (2017) recommended a comprehensive set of policy measures: (i) domestic infrastructure upgrading and logistics regulatory environment improvement, (ii) a more private-sector financed and integrated approach to developing transport corridors, (iii) a more liberal stance on FDI, (iv) a general reduction in the costs of doing business, (v) more transparent and predictable border procedures, (vi) and better connectedness with regional sources of demand and technology-related investment. This report scrutinizes a more specific aspect of pro-trade connectivity for GVC integration and complements the Hollweg, Smith, and Taglioni (2017) recommendations by offering a new analytical 1 This report does not cover trade cost issues, which are addressed in other World Bank papers, including Pham and Oh (2018), and Pham, Artuso et al., (2018). Executive summary xi approach developed according to the value-chain-based connectivity assessment under the World Bank 2013 Report. It does so by collecting and analyzing data on geographic arrangements, links and connectivity of production, and export activities along key value chains of comparative advantage for Vietnam. The principal objective is to identify policy actions to improve connective efficiency for and prioritized investment in pro-trade transport infrastructure to enhance trade competitiveness and GVC integration. The report is organized into six chapters beginning with an introduction to trade-oriented connectivity policy. Chapter two then outlines our approach to assessing value-chain-based connectivity whereas chapters three through six focus on efficient international trade gateways, regional specialization and coordination, economic zones in relation to value chains, and implementing trade-oriented connectivity policy, respectively. The key findings are organized into five main policy recommendations with detailed recommen- dations, objectives, policy actions, lead agencies, timeframe for implementation, and outputs, summarized in Table ES.1. Policy Recommendation One: Make connectivity policy and transport investment more robustly trade oriented by integrating comprehensive value-chain connectivity assessment and trade gateways analysis. At present, the objectives of improving trade growth competitiveness are not clearly linked with the objectives of developing connectivity policies and investment in transport infrastructure. Trade information, especially on value chains, is rarely used in policy formulation and implementation. There remains a lack of in-depth analysis on spatial structure and connective propensity along various linked segments of value chains to inform relevant policies and investment for transport infrastructure development. Related policies, transport master plans, and investment priorities should be formulated and implemented to more strongly support trade. Chapter two provides a new four-step methodology for a comprehensive value-chain connectivity and competitiveness assessment that identifies corridors and gateways critical for key domestic export-oriented value chains. These corridors are defined based on spatial structure of input-output links, industrial concentration, and hierarchical connective propensity, linking all segments of value chains with international trade gateways. This important information could guide related policies for and investment in transport infrastructure for the most effective support to enhance trade competitiveness and improve GVC integration. The report identifies key trade corridors and gateways for ten selected value chains with national comparative advantage, good trade performance, and governmental priority, including textile and garment, leather and footwear, electronics and electrical equipment, motor vehicles, wood products, rubber, rice, coffee, and fruits and vegetables. Subsequently, chapter three scrutinizes international trade gateways and their trade flows and structure, showing the share of total trade via air gateways has increased rapidly from 15.6 percent in 2011 to 39.5 percent in 2016, while the share of total trade via sea gateways plummeted from 78.8 percent in 2011 to 56.1 percent in 2016. This reflects the drastic shift in export structure from xii Vietnam: Connecting value chains for trade competitiveness primary exports including crude oil and non-oil (coal, stone, sand, gravel, aluminum, copper, etc.) and resource-based exports (agriculture-based products) to high-tech exports (electronics, cell phones, incorporated circuits, etc.). This structural change—rapid increases in small but valuable products, like mobile phones, electronic components, high fashion exports, and high-value, processed agricultural products—requires the transport system and its investments supporting exports to consider a shift from logistics perspectives, based not only on trade growth but also (and more important) on structural change and developing domestic value chains. Later, chapter six suggests formalizing the comprehensive value-chain connectivity and competitiveness assessment and trade gateway analysis into new transport and trade strategies. Authorities would need to formalize these analyses and appoint lead agency and research institutions to regularly conduct them, guide interagency coordination, and integrate their outputs and outcomes into trade policy, export-import strategy, and national and provincial socio-economic development strategies (SEDS) and master plans. One proposed action is to incorporate the information and policy analyses on trade flows and key value chains into the transport strategy for 2030. Trade-related indicators should be factored into renewed transport strategy to better benchmark Vietnam against international practice and to monitor policy implementation. Key trade indicators would include trade cost reduction, and Vietnam’s improved position in the interrelationship between efficient connectivity, measured by the quality of trade- related infrastructure, and trade development, measured by trade per capita, etc. In conjunction, the import-export strategy for 2030 should also be renewed to incorporate trade- related infrastructure factors including transport and logistics policies. Similarly, Vietnam should consider including infrastructure-related indicators like trade-related transport capacity (road, airway, seaway and ports, railway) and logistics performance indicators into import-export strategy to promote this critical policy coordination. In selecting ten value chains to demonstrate the four-step methodology for the value-chain-based connectivity and competitiveness analysis in chapter two, the report used existing datasets to produce empirical results. Looking forward, when scrutinizing structural changes in developing GVCs, policy makers should also account for mega trends that may disrupt GVCs, notably the acceleration of the digital transformation and associated de-globalization process. Over the medium to long term, GVCs will consolidate, with fewer countries and firms participating. Automation may result in reshoring manufacturing and therefore the comparative advantage of cheap labor enjoyed by low-and- middle-income countries (LICs) like Vietnam may be quickly eroding. In other words, infrastructure investments should not only support current economic activities (and, therefore, inevitably reinforce the current economic structure), but also be forward-looking and consider emerging trends and future developments. The proposed methodology allows for close monitoring with dynamic change in spatial structure and connective propensity of existing and emerging value chains in Vietnam. Policy makers may face some trade-offs when using information on GVC-based connectivity for master planning given limited resources and capacity. For example, developing connective infrastructures and gateways to support electronics GVCs may come at the expense of aquaculture value chains. Executive summary xiii This is already happening in Vietnam, when infrastructure in the Mekong River Delta does not keep pace with rapidly rising demand, while in the North, activities along some highways are relatively low. Policy Recommendation Two: Establish an efficient mechanism for coordinating trade and transport connectivity and GVC policies proposed in Recommendation One. It is vital to establish an effective interagency coordination mechanism to implement the recomme­ ndation for multisectoral policies and investment related to pro-trade transport infrastructure and GVC integration. This mechanism should be put under a four-pillar framework for trade facilitation and logistics and enhanced trade competitiveness as referenced in Pham and Oh (2018). Chapter six recommends the National Committee for National Single Window, ASEAN Single Window and Trade Facilitation (below reffered to as National Trade Facilitation Committee - NTFC) should take the lead in coordinating trade, trade-related transport, and GVC policies by providing strategic direction and guidance, and supervising related multisectoral policies, particularly for delivering Policy Recommendation One. This committee was established by the Prime Minister’s Decision 1899 /QD-TTg dated April 10, 2016, chaired by Deputy Prime Minister Vuong Dinh Hue, with senior representatives from 20 line ministries, primarily to comply with the WTO’s Trade Facilitation Agreement (TFA). More important, this committee has been assumed to coordinate multiagency efforts to facilitate trade, reduce trade costs, and improve trade competitiveness. In response to the World Bank’s policy recommendation (Pham and Oh, 2018), the Prime Minister issued the Decision 684/QD-TTg dated June 4, 2019 to revise and supplement the Decision 1899/ QD-TTg by adding the role to coordinate interagency efforts on national logistics development. This additional function makes the upgraded NTFC a perfect body to coordinate multisectoral policies on trade, traded-related transport and connectivity, and GVC development for trade competitiveness as proposed in the four-pillar framework and Policy Recommendation One. Chapter six further suggests strengthening this mechanism by recommending the committee appoint an interagency taskforce to assist managing assumed task in the above-mentioned Recommendation One. Policy Recommendation Three: Secure firm-level data for qualified multisectoral policy analyses on trade, transport, and value chains. Chapter six advises relevant data sets should be in place with appropriate and regularly updated statistical indicators on value chains and gateways to ensure reliable policy analysis informs connectivity policy and investment in trade-related infrastructure. Much of the information and data needed for such analyses is missing and/or difficult to collect. This is partly due to a new approach that requires complex datasets and time for statistical systems to respond, but more important due to strict regulations for disclosing raw and firm-level data. Chapter six proposes issuing relevant regulations to make firm-level trade and transport data available for the value-chain connectivity and competitiveness assessment and trade gateways analysis, as well as establishing an effective mechanism for better data collection, processing, and coordination at national and sectoral statistics levels among the General Statistics Office (GSO), the Ministry of Transport, the General Department of Customs, and others to supplement the data. Innovative methodologies using big data for real-time analysis should be explored toward this modern policy formulation process. xiv Vietnam: Connecting value chains for trade competitiveness The analyses proposed in chapters two and three would use considerable firm-level data to address research questions and inform key findings. Disaggregated data would be combined from various sources, including: (i) input-output data for value-chain link identification, (ii) enterprise data (per industry, per province, per commodity, per industrial park, per industry, etc.) for capturing regional concentration of domestic supply chains, (iii) transportation data and origin-destination (OD) flows (both within supply-chain structure and between cluster locations and trade gateways), and (iv) border/port trade data (land gateway, seaport, and airport with the Harmonized System (HS) code of export and import volume). Because this value-chain analysis is critical for businesses and the private sector, the report recommends building an information point with convenient access to publicly available information on value-chain links and spatial structure, including but not limited to geographic location of value- chain links, provincial specialization, international gateway statistics, etc. To be sustainable, such a multidisciplinary, cluster-development data center would require strong interagency coordination and a government-private sector partnership. Optimally, this center would be managed by a government agency, strongly motivated to use the data (which could oversee development master planning, competitiveness enhancement, and connectivity policy and investments). For coordinating data inputs, this agency would be mandated to work with the various sources of this data (GSO, customs, transport, other development partners, etc.), and empowered to manage data sharing with the private sector. Preferably, the center would be overseen by the National Trade Facilitation Committee. Regular updates and visualization of industrial concentration indicators and commodities flows is useful for all stakeholders, policy makers, academics, researchers, and businesses alike. Chapter six recommends sharing the information on provincial specialization for all concerned parties including central and local governments, the private sector, and development partners. The information on value-chain links, spatial structure, and connectivity is important not only for policy formulation, but also for the private sector to proactively participate in domestic supply chains as a significant part of GVCs. This is especially essential in Vietnam where more than 90 percent of the domestic private sector are small firms that lack this information and have weak links to foreign invested firms. The information on value-chain links, spatial structure, and connectivity could be made available on a cluster-mapping website following a U.S. model, with information collected and analyzed via the comprehensive value-chain connectivity assessment and trade gateways analysis, using big data in real time. It would be developed and shared publicly, for both policy makers and the private sector not only to implement Policy Recommendation One but also to fulfill Vietnam’s e-government initiative. Vietnam should consider developing a similar cluster-mapping project as in the United States, with data sources and operational organizations properly ascribed. In addition to a cluster website, online freight-flow modeling can be developed based on OD flow data. The website and online freight-flow modeling would provide dynamic, visual information for governments and businesses to understand and shape the competitive landscape for a range of industries. The website would also help local governments understand local specialization and regional comparative advantages to promote strategic investment and lay the groundwork for new industries. Executive summary xv Policy Recommendation Four: Consider regional specialization and inter-regional cooperation in transport infrastructure investment policy. Chapter four shows manufacturing density and value-chain concentration have a positive interrelationship with income, exports, and employment in local areas. Regional specialization measured by location quotient (LQ) is a dynamic metric, changing over time. Some provinces change their participation in the value chain for various reasons. For example, emerging electronics value- chain expansion in Thai Nguyen Province and Bac Ninh Province resulted from the foreign-led firm Samsung’s move to Vietnam. Other provinces change their specialization because of increasing competition in obtaining necessary skilled labors. The information on local specialization is important for understanding the geographic structure of value chains. The government needs this information to formulate policy in regional coordination integrating value-chain links, and for planning human capital in appropriate regions and areas accordingly. An example of spatial structural analysis using this method of calculating LQ on the aquaculture value chain shows that, although aqua-culturing is spread across the country, this export-oriented value chain that has a spatial layout along key segments including aquaculture, capture, and export processing, is concentrated mainly in the South, especially in the Mekong Delta. The value-chain connectivity follows a specific trend of sequential transport connection, reflecting the input-output relationship of production, and especially the connection between export processing location and related national trade gateways. This report recommends integrating effective regional value-chain connectivity into regional development planning and implementation. Moreover, decisions on transport infrastructure investment should be based on a conducive environment for regional specialization and inter- regional cooperation (rather than unhealthy competition) for public investment sources. Another example of dynamic specialization is Ha Nam Province, which currently specializes in the textile and garment, electronics, cement, auto-parts, and animal food subsectors. Between 2011 and 2016, textile and garment specialization has moved away from fabric and clothing and more toward yarns. Meanwhile, electronics emerged during this period. This specialization change would require a shift in labor force skills, and a conducive policy environment for increased competition in the labor force. Remarkably, during this same period, Ha Nam Province has become more specialized in cement production, but with fewer local inputs. This may reflect a shortage of local inputs or the provincial government’s increasing awareness of environmental protection. Changes in provincial specialization can create opportunities for lagging regions like poor and remote provinces. This report shows the garment segment of the textile and garment value chain has shifted from Red River Delta provinces (Hai Duong, Bac Ninh, Ha Nam) to a lagging province (Tuyen Quang) between 2011 and 2016. The shift in provincial specialization needs to be reviewed more closely in separate focused studies. This report recommends developing and implementing a National Action Plan for Inter-regional Linkages and Coordination, built on an in-depth assessment of static and dynamic provincial specialization and an analysis of inter-regional links and connectivity, and based on spatial structure and developing key value chains. Among other things, the Action Plan should direct public investment xvi Vietnam: Connecting value chains for trade competitiveness accordingly and avoid unnecessary regional competition for inefficient and fragmented public investment by each province. Policy Recommendation Five: Industrial and economic zones should support the development of domestic supply chains for better GVC integration. Chapter five uses value-chain analysis to differentiate between industrial agglomeration and concentration through value chains versus economic concentration in economic zones. The characteristics and spatial structure of industrial parks and economic zones are far different than those of value chains. Industrial and economic zones are organized in specific areas designed for multiple sectors, and sometimes apply preferential policies for firms located within them. Instead, the spatial structure of a value chain usually spans larger geographic areas with many more actors, and preferential policies are not usually applied to entire value chains. The difference in the spatial structure and the policy disparity inside and outside industrial and economic zones can prevent or constrain links across the entire value chain, as international experience shows (Zeng, 2010). Establishing the needed interactions between various segments of value chains, physically and institutionally, is one challenge for economic zones models. Furthermore, zones are often located near big cities or important transport corridors, especially in the final corridor connecting the export processing point to the international trade gateway. This report emphasizes the necessity of rethinking and modernizing industrial and economic zones to make them best support domestic value-chain links and connectivity. Policies should be enacted so industrial and economic zones are developed to promote better GVC integration. Zone policies also need to be revised and supplemented to promote cluster development. Cluster links, while they may not fully reflect the entire value chain, include important value-chain link(s) that require spatial and policy priorities to facilitate input-output links of the broader chain. Policies should also address the impacts of urbanization and spontaneous zone development along main transport corridors. Chapter five provides international examples from China and other countries of developing cluster-based economic zones and facilitating localized cluster growth. This report provides several policy recommendations and actions. First, we recommend restructuring industrial and economic zones so they best support value-chain-based connectivity and competitiveness. This would require forming an industrial and economic zones development plan accounting for comparative advantages of each province and region, and integrating them into national, regional, and provincial master plans. In addition, industrial and economic zones regulations should be revised to drive better GVC integration. Second, in the longer term, the government should prioritize leveraging industrial and economic zones to attract FDI and accelerate industrial and trade growth. To achieve this, an FDI promotion plan should identify appropriate types of FDI to be attracted with sectoral priority. Furthermore, the government should design an industrial and trade strategy and development plan considering regional comparative advantages, value-chain development, and upgrading Vietnam’s position in GVCs. This strategy would then be integrated into national, regional, and provincial master plans. Executive summary xvii Last but not least, Vietnam should identify existing industrial and economic zones that are not fully occupied and promote them as “greenhouses” for potential industrial clusters based on the host province’s advantages and industrial agglomeration (LQ index), and reflect this direction in national, regional, and provincial master plans. TABLE ES.1. Policy priorities for value chain connectivity and trade competitiveness Objective Policy action Lead agency Time Outputs 1. Policy Recommendation One: Make connectivity policy and transport investment more robustly trade oriented by integrating comprehensive value-chain connectivity assessment and trade gateways analysis. 1.1. Renew the transport Incorporate trade-related indicators MOT, MOIT, 2020-2021 - A new transport strategy for 2030 in transport strategy, including but MPI strategy to include links not limited to (i) reducing trade - Inputs for ten- between transport and costs (“doing business”, etc.); year social development of trade (ii) improving Vietnam’s position and economic and key value-chains. in the interrelationship between development efficient connectivity (measured strategy (SEDS) by the quality of trade-related 2021-2030 infrastructure Logistics Performance Index (LPI)) and trade development (measured by trade per capita). 1.2. Renew the import- Incorporate infrastructure-related MOIT, MPI, 2020-2021 - A new import- export strategy for 2030 indicators in import-export strategy, MOT export strategy to include trade-related including but not limited to (i) trade- - Inputs for ten-year infrastructure factors related transport capacity (road, SEDS 2021-2030 (transport and logistics). airway, seaway and ports, railway), (ii) trade-related logistics (LPI). 1.3 Formalize a - Issue an instruction to formalize MPI (VIDS), Annually - Regular value- comprehensive value- a comprehensive value-chain MOT (TDSI), or every chain-based chain connectivity connectivity assessment and MOIT (VIA), two years connectivity and assessment and trade trade gateways analysis by MARD competitiveness gateways analysis that appointing lead agency and analysis. collects and analyzes research institutions to regularly - Value-chain-based data on geographic conduct these studies, and guide connectivity arrangements of interagency coordination. analysis to be production and export - Integrate value-chain connectivity integrated in trade activities along key value assessments into trade policy and policy and export- chains of comparative export-import strategy. import strategy. advantage for Vietnam, - Integrate value-chain connectivity - Value-chain-based and key international assessments into regional and connectivity gateways into transport provincial socio-economic analysis to be and trade strategies. development strategy and master a component plans. in regional and provincial SEDS (ten year) and master plans (five year) xviii Vietnam: Connecting value chains for trade competitiveness Objective Policy action Lead agency Time Outputs 2. Policy Recommendation Two: Establish an efficient mechanism for coordinating trade and transport connectivity and GVC policies proposed in Recommendation One. 2.1. Establish an effective - Consolidate the current National NTFC, MOIT, 2020-2021 - A consolidated mechanism for Trade Facilitation, Logistics MARD, MPI, NTFC implementing Policy and National Single Window MOT - Relevant NTFC Recommendation One. Committee (NTFC) to coordinate decisions trade, trade-related transport, and GVC policies, to provide strategic direction and guidance, and to supervise related multisectoral policies. - Appoint a special taskforce or secretary to assist the NTFC in managing multisectoral policy coordination for trade, trade- related transport, and GVCs. 2.2. Establish an interagency - Issue a decision for establishing NTFC, MOIT, 2020-2021 A taskforce under taskforce under the the special taskforce or secretary. MARD, MOT, NTFC oversight NTFC to implement - Formalize the development GSO, Customs actions under Policy and implementation of a firm- Recommendation One. level data system to support multisectoral policy analyses on trade, trade-related transport, and GVCs. 3. Policy Recommendation Three: Secure firm-level data for qualified multisectoral policy analyses on trade, transport, and value chains. 3.1. Develop a relevant Promulgate regulations to GSO, Customs, 2020-2021 PM’s decision making dataset with appropriate make data available for a value- MOIT, MARD, a dataset available and regularly updated chain-based connectivity and MOT (TDSI) for a value-chain- statistical indicators competitiveness analysis, including based connectivity on value chains to (i) adding missing indicators in and competitiveness ensure reliable policy enterprise surveys and census analysis analysis and appropriate (OD freight flows information, investment in trade- etc.), (ii) adding missing indicators related transport in customs data (trade volume, infrastructure. OD information, etc.), (iii) adding freight flows data, and (iv) exploring innovative methods using big data for real-time analysis. 3.2. Share publicly the Develop and maintain a GSO, 2020-2021 - A cluster-mapping information on Vietnamese cluster-mapping Customs, website value-chain links, website like the U.S. model, and MOIT, MARD, spatial structure, and collect and analyze information MOT connectivity, especially under the value-chain-based for the private sector to connectivity and competitiveness proactively participate analysis. in GVCs. Executive summary xix Objective Policy action Lead agency Time Outputs 3.3. Establish a digital - Develop a pilot project for a MOF (GDVC), 2021-2023 - A demonstrated traceability system digital traceability system. MOC, MOST, digital traceability for key value chains - Regulate enterprises’ MOIT, MARD, system to enable relevant responsibility in goods’ origin business - Regulations stakeholders to collect traceability. associations on enterprises’ and analyze value-chain of key value responsibility data and improve the chains in goods’ origin performance of the traceability supply chains. 4. Policy Recommendation Four: Consider regional specialization and inter-regional cooperation in transport infrastructure investment policy. 4.1. Develop and implement Develop an in-depth assessment MPI, NTFC 2020-2021 - National Action a National Action Plan for on static and dynamic provincial taskforce, DPIs Plan for Inter- Inter-regional Links and specialization and an analysis on regional Links and Coordination taking into inter-regional links and connectivity Coordination account spatial structure that integrates spatial structure and - Inputs for regional and development of key development of key value chains as and provincial value chains. foundation for the National Action master plans Plan for Inter-regional Links and - Inputs for ten-year Coordination. SEDS 4.2. Direct public investment Prioritize and integrate MPI, MOIT, 2020-2021 Inputs for regional based on the National development of some high-value MARD, and provincial master Action Plan for Inter- agriculture processing chains and MOF, NTFC plans regional Links and manufacturing chains into the taskforce, PCs Coordination. National Action Plan for Inter- regional Links and Coordination to avoid inefficient and fragmented public investment by each province. 5. Policy Recommendation Five: Industrial and economic zones should support the development of domestic supply chains for better GVC integration. 5.1. Restructure industrial - Formulate an industrial and MPI, DPIs 2020-2021 - Industrial and and economic zones so economic zones development economic zones they best support value- plan accounting for comparative development plan chain-based connectivity advantages of each province and - Inputs for national, and competitiveness. region, and integrating them into regional, and national, regional, and provincial provincial master master plans to best support plans value-chain-based connectivity and competitiveness. - Revise industrial and economic zones regulations to drive better GVC integration. xx Vietnam: Connecting value chains for trade competitiveness Objective Policy action Lead agency Time Outputs 5.2. Prioritize leveraging - Develop an FDI promotion plan MPI, DPIs, 2020-2021 - FDI promotion plan industrial and economic to identify appropriate types of MOIT, MARD - Industrial and trade zones for FDI attraction FDI to be attracted with sectoral strategy to accelerate industrial priority. - Inputs for national, and trade growth. - Design an industrial and trade regional, and strategy and development plan provincial master considering regional comparative plans advantages, value chain development, and Vietnam’s improvement in GVCs and integrate it into national, regional, and provincial master plans. 5.3. Select existing industrial Identify industrial and economic MPI, DPIs, 2020-2021 - List of industrial and economic zones zones that are not fully MOIT, DOITs and economic as “greenhouses” for occupied and promote them as zones and their potential industrial “greenhouses” for industrial clusters potential industrial clusters based on based on the host province’s clusters provincial industrial advantages and industrial - Inputs for national, agglomeration (denoted agglomeration, and reflect this regional, and by LQ index). direction in national, regional, and provincial master provincial master plans. plans that consider emerging industrial clusters. Executive summary 1 CHAPTER 1 Toward trade-oriented connectivity policy This chapter recommends making connectivity policies and transport investments more robustly trade- oriented and integrating comprehensive value-chain analyses. 2 Vietnam: Connecting value chains for trade competitiveness Transport and connectivity policies, master plans, and investment priorities should be formulated and implemented to support trade more strongly. At present, the objectives of improving trade growth and competitiveness are not clearly linked with the objectives of developing connectivity policies and transport infrastructure investment. Global empirical evidence shows trade promotes growth. And by promoting growth, trade openness can be an important driver of poverty reduction. The export-led growth strategy adopted by a number of countries in the East Asia region provides powerful examples, including Vietnam’s. Trade has played a particularly remarkable role in economic growth and poverty reduction over the past three decades. Figure 1.1 illustrates the link between trade (denoted by export-to-GDP ratio), the growth trend (reflected by GDP per capita), and poverty reduction (represented by poverty gap at US$ 1.90 a day) for an extended period, from 1992 to 2017 in Vietnam. FIGURE 1.1. Export-led growth and poverty reduction (1992-2017) 2,500 GDP per capita in constant 2010 US$ Poverty gap at $1.90 per day (2011PPP) (%) 100 Export of goods and services to GDP (%) 2,000 80 1,500 52.9 60 1,000 38 35.5 40 26.5 19.5 14.8 500 20 4.2 2.8 2.7 2 - - 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 GDP per capita (in constant 2010 US$) Export of goods and services to GDP (%) Log. (Poverty gap at $1.90 a day (2011 PPP) (%)) Source: World Development Indicators (WDI). In the last nearly three decades Vietnam has become one of the most open trade countries in the world. The export-to-GDP ratio has steadily increased from 34.7 percent in 1992 to more than 100 percent in 2017, excluding the 2009-2011 drop during the global financial crisis. Imports have increased alongside exports, and trade openness, measured by the ratio of trade to GDP, reached nearly 200 percent in 2017. Vietnam’s spectacular achievement has resulted largely from trade liberalization underpinned by their removal of trade and nontariff barriers committed in a number of regional trade agreements, like the Association of Southeast Asian Nations (ASEAN) Free Trade Area (AFTA) in 1996, the Bilateral Trade Agreement between the United States and Vietnam (US-BTA) in 2000, the World Trade Organization (WTO) in 2006, and the Comprehensive Progressive Trans-Pacific Partnership (CP-TPP) in 2017. Chapter 1 – Toward trade-oriented connectivity policy 3 As a result of this remarkable trade performance, Vietnam’s GDP per capita measured by constant price in 2010 US$ has increased from less than US$ 500 in 1992 to more than US$ 1,800 in 2017, nearly four-fold during this period. Similarly, Vietnam is a leading example of how trade can contribute to significant poverty reduction. The poverty rate of Vietnam has decreased remarkably during the same period. At the threshold poverty line of US$ 1.90 a day, the poverty headcount as percentage of Vietnam’s total population has decreased from nearly 52.9 percent in 1992 to 2.0 percent in 2017. Vietnam’s trade grew alongside its deepening global integration and participation in GVCs, as discussed in Hollweg, Smith and Taglioni (2017). Since 1995, Vietnam has shown higher integration as a buyer and a seller in a number of GVCs. An improved business environment for attracting qualified foreign direct investments (FDI) has enabled Vietnam to upgrade in GVCs. Vietnam grew its domestic value addition embodied in its increase in gross exports by 16.6 percent annually between 1995 and 2011 – just below what had been achieved by China. This combined integration and upgrading has created better jobs, supported economic growth, and reduced poverty. Despite remarkable achievement during the past two-and-half decades, challenges remain. First, many of Vietnam’s higher-value manufacturing exports have high value content but low domestic value addition. Second, Vietnam is facing a situation in which export performance is largely driven by the FDI sector, which contributes up to 70 percent of total exports, while domestic firms contribute only 30 percent of total exports. The dual-track economy has resulted from weak value-chain links and the limitation of domestic firms’ participation in key GVCs. In addition, there is low productivity growth and weak competitiveness in the domestic private sector, especially small and medium enterprises (SMEs). Third, Vietnam’s exports are strong on quantity but weak on quality. Although a leader in export quantities of several agricultural commodities, they are often low quality, hence sell at low unit prices. Last but not least, Vietnam’s trade costs are higher than the ASEAN average in logistics and compliance costs, with complicated on-border and behind-the-border regulations according to annual Doing Business report of the World Bank (2019). These four challenges have contributed to Vietnam’s weak trade competitiveness. Trade competitiveness – one key element of a productive economy – can be enhanced in three key ways, among others: (i) lowering trade costs associated with policy barriers to trade, (ii) improving the efficiency and reliability of transport infrastructure, and (iii) enhancing the integration of domestic production into GVCs as indicated in the four-pillar policy framework on trade competitiveness in Pham, Mishra, Chong et al. (2013). This report looks at the last two pillars, recommending policies to support trade competitiveness by improving the efficiency of transport infrastructure to enhance the country’s ability to integrate into GVCs2. Scrutinizing the export growth from trade-related connectivity perspectives, Vietnam’s export structure has changed dramatically in terms of technology embodied in export and product-based export. 2 This report does not cover trade-costs issues, which are addressed in other World Bank papers, including Pham and Oh (2018), and Pham, Artuso et al. (2018). 4 Vietnam: Connecting value chains for trade competitiveness Figure 1.2 shows significant structural change in technology embodied in export between 1997 and 2017 based on the five international classifications of technology level of exports: low tech, medium tech, high tech, primary products, and resource based3. The share of low-tech exports, medium-tech exports, and resource-based products has not changed significantly during this period. However, the proportion of primary products including crude oil and non-oil (coal, stone, sand, gravel, aluminum, copper etc.) declined markedly, mainly FIGURE 1.2. Structural change in technology due to a decline in crude oil export, and embodied in export (1997-2017) export-control policy limiting minerals and raw material exports. Furthermore, there 100% has been a dramatic shift from primary 80% and resource-based products (agriculture- based products) to high-tech exports 60% (electronics, cell phones, incorporated 40% circuits, etc.). This trend is closely associated 20% with significant investment projects of some leading multinational firms including 0% Samsung Group and Intel Corporation, 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 who selected Vietnam as a production Resource Based Primary Products Medium Tech Low Tech High Tech base for mobile phone and tablet products for export worldwide. Source: UNComtrade. Figure 1.3 shows the share of key exports FIGURE 1.3. Structural change in total export by (the ten selected value chains discussed in value chain (1997-2017) chapter two) has changed over time. 80% Despite increasing in absolute values, the Share in Total Export 60% share of agricultural exports (aquaculture, rice, vegetables and fruit, and coffee) 40% decreased, from over 20 percent in 1997 to about 10 percent in 2017. Meanwhile, other 20% traditional exports from labor-intensive value chains like leather and footwear and 0% textile and garment have maintained their 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 share in total exports (around 10 percent Wood Textile and garment Autotomotive Rice from leather and footwear and about 20 Co ee Leather and footwear Electronics Rubber Vegetables and fruit Aquaculture percent from textile and garment). The share of these ten value chains in total Source: UNComtrade. exports has increased from 60 percent to 75 percent in this period, confirming their increasing importance in the economy in general and exports in particular. The FDI sector plays a crucial role in these value chains’ export, accounting for 60 percent of total textile and garment exports, 70 percent of total leather and footwear exports, and 100 percent of electronics exports. 3 UNComtrade. Chapter 1 – Toward trade-oriented connectivity policy 5 There are two important implications concerning this structural change. First, Vietnam is at the low end of GVCs, performing primarily assembly functions with limited value addition, and weak links from FDI firms to the domestic private sector. Second, since exports matter for connectivity, a change in product- based export structure should be considered when prioritizing better connectivity and logistics for higher-value product export. A transport system which promotes exports should account for this shift in logistics perspectives, especially for prioritized investment in appropriate trade gateways. Clearly, Vietnam is in a transition to diversify and move up the chain to higher, value-added functions, and to support backward integration of domestic, input-supplying firms. Policies for export-led growth and enhanced competitiveness in general and for improved connectivity specifically will need to become more value-chain based, moving toward higher-value products. A value-chain-based approach will allow a comprehensive consideration to promote trade competitiveness, including policies to facilitate GVC integration. Along with asset protection, seamlessly connecting factors across or within borders is one of two fundamental needs of the private sector to participate in GVCs and is a key driver of international competitiveness. Lead firms must be able to move component parts quickly, reliably, and cheaply between different points in the production network and assembly facilities. However, the logistics needs vary depending on the composition of exports (Hollweg, Smith, Taglioni, 2017). For example, in agri-processing, logistics are needed to move higher value-added goods for export and/or domestic processing, and competitiveness depends a lot on the quality of domestic infrastructure. In apparel, logistics are needed to reduce lead times for assembly activities. In information and communication technology (ICT), which relies on lightweight or digital inputs, logistics in soft and hard infrastructure are necessary. FIGURE 1.4. Quality of trade-related infrastructure versus trade per capita 6.00 Quality of trade and transport related 5.00 Germany Netherlands UK Luxembourg infrastructure ( 2016 Score) Japan USA South Africa 4.00 Turkey India Thailand Egypt Brazil 3.00 Vietnam Kazakhstan 2.00 Bhutan Lao PDR Iraq SomaliaHaiti Guinea 1.00 - 150 1,500 15,000 150,000 2015 Trade per capita (Log ) Source: LPI, WDI, and calculation by authors. 6 Vietnam: Connecting value chains for trade competitiveness Efficient connectivity is essential for trade development as shown in Figure 1.4, which illustrates the interrelationship globally between efficient connectivity, measured by the quality of trade and transport-related infrastructure,4 and trade development, measured by trade per capita. Vietnam’s score on quality of trade-related infrastructure as measured by the logistics performance index (LPI) in 2016 was 2.7, roughly equal to the average global level (2.75) but below the East Asia and Pacific (EAP) level (3.02) when including the high-income countries. It was, however, higher than the average EAP level (2.58) when excluding the high-income countries. From this global landscape, Vietnam’s quality of trade and transport-related infrastructure apparently has not been able to keep pace with the EAP’s trade growth and development. Vietnam typically performs well on connectivity metrics, but EAP is an increasingly competitive environment, and Vietnam will need to upgrade its performance quickly and comprehensively (Hollweg, Smith, Taglioni, 2017). Vietnam’s trade competitiveness potential is constrained by the lack of policy aimed toward promoting trade-oriented connectivity. This study looks at the spatial dimension of Vietnam’s domestic participation in value chains, to identify key value-chain sectors, as well as policy recommendations that can support domestic links to emerge within these sectors with a spatial consideration5. 4 Quality of trade-related infrastructure is one key indicator surveyed by the World Bank under the Logistics Performance Index (LPI) for 160 countries. 5 Given the breadth of issues covered and the limited resources, this report focuses on domestic connectivity only. Another study on external trade linkages and connectivity, including within ASEAN or CLMV, is planned. Chapter 1 – Toward trade-oriented connectivity policy 7 CHAPTER 2 A new approach for value-chain-based connectivity This chapter proposes a new approach for value-chain-based connectivity and competitiveness that collects and analyzes data on geographic arrangements of production and export activities along key value chains of comparative advantage for Vietnam. 8 Vietnam: Connecting value chains for trade competitiveness Currently, trade information, especially on value chains, is rarely used in formulating connectivity policy and transport investment in Vietnam. There remains a lack of in-depth analyses on spatial structure and connective propensity along various linked segments of value chains to inform respective policy and investment. The principal objective is to identify policy actions to improve connective efficiency for investment into pro-trade transport infrastructure in order to enhance trade competitiveness and GVC integration. FIGURE 2.1. Methodological overview The proposed value-chain-based connectivity assessment is built on Pham, Mishra, Cheong • Selecting value chains of high comparative et al.’s methodological approach in their advantage and trade performance that Selection review of Vietnam’s trade competitiveness contribute to trade growth, industrialization, and global value chain integration (2013). Following this approach, there are four integral steps, summarized in Figure 2.1. • Identifying domestic input - output linkages and operational structure of selected value Hierarchical structure of a value chain is defined Linkages chains by input-output links, and spatial industrial • De ning the spatial structure of selected concentration of all segments is measured Spatial value chains based on linkages, regional by location quotients (LQs) within the value Structure specialization, and gateway analysis chain. This report adopts the concept of local • Outlining the connective propensity of cluster (Porter, Michael E., 2000 and http:// Connective selected value chains based on spatial www.clustermapping.us), but focuses on the Propensity structure and linkages and quantifying supply chain-based corridors domestic production part of value chains including companies, and suppliers among other cluster’s stakeholders.6 Value-chain Source: Built on Pham, Mishra, Cheong et al., 2013. connective propensity follows the sequential links of all value chains segments to international trade gateways based on product mix and the structure of export-oriented domestic production. Value-chain connective propensity is important information that could guide transport infrastructure policies and related investment for the most effective support to enhance trade competitiveness and GVC integration. 2.1. Selecting key value chains The first step in this methodological approach is to analytically identify and select key value chains where Vietnam is already competitive. The three criteria for selecting key value chains are: (i) high trade performance and high importance in the economy, (ii) high comparative advantage, and (iii) governmental priority. The idea is, if Vietnam’s performance in these value chains is improved, there is potential to enhance national competitiveness and promote Vietnam’s continued participation in GVCs. Value-chain trade performance is measured by (i) the sector’s share in total exports and imports nationally and globally, (ii) the average annual growth rate of the sector’s exports and imports 6 See the distinction between value chain, supply chain, and industrial cluster in Box 2.1 below. Chapter 2 – A new approach for value-chain-based connectivity 9 nationally, and (iii) and the sector’s net trade balance nationally. In addition, domestic value addition, estimated by OECD-WTO in “Trade in value-added database” (https://stats.oecd.org/index. aspx?queryid=75537), is also considered when reviewing the value-chain trade performance. Value-chain comparative advantage is measured by the revealed comparative advantage index (RCA)7 and Lafay index. The RCA index indicates relative advantage of a certain product as evidenced by a country’s trade flows in that product relative to global trade flows in that product. It is calculated as the ratio of the product’s national export share to the product’s global export share. For an RCA > 1 (the country has a comparative advantage in exporting the product and for an RCA < 1, a disadvantage. The Lafay index reveals comparative advantage of an industry. If the index has a value of more than zero, the country being reviewed has a comparative advantage on global competitors in exporting the industry’s products. (An index value of less than zero shows a country does not have a comparative advantage). Since it takes account of exports and imports, it is suitable for some export-oriented industries that have high imports. Figure 2.2 shows Vietnam’s key products by trade performance and RCA. FIGURE 2.2. Value Chain Selection Based on Revealed Comparative Advantage (RCA) and Trade Performance 40 87 Vehicles other than railway, tramway 85 Electrical, electronic equipment 72 Iron and steel 08 Edible fruit, nuts, peel of citrus fruit, melons Annual growth of export value, 2011 - 2016 (%) 30 16 Meat, sh and seafood food preparations nes 61 Articles of apparel, accessories, knit or crochet 20 62 Articles of apparel, accessories, not knit or crochet 64 Footwear, gaiters and the like, parts thereof 10 09 Co ee, tea, mate and spices 0 40 Rubber and articles thereof 94 Furniture, lighting, signs, prefabricated buildings -10 73 Articles of iron or steel 03 Fish, crustaceans, molluscs, aquatic invertebrates nes 10 Cereals -20 27 Mineral fuels, oils, distillation 44 Wood and articles of wood, wood charcoal products, etc -30 0 1 2 3 4 5 6 7 8 9 10 11 RCA Index, 2016 Size of bubble: export value, 2016 Source: Trade Map, 2016. 7 The revealed comparative advantage of commodity i of country j in a given period t is calculated as: xij xtj RCA= Xiw Xtw where xij is the export value of commodity i of country j, xtj: total export value of country j, Xiw: export value of commodity i of the world, and Xtw: total export value of the world. 10 Vietnam: Connecting value chains for trade competitiveness The Government priorities can be found in Prime Minister’s Decision No.879/QD-TTg dated June 9, 2014 approving Vietnam’s industrial development strategy through 2025, with a vision toward 2035 (Table 1.1), and Prime Minister’s Decision No.32/QD-TTg dated January 13, 2015, approving an integrated program to develop and upgrade value chains and clusters that are considered to be of competitive advantage. TABLE 2.1. Industrial priorities through 2025 with a vision toward 2035 Industrial priority Through 2025 Toward 2035 Agri-processing - Fisheries and aquatic processing - Wood processing Light manufacturing (garments and - Auxiliary materials - Fashion footwear) - Luxury products Electronics and telecommunications - Computers - Telephones and components Mechanicals - Agricultural machinery - Shipbuilding - Nonferrous metal Chemicals - Basic chemicals - New materials - Plastic and rubber products - Pharma-chemicals - Petrochemicals Energy - New and renewable (wind, solar, - Renewable (geothermal, wave, biomass) nuclear) Fourteen value chains met these three criteria: textile and garment, leather and footwear, electronics, automotive vehicles, wood products, rice, aquaculture, coffee, rubber, fruit and vegetable and, cement, iron and steel, and oil and gas. We selected ten of those (aquaculture, textile and garment, leather and footwear, electronics, wood products, motor vehicles, rubber, rice, coffee, fruit and vegetable) for analysis because they are agricultural and manufacturing value chains with high RCA and increasing share in total exports (more than 70 percent of total exports in 2017). This final selection has nothing to do with traditional industrial policy perspectives. Instead, it is to ensure, while covering these “backbone” value chains of the economy, the analytical result could build on sufficient evidence to support policy recommendations to promote the connectivity associated with these competitive sectors. As an example, the aquaculture value chain is analyzed later in the chapter, using the four-step approach (summarized in Figure 2.1). Detailed analyses for other selected value chains can be found in Annexes 1-9. 2.2. Identifying value-chain links The second step is to determine the production structure to identify the domestic links of the ten selected value chains by analyzing domestic input-output (I/O) data. This study limits its focus to the domestic “production” links of the value chains rather than the entire value chain, which would also include service-related and external links. The domestic links are identified following four stages: (i) determine first-tier supplying sectors (a sector’s backward links) using data from Vietnam’s I/O tables from Enterprise Censuses 2011 and 2016; (ii) repeat the exercise in the first step multiple times to Chapter 2 – A new approach for value-chain-based connectivity 11 compute second-, third- or lower-tier supplying sectors; (iii) diagram the value-chain links; and (iv) refine links and diagrams based on expert views and sectoral data. Stage one, determining first-tier supplying sectors, is based on a sector’s sourcing intensity, computed as the share of inputs from a supplying sector as percent of total intermediate inputs. We consider both imported and domestically purchased inputs, since both are combined in the I/O data for 2011 and 2016. Lastly, we only consider nonservices and noncapital inputs. In most cases, we apply a sourcing intensity threshold of 2%. That is, we only consider non-services and noncapital inputs that represent at least 2% of total inputs. Sourcing intensity (SI) is defined as follows: SIs,i = ( purchase of input s by sector i total intermediate inputs used by sector i ) * 100% where i is a key value-chain sector, and s is a sector supplying inputs to the key value-chain sector. The second stage repeats the first stage for various layers of backward links. That is, it considers the sectors that supply inputs to sector s. Since SIs of lower-tier sectors can vary across value chains, we selected the most important supplying sectors ad-hoc, focusing on the most critical in the value chain. In other words, once the most important first-tier supplying sectors are identified, this process is repeated for the most critical inputs, all the way back to the third-tier supplier. The third stage is to develop a diagram of value-chain links for each of the value-chain sectors as well as the first-, second-, and third-tier supplying sectors. Figure 2.3 shows the diagram from the aquaculture value chain for 2016. FIGURE 2.3. Aquaculture value-chain links, inputs-outputs table 2016 Processed and preserved sh, crustaceans and molluscs (IO 36) Producer Live sh, freshwater, marine, farmed Live sh, freshwater, marine, not farmed First (IO 27) (IO 26) tier SI = 47.6% SI = 22.9% supplier Prepared animal feeds Live sh, freshwater, marine, farmed Second (IO 46) (IO 27) tier SI = 51.1% SI = 33.6% supplier Maize and other cereals Grain mill products, starches, Edible roots and tubers Third (IO 2) and starch products (IO 40) with high starch (IO 3) tier SI = 28.1% SI = 17.3% SI = 13.7% supplier Source: Authors. 12 Vietnam: Connecting value chains for trade competitiveness Box 2.1. Distinction between value chain, supply chain, and industrial cluster It is important to distinguish between the terms “value chain” and “supply chain”, which are often used interchangeably, as well as “industrial cluster”. A supply chain refers only to input-output links of productive activities in a product chain. A value chain embraces the full range of value-adding activities in a supply chain, and supplementary services including research and development, design, input sourcing, processing, marketing, distribution, and customer support. Figure 1 (Box 2.1) shows an example distinguishing an apparel value chain (in blue) from its supply chain (in green). The difference between the value chain and supply chain is that a supply chain focuses on a physical transformation and transportation of raw materials (or inputs) to final products, whilst a value chain focuses on activities that add economic value to products but are not necessarily manufacturing or logistics related. FIGURE 1 (BOX 2.1). Apparel value chain versus supply chain UPSTREAM DOWNSTREAM Research and Value-adding Design Production Logistics Marketing Services development activities Raw materials Textiles Final products Distribution and sales Market Market Natural and Yarn and Apparel Distribution Supply synthetic fabric Market production and sales chain bers production Market Second-tier supplier First-tier supplier Source: Modified from Staritz and Fredrick (2014) and Fredrick (2010). Industrial clusters, by contrast, refer to the geographic and sectoral components of economic activity. They are defined as “regional concentrations of economic activities in related industries connected through local linkages and spill-overs” (Ketels, 2017). Clusters are not the same as sectors because they recognize the geographic location of production and also consider groups of sectors that are related through links and spillovers. Clusters can further be differentiated from agglomerations which tend to focus on geographic locations of economic activity in the context of related sectors. This report uses the concept of industrial cluster to refer to the spatial structure of related industrial supply chains of domestic production. Chapter 2 – A new approach for value-chain-based connectivity 13 The information from the I/O analysis is refined in the fourth stage in combination with external information from existing value-chain maps or clusters. This allows us to create value-chain specific concordance tables that link the sectors of the I/O tables to the Vietnam Standard Industrial Classification (VSIC) codes. These concordance tables are value chain specific because an I/O sector can be linked to multiple VSIC codes hence the I/O tables are more aggregated than the VSIC codes (138 sectors vs. 734 sectors). Since VSIC is based on the International Standard Industrial Classification (ISIC), the selection of VSIC sectors has been supported by the accompanying United Nations document (2008) and insights from industry experts. VSIC is crucial to ensure an accurate reference for value-chain structure and a reliable source of analytical data. This study uses the five-digit VSIC 2018, developed by the General Statistics Office (GSO) based on the four-digit International Standard Industrial Classification (ISIC Rev.4, 2006). (See Annex 1.3 Terminology). Figure 2.4 shows the refined aquaculture value-chain links with the VSIC codes of related commodities in the value chain. FIGURE 2.4. Refined links: aquaculture value chain HARVESTING PROCESSING Frozen FISHING Prepared/preserved AQUA-PROCESSING Live sh, freshwater, marine, not farmed Salted/smoked/dried Processed and (VSIC03110, 03121, 03122) Canned preserved sh and crustaceans (VSIC1020) AQUA-CULTURING Live sh, freshwater, marine, farmed (VSIC03210, 03221, 03222) AQUA-FEEDING AQUA-FEEDING Key segments Prepared animal feeds Live sh, freshwater, marine, (VSIC10800) farmed (VSIC03230) Maize and other Grain mill products, starches, cereals and starch products Indirect segments (VSIC01120) (VSIC10612, 10620) Source: Authors. 14 Vietnam: Connecting value chains for trade competitiveness Box 2.2. Industrial classification and harmonized commodity coding system The Harmonized Commodity Description and Coding System, also known as the Harmonized System (HS) of tariff nomenclature is an internationally standardized system of names and numbers to classify traded products. It came into effect in 1988 and has since been developed and maintained by the World Customs Organization (WCO), an independent intergovernmental organization based in Brussels, Belgium, with over 200 member-countries. The HS is regularly updated by the World Customs Organization (WCO) to accommodate new products and remove outdated/obsolete products. The fourth edition, HS 2007 (which is a substantial revision from previous versions), came into effect on January 1, 2007, and on January 1, 2012 the fifth edition, HS 2012, came into effect. In HS code, the first two digits designate the HS chapter (HS2), the second two digits designate the HS heading (HS4), and the third two digits designate the HS subheading (HS6). The ten selected value chains in this report will be identified by the two- or four-digit Vietnam Standard Industrial Classification (VISC), and converted to a Harmonized System (HS2, HS4, and HS6) level when analyzing the related value chains’ trade data. Vietnam Standard Industrial Classification 2018 (VISC 2018) was developed by the General Statistics Office (GSO) based on the International Standard Industrial Classification of All Economic Activities (ISIC) adopted by the United Nations Statistics Committee in March 2006 (ISIC Rev.4, at four digits) and the ASEAN Common Industrial Classification (ACIC, at three digits). The ISIC and other relevant classification systems provide a set of activity categories for collecting and reporting statistics. They provide a comprehensive framework for collecting, reporting, and analyzing economic data for decision-making and policy making. The classification structure represents a standard format to organize detailed information about the state of an economy according to economic principles and perceptions. These economic activities are subdivided in a hierarchical, four-level structure of mutually exclusive categories, facilitating data collection, presentation and analysis at detailed levels of the economy in an internationally comparable, standardized way. To adapt ISIC in Vietnam, the GSO has developed the VSIC 2018 up to five digits, which has been promulgated by Prime Minister’s Decision No 27/2018/QD-TTg dated July 6, 2018. VSIC includes five levels: level one includes 21 sections hierarchically designated by alphabetical order from A to U, level two includes 88 divisions, each assigned a two-digit code according to the corresponding section, level three includes 242 groups, each assigned a three-digit code according to the corresponding division, level four includes 486 classes, each assigned a four-digit code according to the corresponding group, and level five includes 734 subclasses, each assigned a five-digit code according to the corresponding class. VSIC/ISIC classifies economic activities and is used mainly for statistical purposes like analyzing national accounts, demography of enterprises, employment, and others. The HS classifies products according to its composition, form, or function, and is used mainly for trade-related work like analyzing import/export and tariff statistics, or trade negotiations. Both VSIC/ISIC and HS are useful for research: the former is more commonly used for research focusing on local economic activities/products (like industrial agglomeration, clusters, local value chains), while the latter is more commonly used for research focusing on trade-related products/activities (such as global value chain of export-oriented products, competitiveness). Source: Authors, based on government regulations on VSIC. Chapter 2 – A new approach for value-chain-based connectivity 15 2.3. Defining the spatial structure of value chains The third step is to determine the spatial structure of value chains using the analytical results on I/O links of the domestic value chains from the previous step. In other words, the main purpose of this step is to determine geographic distribution and industrial concentration of selected chains, including their backward segments of the value chain. There was a way defining industry spaces as representation of technological relatedness (Hausman, Tran, Butos, 2017). However, this approach does not take into account backward and forward linkages across supply chains. From another perspective, cluster mapping was an initial step for developing a cluster initiative (Shakya, Mallika, the World Bank, 2009). While this approach introduced various tools for cluster mapping, including value chain analysis, it does not specify steps taken. This methodology defines spatial structure of value chains in two stages: (i) identifying regional specialization and value-chain locations by calculating location quotients (LQs) of all selected value chains and their segments, and (ii) mapping the spatial structure of value chains and their segments. LQs were calculated for each segment of the selected value chains and reflect the economic concentration of activities. They are a statistical indicator measuring the deviation of a series of specific economic activities in a region from the overall economy. Data from Vietnam’s Enterprise Censuses for 2011 and 2016 were used to calculate the key value chains’ LQs at the district and provincial level, based on three parameters: labor, employment and turnover, and revenue. Employment LQs are calculated as a value-chain segment’s regional employment share divided by the its national employment share. ei /e LQEmployment= Ei /E Where LQEmployment is LQ measured by employment, ei is the number of employees in the value-chain segment i at a locality, e is the total number of employees in a locality, Ei is the number of employees in the value-chain segment i of country, and E is the number of employees in a country. An LQ >1 indicates the value-chain segment has a greater regional employment share than the total value-chain segment has in national employment. An LQ 2016 > 1 reflects the current extent of relative provincial specialization. When LQ 2016 > LQ 2011, it shows positive change in relative provincial specialization expanded between 2011 and 2016 (and vice versa). Defining spatial structure, or value-chain mapping, can be generated from understanding geographic distribution of various segments in the value chain. The locational distribution of value-chain segments, in turn, is defined by scrutinizing regional specialization for respective segments. For example, the aqua-culturing segment is sub-segmented into the fishing segment and aqua-processing segment. Two other subsectors of the aqua-culturing segment (aqua-breeding and aqua-feeding) are also scrutinized. 16 Vietnam: Connecting value chains for trade competitiveness Figure 2.5 illustrates locational distribution of aqua-culturing segments. Aqua-culturing activities are spread-out all over the country. However, from statistics on international trade gateways, aqua- culturing activities in the North and Central Vietnam are mainly for domestic consumption, whereas in the South they are mainly for export. FIGURE 2.5. Locational distribution of the aqua-culturing segment Upper Right Quadrants: LQ 2016 > 1; Upper 45 - Degree Line: LQ 2016 > LQ 2011 30 Son La Ha Tinh Ba Ria - Vung Tau Khanh Hoa Bac Lieu Soc Trang 25 Ca Mau Ben Tre Quang Ninh 20 Ha Giang Quang Binh 15 LQ 2016 Lang Son Kien Giang Binh Thuan 10 5 Nghe An Vinh Long 0 Dong Thap Phu Yen -5 -5 0 5 10 15 20 25 LQ 2011 Source: Enterprise Census 2011 and 2016, and calculation by authors. Figure 2.6 illustrates provincial specialization in fishing activities. Fishing is concentrated in a few provinces in the South, including Khanh Hoa, Kien Giang, and Ben Tre. Although Khanh Hoa Province remains specialized in fishing (LQ2016 > LQ2011), other provinces such as Kien Giang, Ben Tre, and Tien Giang have reduced their specialization in fishing. FIGURE 2.6. Locational distribution of the fishing segment Upper quadrants: LQ2016>1; Upper Right Quadrant: LQ2016>LQ2011 130 Kien Giang, 108.7 100 LQ 2016 70 Ben Tre, 21.7 Tien Giang, 7.2 Ninh Thuan, 1.0 Ca Mau, 1.1 Ha Tinh, 1.7 Khanh Hoa, 10.2 40 10 -13 -9 -5 -1 3 -20 Changing in LQ between 2016 - 2011 Circle size = number of employee 2016, Min Ninh Thuan = 65, Max Kien Giang = 23,700 Source: Enterprise Census 2011 and 2016, and calculation by authors. Chapter 2 – A new approach for value-chain-based connectivity 17 The Aqua-processing segment is located mostly in the Mekong Delta Region as shown in Figure 2.7. Key localities have enhanced their specialization in aqua-processing including Ca Mau, Soc Trang, Bac Lieu, and Dong Thap provinces. Provinces that have reduced their aqua-processing specialization are Hau Giang, An Giang, Can Tho, Tien Giang, and Kien Giang. FIGURE 2.7. Locational distribution of the aqua-processing segment Upper quadrants: LQ 2016 >1; Upper Right Quadrant: LQ 2016 > LQ 2011 35 Ca Mau, 27.7 Bac Lieu, 25.5 An Giang, 14.0 30 Hau Giang, 15.7 Soc Trang, 23.1 25 Binh Thuan, 4.6 20 Can Tho, 11.3 Dong Thap, 24.3 LQ2016 Tien Giang, 5.7 15 Khanh Hoa, 3.9 Ninh Thuan, 4.4 Kien Giang, 6.1 10 Ben Tre, 5.2 Phu Yen, 4.0 5 Tra Vinh, 1.4 Thua Thien Hue, 1.5 0 -6 -4 -2 0 2 4 6 Quang Ngai, 1.4 Ba Ria - Vung Tau, 3.1 -5 Quang Binh, 1.1 Da Nang, 1.2 Long An, 1.2 Changing in LQ between 2016 - 2011 Circle size = number of employee 2016, Min Quang Binh = 700, Max Dong Thap = 23,000 Source: Enterprise Census 2011 and 2016, calculation by authors. Table 2.2 shows calculated LQs identifying geographic locations of the aquaculture value chain and its various segments. 18 Vietnam: Connecting value chains for trade competitiveness TABLE 2.2. Spatial Structure of the Aquaculture Value Chain Province LQ Diff btw Province LQ Diff btw Province LQ Diff btw 2016 LQ 2016 2016 LQ 2016 2016 LQ 2016 >1 and LQ >1 and LQ >1 and LQ 2011 2011 2011 Aqua-feeding Aqua-culturing Fishing Hoa Binh 1.08 0.52 Tuyen Quang 1.30 1.30 Ha Tinh 1.73 1.73 Vinh Phuc 1.58 -0.56 Lai Chau 1.38 1.38 Khanh Hoa 10.19 1.96 Hai Duong 1.89 0.43 Son La 25.46 25.46 Ninh Thuan 1.04 -2.31 Hung Yen 4.28 -1.22 Yen Bai 1.03 -1.02 Tien Giang 7.18 -6.95 Ha Nam 2.93 -0.16 Lang Son 5.61 5.61 Ben Tre 21.74 -11.18 Quang Nam 1.38 0.42 Quang Nam 7.46 3.46 Kien Giang 108.66 -1.19 Binh Dinh 1.55 1.01 Bac Giang 2.38 2.28 Ca Mau 1.08 0.81 Binh Thuan 1.60 -0.49 Nam Dinh 1.66 0.55 Aqua-processing Tay Ninh 1.05 0.77 Nghe An 1.92 -1.18 Quang Binh 1.13 0.64 Binh Duong 1.08 0.18 Ha Tinh 19.60 17.62 T.Thien - Hue 1.46 0.75 Dong Nai 5.84 -0.29 Quang Binh 7.57 1.13 Da Nang 1.18 0.00 Long An 4.47 -0.46 Quang Tri 1.49 1.07 Quang Ngai 1.37 0.80 Tien Giang 2.07 -0.38 Phu Yen 2.78 -7.63 Phu Yen 4.01 2.74 Vinh Long 1.57 -0.92 Khanh Hoa 9.52 0.52 Khanh Hoa 3.92 0.11 Dong Thap 9.76 -4.55 Binh Thuan 2.56 -9.20 Ninh Thuan 4.41 2.08 An Giang 1.39 1.02 Dak Lak 2.39 1.33 Binh Thuan 4.59 -1.02 Kien Giang 1.20 -0.11 Lam Dong 1.65 -0.44 B.Ria - Vung Tau 3.15 -0.14 Can Tho 2.05 -1.76 B.Ria-Vung Tau 3.69 0.90 Long An 1.23 0.17 Hau Giang 3.15 1.15 Ben Tre 9.02 7.06 Tien Giang 5.70 -1.52 Aqua-breeding Tra Vinh 2.14 1.60 Ben Tre 5.16 1.66 Nghe An 1.87 -0.83 Vinh Long 1.12 -14.02 Tra Vinh 1.44 -4.08 Ha Tinh 3.70 3.70 An Giang 2.45 2.06 Dong Thap 24.29 5.37 Binh Dinh 4.09 -1.95 Kien Giang 8.13 -8.34 An Giang 13.99 -1.67 Phu Yen 2.82 2.44 Can Tho 1.74 -0.36 Kien Giang 6.10 -1.71 Khanh Hoa 1.62 -0.55 Soc Trang 19.81 0.22 Can Tho 11.29 0.15 Ninh Thuan 152.10 35.20 Bac Lieu 15.59 4.95 Hau Giang 15.67 -2.05 Binh Thuan 39.68 2.98 Ca Mau 5.06 3.08 Soc Trang 23.13 1.50 Ben Tre 6.94 5.62 Bac Lieu 25.47 4.35 Can Tho 2.11 2.11 Ca Mau 27.69 1.42 Bac Lieu 96.52 -14.52 Ca Mau 20.30 7.07 Source: Enterprise Census (2011 and 2016), calculation by authors. The spatial structure of the aquaculture value chain is illustrated in Map 2.1 based on analyses of the locational distribution of its five respective value-chain segments. 1. Aqua-feeding: a small-scale segment concentrated in the South, in Dong Nai, Long An, and Dong Thap provinces. Chapter 2 – A new approach for value-chain-based connectivity 19 2. Aqua-breeding: a small-scale segment concentrated in the South, in Ninh Thuan, Binh Thuan, and Ca Mau provinces. 3. Aqua-culturing: spread-out across the country though aqua-culturing activities in the North and Central Vietnam are mainly for domestic consumption, and in the South for export. 4. Fishing: concentrated in few localities in the South (Khanh Hoa, Kien Giang, and Ben Tre provinces). 5. Aqua-processing: concentrated mostly in the Mekong Delta Region. Processed products are mostly exported through Ho Chi Minh City (HCMC) seaports. Although the aqua-culturing segment is spread across the country, the export-oriented value chain that has a spatial layout along key segments including aquaculture, capture, and export processing, is concentrated mainly in the South, especially in the Mekong Delta. MAP 2.1. Spatial structure of the aquaculture value chain LQ Aqua-processing 1.2 - 2.0 2.1 - 5.0 5.1 - 10.0 10.1 - 27.7 Disclaimer: The boundaries, colors, LQ Aqua-fishing denominations and other 7.2 - 10.0 information shown on 10.1 - 30.0 any map in this work do not imply any judgement 30.1 - 108.7 on the part of The World Bank concerning the legal LQ Aquaculturing status of any territory 1.0 - 2.0 or the endorsement 2.1 - 5.0 or acceptance of such 5.1 - 10.0 boundaries. 10.1 - 25.5 LQ Aqua-breeding 4.1 - 5.0 5.1 - 10.0 10.1 - 30.0 30.1 - 152.1 LQ Aqua-feeding 1.1 - 2.0 2.1 - 5.0 5.1 - 5.8 Top 10 Aqua-processing firms (income) 1,000,000 10,000,000 Source: I/O Table (2011 and 2016) and Enterprise Census (2011 and 2016), calculation by authors. 20 Vietnam: Connecting value chains for trade competitiveness 2.4. Value-chain-based connective propensity and key corridors The objective of the fourth step is to use the results of the geographic analysis from the third step to analyze the structure of domestic production chains in the context of regional specialization. In other words, this analysis helps identify the connection of different production-chain segments by connectivity of regional specialization links. The provincial specialization degree was determined not only by the LQs of all selected chains in that province, but also by the change in the LQ measured in two points in time. A specialization map for each production chain for all provinces helps define the connective propensity of production chains based on regional specialization links. The methodology of the connective propensity analysis is based on the LQ index and the value-chain spatial structure. To define the connective propensity of a value chain we relied on the following four hypotheses: 1. Localities having a productive concentration for value chain segments with LQ > 1 defined as linked nodes to attract and generate connections emerged throughout the domestic part of value chain. 2. Localities having productive concentration with LQ2016 > LQ2011, are becoming more specialized in the segments and/or value chain, and therefore may potentially generate stronger connectivity in the future (and vice versa). 3. The connective propensity of a value chain is the trend of related commodities flowing from and to international gateways and along various productive points of a domestic value chain as shown in the connective model in Figure 2.8. A connective propensity is defined based on spatial locations of productive segments of the domestic value chain and the structure of value- chain links. It begins with materials imported via a gateway, connected by various intermediate points for supplying and processing raw materials, concluding with finished products exported via gateways or consumed domestically. 4. Transport distance is assumed to define priority of movement of goods and transport corridor formation. Import and export gateways are at the end FIGURE 2.8. Connective model for a value nodes in this model. For all identified value chain chains, trade gateways (air, sea, inland waterways, rail, and road) are imports origins Imported Materials Exports and exports destinations, whereas, location of Assemblers/ Processors cluster segments (suppliers, processors) are the Domestic Domestic internal export origin and import destination. Materials Consumption The information on trade gateways8 for all Source: Authors. forms of transport is important for value-chain- based connectivity. 8 A comprehensive gateways analysis is done in chapter three. Chapter 2 – A new approach for value-chain-based connectivity 21 Table 2.3 and Map 2.2 depict the connective propensity and key corridors of the aquaculture value chain. TABLE 2.3. Key corridors of the aquaculture value chain Origin Destination Main corridor Aqua-feeding Aqua-culturing Quan Lo – Phung Hiep expressway, HCMC - Trung Luong – My Thuan expressway, Ben Luc – Long Thanh expressway, NR1, NR91, NR80, NR54, NR62, NR63, Ringroad No4 HCMC, NR22, NR13 (Vinh Binh – Binh Duong – Binh Phuoc), Ringroad No3 HCMC, NR51, NR56, NR20, NR55, Cau Gie – Ninh Binh expressway, Noi Bai – Bac Ninh expressway, Noi Bai – Lao Cai expressway (through Vinh Phuc), NR21, NR32B, AH13, NR18, NR5 Breeding Aqua-culturing HCMC - Trung Luong – My Thuan expressway, NR91, NR80, NR54, NR62, NR63, NR60, NR57, Ringroad No4 HCMC, NR22, NR13 (Vinh Binh – Binh Duong – Binh Phuoc), NR14, AH17, Ringroad No3 HCMC, NR51, NR56, NR20, NR55, NR28B, NR7, NR1, NR19, Noi Bai – Lao Cai expressway Aqua-culturing Processing Quan Lo – Phung Hiep expressway, HCMC - Trung Luong – My Thuan expressway, Ben Luc – Long Thanh expressway, NR1 (Phu Yen - HCMC – Ca Mau), NR91, NR80, NR54, NR62, NR63, NR60, NR57, Ringroad No4 HCMC, NR22, NR13 (Vinh Binh – Binh Duong – Binh Phuoc), Ringroad No3 HCMC, NR51, NR56, NR20, NR55, NR28B, NR7, NR1, AH13, Noi Bai – Lao Cai expressway, NR26, NR29, NR10, NR18, Ha Noi – Bac Giang expressway, NR12B Fishing Processing Quan Lo – Phung Hiep expressway, HCMC - Trung Luong – My Thuan expressway, NR80, NR63, NR62, NR54, NR1 (HCMC – Ca Mau), NR51, Ringroad No4 HCMC, NR20, NR56, NR1 (Ninh Thuan – Dong Nai), NR27, NR28 Processing Export Quan Lo – Phung Hiep expressway, HCMC - Trung Luong – My Thuan expressway, Ben Luc – Long Thanh expressway, NR80, NR91, NR91C, NR1, NR22, NR51, NR19C, NR24B, NR9 Source: Enterprise Census (2016), and estimates by authors. 22 Vietnam: Connecting value chains for trade competitiveness MAP 2.2. Connective propensity of the aquaculture value chain LQ Aqua-processing 1.2 - 2.0 2.1 - 5.0 5.1 - 10.0 10.1 - 27.7 LQ Aqua-fishing 7.2 - 10.0 10.1 - 30.0 30.1 - 108.7 LQ Aquaculturing 1.0 - 2.0 2.1 - 5.0 5.1 - 10.0 10.1 - 25.5 LQ Aqua-breeding 4.1 - 5.0 5.1 - 10.0 10.1 - 30.0 30.1 - 152.1 LQ Aqua-feeding 1.1 - 2.0 2.1 - 5.0 5.1 - 5.8 Top 10 Aqua-processing firms (income) Disclaimer: 1,000,000 The boundaries, colors, denominations and other 10,000,000 information shown on any map in this work do not imply any judgement Airport on the part of The World Bank concerning the legal Border gate status of any territory or the endorsement Seaport or acceptance of such boundaries. Processing(1V) - Export Fishing(1IV) - Processing(1V) Aquaculturing(1III) - Processing(1V) Breeding(1II) - Aquaculturing(1III) AquaFeeding(1I) - Aquaculturing(1III) Source: I/O Table (2012 and 2016) and Enterprise Census (2011 and 2016), calculation by authors. Chapter 2 – A new approach for value-chain-based connectivity 23 This connective propensity analysis concludes with a model of the value-chain link as a result of the previous study, to simulate freight flows according to the origin-destination (OD) model along the spatial organization of value chains and connect chains to international gateways. The results of the mapping allowed us to identify trade corridors organized to promote value-chain links, and to evaluate supporting logistics services to improve their overall efficiency and competitiveness. It is necessary to develop a model of disaggregated freight flows for traffic between the main freight centers in Vietnam (at district level if possible, including cluster locations) and the trade gateways. Freight-flow models, however, will require freight-volume data for transportation I/O within a value chain, and between processing location and trade gateway. Given the lack of readily accessible relevant freight- volume data needed, available HS trade data should be converted into equivalent freight-volume data, using proven techniques for converting export and import commodity values with appropriate assumptions. The key challenge is the lack of provincial I/O tables to define links, and OD flows per commodity within the value chains of each selected clusters. 2.5. Critical consolidated transport network for ten selected value chains Map 2.3 shows the consolidation of connective propensity, transport network, and critical corridors for the ten selected value chains. The transport link’s thickness represents its significance to value- chain links and freight flows of final exported and imported products as well as all backward links and intermediate products of the domestic supply chains. Key corridors are found (i) around the largest economic centers—Hanoi and HCMC—connecting nearby provinces that participate in the value chains, (ii) between the Mekong Delta Region and HCMC, (iii) between Hanoi and northern Chinese borders, (iv) along the north-south coastal line, and (v) between the central highlands and the south. Ensuring quality infrastructure and necessary logistics services along these corridors would help lower the trade and transport costs associated with these value chains, which are crucial for Vietnam’s export competitiveness. On one hand, transport and logistics networks should be master-planned and new transport investments should be based at least partially on changes in the structure of trade flows and value chains. On the other hand, they have the reverse relationship because infrastructure development of existing transport networks, needs on trade and cluster development could be created. In any case, value-chain connectivity information and analysis are crucial for optimizing the efficiency of transport networks and transport infrastructure investment. The relationship between transportation infrastructure and trade-related activities are elaborated further in the Vietnam Development Report 2019 (VDR) “Connecting Vietnam for Growth and Shared Prosperity”. 24 Vietnam: Connecting value chains for trade competitiveness MAP 2.3. Connective propensity of ten export-oriented value chains Disclaimer: The boundaries, colors, denominations and other information shown on any map in this work do not imply any judgement on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Multi-source: I/O Table (2012 and 2016) and Enterprise Census (2011 and 2016), calculation by authors. Chapter 2 – A new approach for value-chain-based connectivity 25 CHAPTER 3 Efficient international trade gateways This chapter recommends transport infrastructure investment, especially for developing international trade gateways, take into account not only increasing cargo transportation needs, but also changes in the structure of imports and exports. 26 Vietnam: Connecting value chains for trade competitiveness Because connectivity matters for export competitiveness, changes in the product-based export structure should be considered when prioritizing better connectivity and logistics for more competitive exports. A transport system that supports exports should account for this shift from a logistics perspective, especially in terms of investment into appropriate types of trade gateways. Therefore, the efficiency improvement of the international trade gateway structure and performance should take into account not only trade growth but also (and more important) structural changes and the continued development of domestic value chains. 3.1. Overview of trade gateways A trade gateway (also known as a border gate) is the first destination of imported goods and the final departure point of exported goods. Therefore, gateways play an important role in foreign trade activities both in terms of cost and reliability of traded goods, and the performance and development of domestic value chains. There is a distinction between a customs clearance point and a trade gateway. Customs statistics show there are 478 customs clearance points including at borders and at Inland Clearance Depots (ICDs) or other inland clearance points. However, the number of international gateways is much less. Among Vietnam’s 48 major international gateways located in 31 provinces and cities, there are four main types: land (20), air (8), sea (16) and other (two rail and two inland waterways (IWT)). Although airports and seaports have broader international connections, the land gateways are to China, Laos and Cambodia. Among other major international gateways, there is one rail gateway to China and two IWT gateways to Cambodia. Table 3.1 represents Vietnam’s main international gateways. TABLE 3.1. Vietnam’s main gateways Transport mode Location Name Connection Sea Air Land Other Ha Noi Noi Bai Airport X Multi-national Hai Phong Cat Bi Airport X Multi-national Hai Phong Seaport Complex X Multi-national Quang Ninh Van Don Airport X Multi-national Mong Cai X China Cai Lan Seaport X Multi-national Lang Son Huu Nghi X China Dong Dang Rail China Cao Bang Ta Lung X China Ha Giang Thanh Thuy X China Lao Cai Lao Cai X Rail China Dien Bien Tay Trang X Lao PDR Son La Chieng Khuong X Lao PDR Ninh Binh Ninh Phuc Seaport X Multi-national Thanh Hoa Na Meo X Lao PDR Nghi Son Seaport X Multi-national Chapter 3 – Efficient international trade gateways 27 TABLE 3.1. Vietnam’s main gateways (cont.) Transport mode Location Name Connection Sea Air Land Other Nghe An Vinh Airport X Multi-national Nam Can X Lao PDR Cua Lo Seaport X Multi-national Ha Tinh Cau Treo X Lao PDR Vung Ang Seaport X Multi-national Quang Binh Cha Lo X Lao PDR Quang Tri Lao Bao X Lao PDR Hue Phu Bai Airport X Multi-national Chan May Seaport X Multi-national Da Nang Da Nang Airport X Multi-national Tien Sa Seaport X Multi-national Quang Nam Ky Ha Seaport X Multi-national Quang Ngai Dung Quat Seaport X Multi-national Kon Tum Bo Y X Lao PDR Gia Lai Le Thanh X Cambodia Binh Dinh Quy Nhon Seaport X Multi-national Khanh Hoa Nha Trang Seaport complex X Multi-national Cam Ranh Airport X Multi-national Binh Phuoc Hoa Lu X Cambodia Tay Ninh Xa Mat X Cambodia Moc Bai X Cambodia Ba Ria - Vung Tau Phu My Seaport X Multi-national Vung Tau Seaport Complex X Multi-national Ho Chi Minh City Saigon Seaport Complex X Multi-national Tan Son Nhat Airport X Multi-national Long An Binh Hiep X Cambodia Dong Thap Dinh Ba X Cambodia Thuong Phuoc IWT Cambodia An Giang Vinh Xuong IWT Cambodia Tinh Bien X Cambodia Can Tho Can Tho Seaport X Multi-national Kien Giang Ha Tien X Cambodia Source: Authors, combined sources of Vietnam Customs and Wikipedia. Despite that there are 48 main international gateways, as in Map 3.1, just 12 international gateway complexes accounted for more than 85.6 percent of Vietnam’s trade value in 2016. These 12 gateways are the two largest airports, Noi Bai in Hanoi and Tan Son Nhat in Ho Chi Minh City (HCMC), the five most important seaports (complexes in HCMC, Hai Phong, and Vung Tau, Cai Lan in Quang Ninh, and Tien Sa in Da Nang), and five land-gate complexes in Lang Son, Quang Ninh, Lao Cai, Quang Binh-Quang Tri, and Tay Ninh. Among the land gateways, the complexes of Lang Son, Quang Ninh, and Lao Cai are the main gateways to China, the Quang Binh-Quang Tri complex is the key gateway to Lao PDR, and the Tay Ninh complex is the main gateway to Cambodia. 28 Vietnam: Connecting value chains for trade competitiveness MAP 3.1. Main trade gateways Lang Son Land-Port: 2.4% of total trade (2016) 8 1.2 2.3 6 1.1 Billions US$ 4 0.9 6.8 5.5 6.0 2 0.7 3.9 0.4 1.6 1.8 0 2011 2012 2013 2014 2015 2016 Import Export Hai Phong Seaport Complex: 17% of total trade (2016) 60 23.7 25.3 26.8 40 21.0 Billions US$ 14.7 17.9 20 30.7 32.0 33.6 32.8 24.4 25.1 0 2011 2012 2013 2014 2015 2016 Import Export Disclaimer: The boundaries, colors, denominations and other Noi Bai Airport: 21.9% of total trade (2016) information shown on any map in this work do 80 not imply any judgement on the part of The World 44.3 Billions US$ Bank concerning the legal 39.5 status of any territory 26.8 30.8 or the endorsement 30 15.2 or acceptance of such 28.8 32.4 8.3 22.5 23.3 boundaries. 8.3 15.5 0 2011 2012 2013 2014 2015 2016 Import Export Tan Son Nhat Airport: 8.0% of total trade (2016) 30 12.5 20 Billions US$ 10.5 7.8 10 6.6 7.3 7.1 15.40 12.62 7.99 7.71 7.26 9.41 0 2011 2012 2013 2014 2015 2016 Import Export Air port HCMC Seaport Complex: 29.6% of total trade (2016) Vung Tau Seaport Complex: 1.7% of total trade (2016) 15 Railway gate 100 54.6 48.0 8.7 10.0 8.7 Billions US$ Billions US$ 54.6 47.6 10 46.0 51.2 Waterway gate 50 4.1 5 3.0 Seaport 46.6 45.9 49.9 52.9 52.2 55.9 6.3 6.2 6.3 1.8 4.2 3.1 2.5 0 0 Road gate 2011 2012 2013 2014 2015 2016 2011 2012 2013 2014 2015 2016 Import Export Import Export Source: Customs data, calculation by authors. Chapter 3 – Efficient international trade gateways 29 When viewing the geographic distribution of international gateways in proximity to economic regions, Noi Bai Airport in Hanoi, Hai Phong Seaport Complex, Cai Lan Seaport, and land border-gate complexes in Lang Son, Quang Ninh, Lao Cai are international gateways in the Northern and Red River Delta Regions. Tien Sa Seaport in Da Nang and Quang Binh-Quang Tri land gateways are key international gateways in the Central Region. Tan Son Nhat Airport in HCMC, Saigon Seaport Complex, Vung Tau Seaport Complex, and Tay Ninh land gateway are the most important international gateways in the Southeast and Mekong Delta Regions. Among the above-mentioned 12 gateways that represent all regions and transport modes, the actual flows of goods across borders are concentrated in only six gateways, which are also highlighted in the Map 3.1 with graphs showing shares of the value of export and import going through these gateways in 2016. These six most significant gateways accounted for 80.5 percent of the total trade value in 2016 and include HCMC Seaport Complex, Hai Phong Seaport Complex, Vung Tau Seaport Complex, Lang Son land gateway, Noi Bai Airport, and Tan Son Nhat Airport. 3.2. Trade by gateway type Vietnam’s trade value has increased rapidly over time, with trade through all gateways increasing, though not uniformly across all types. Figure 3.1 shows this increasing trend between 2011 and 2016 and reflects the change in the domestic value-chain structure where trade value has increases unevenly across different gateway types. The rapid increase in trade flows through air compared to other gateways shows stronger growth in production, export, and import of small but valuable products like mobile phones, electronic components, high fashion exports, and high-value, processed agricultural products. Box 3.1 presents the case of Samsung Vietnam, which has significantly changed the electronics value-chain structure and the development of Vietnam’s air trade. FIGURE 3.1. Trade value by gateway type (2011-2016) 200 185 182 179 180 171 171 161 160 140 120 105 Billions USD 100 92 80 72 64 60 46 40 31 13 14 15 15 20 9 10 0 2011 2012 2013 2014 2015 2016 Sea Road Air Linear (Sea) Linear (Road) Linear (Air) Source: Customs data, calculation by authors. 30 Vietnam: Connecting value chains for trade competitiveness Box 3.1. Electronics Value Chain and Samsung Vietnam Electronic products are often categorized into three groups: computers, communications, and consumer electronics (3C). In electronic value chains, 3C are the final products that have been assembled from a number of components and parts, such as semiconductors, integrated circuits, printed circuit boards (PCB), etc. Electronics, especially cellphone parts, are usually small and light, and more likely to be high value, thus easy to transport by air. These characteristics enable electronics GVCs to spread across countries and continents to optimize comparative advantages and country specialization, and build supply chains around air transportation and logistics. I. COMPONENT II. SUBASSEMBLIES III. FINAL PRODUCTS Integrated Consumer electronics circuits Semiconductor wafer Active discrete Product speci c PCBA Communications parts Electronic PCB Passives Display Enclosure/housing Computers/Storage/O ce Cables Electrical Batteries Samsung started their business in Vietnam in 1996 with a small joint venture named Samsung Vina (with the total investment of 36.5 US$ million) that ultimately turned 100 percent FDI after Samsung bought all the shares from its partner in 2013. In 2008, Samsung made a big move with a new project producing smartphones in Bac Ninh (with the total investment of 670 US$ million) and since then has expanded their production in Vietnam, opening several new factories. Samsung now has six factories in three provinces (Bac Ninh, Thai Nguyen, and Ho Chi Minh City), and an R&D center in Ha Noi with a total registered investment capital of 17 US$ billion. A decade after starting the first project on smartphones, Samsung has cultivated its supply chain in Vietnam with 35 suppliers in 2018, up from four in 2014, and is expected to expand to 50 by 2020. Although the local supply chain has expanded, Samsung still relies on imported parts mostly delivered from China through Huu Nghi land gateway, or from other countries via Dinh Vu seaport, but most importantly from the Noi Bai airport. Samsung Vietnam makes the majority of its global smartphones output in Vietnam, including its latest flagship device, which is exported all over the world via Noi Bai airport. This is one of main reasons Vietnam’s smartphone export has increased significantly, from nothing to 20 percent of total exports over the last decade, and trade value via air gateway has increased rapidly, from 15.6 percent in 2011 to 39.5 percent in 2016. Samsung Vietnam has developed a unique supply chain and logistics system to ensure the shortest delivery time for its just-in-time production in Vietnam. In Noi Bai Airport, there is a separate warehouse area and customs clearance line for Samsung to ensure their imports and exports go smoothly. Samsung’s investment has caused structural changes in logistics in the North, with the rapid growth of air cargo and transportation that requires a change in corresponding connective policy and logistics investment. Source: Authors. Chapter 3 – Efficient international trade gateways 31 As a result, although trade through sea and land gateways maintained a certain growth rate, they did not reach the rate of air gateways in the same period. Figure 3.2 shows that the share of trade via air has increased rapidly from 15.6 percent in 2011 to 39.5 percent in 2016, the share of trade via land has decreased slightly from 5.7 percent in 2011 to 4.3 percent in 2016, and the share of trade via sea fell dramatically from 78.8 percent in 2011 to 56.1 percent in 2016. FIGURE 3.2. Share of trade value by gateway As we will discuss later, the change in trade type (2011-2016) gateways used reflects a change in the domestic value-chain structure. Analyzing 100% the product mix of exports and imports 90% 80% through different gateways and their change 70% over time provides a better understanding 60% of this structural change. Figures 3.3-A and 50% 40% 3.3-B demonstrate the structure of imported 30% products through sea gateways for 2011 and 20% 2016, respectively. Generally, sea gateways 10% 0% carry a variety of durable and high-volume 2011 2013 2014 2015 2016 imports, like machinery and mechanical Seagateway Roadgateway Airgateway equipment (HS84), electronics and electrical equipment (HS85), mineral fuels and oils Source: Customs data, calculation by authors. (HS27), iron and steel (HS72), plastic products (HS39), vehicles (HS87), cotton (HS52), etc. Moreover, this product mix has not changed much overtime. Comparing the top four products imported via sea gateways between 2011 and 2016, the share of mineral fuels and oils (HS27) has declined sharply from 11.8 percent to 2.6 percent. While imports of machinery and mechanical equipment (HS84) have increased from 12.4 percent to 14.6 percent, imports of electronics and electrical equipment (HS585) have decreased from 13.5 percent to 10.4 percent, and imports of plastic products (HS39) have risen from 6.3 percent to 8.8 percent. 32 Vietnam: Connecting value chains for trade competitiveness FIGURE 3.3. Products imported through sea gateways (2011 and 2016) Figure 3.3.A: Products imported through sea gateways (2011) HS85 HS27 HS39 HS55 HS23 HS71 HS73 Arti cial Residues Natural/ Products of staple ber, from food cultured iron or 2.2% industry, 2.2% pearls, 2.1% stell, 2.1% HS29 Other Ruber Paper Organic chemical and Medicin... and chemicals products rubber... 1.5% paperb...  Electronics and electrical equipment (HS8517 Telophone 2.0% Man- sets; HS8544 wire, cable; HS85 Plastic Cereals, Hides Fats Inor... HS60 made Electrical apparatus), 13.5% products, 1.1% (exce... and... che... Types of lame... Mineral fuels and oils, 6.3% HS84 hosiery 11.8% or crochet Copper Dye Flyi... Dai... Oil... Fis... HS52 and or... HS72 copper... W... E... P... G... Im... ... coa... S... ... ... ... P... Cotton, 2.6% Fertilizer, Wood Machines and mechanical 1.7% S... ... Z... ... ... ... and... Nuts S... equipment (HS8471 Automatic HS87 Optical, an... S... ... ... data processing machines, S... photogra- ... HS8466 Parts and accessories), Vehicles, Alumi... ... 12.4% Iron and steel, 7.2% 2.3% phic, 1.6% and... Sp... ... ... ... ... ... Figure 3.3.B: Products imported through sea gateways (2016) HS84 HS39 HS87 HS27 HS76 HS23 HS73 Aluminum Residues Products of Mineral fuels and aluminum from food inron of and oils, 2.6% products, 2.5% industry, 2.5% steel, 2.0% HS10 Vehicles, 3.9% Organic Copper Optical, Nuts and chemicals, and photogr... nuts are HS60 Cereals, 1.8/% copper... 1.6% edible,... Machines and mechanical 2.0% equipment (HS8471 Automatic Types of Rubber Dye or Fish and Impr... Spe HS54 data processing machines, hosiery or and tanning... other... coate... wov... Man-made HS8466 Parts and accessories), Plastic products, crochet, 3.4% rubber... lament 14.6% 8.8% yarn; stri... Wood Innor... Gro... Oth... Ess... Pre... HS52 che... HS85 HS72 HS38 and Other woode... Gla... Fe... Sh... D... A... chemical Oil products, Hides see... Zi... ... ... ... ... Cotton, 3.1% 1.9% (except...) Fat... HS55 Ot... ... P... ... ... ... ... Electronics and electrical HS48 ite... equipment (HS8517 Telephone Paper and So... ... ... ... F... P... ... sets; HS8544 wire, cable; HS85 Electrical appartus), 10.4% Iron and steel, 6.2% Arti cial staple paperbo... ber, 2.7% products... Medici... 1.2% Wa... Fu... ... L... ... ... ... Source: Customs data, calculation by authors. Similar to imports, sea gateways carry a variety of durable and high-volume exports. The product mix looks like that of imports including machinery and mechanical equipment (HS84), electronics and electrical equipment (HS85), mineral fuels and oils (HS27), iron and steel (HS72), plastic products (HS39), vehicles (HS87), cotton (HS52), etc. This structure has remained unchanged over time. Figure 3.4-A and 3.4-B show the top 15 products exported via sea gateways for 2011 (accounting for 77.8 percent Chapter 3 – Efficient international trade gateways 33 of total sea exports) and 2016 (accounting for 73.2 percent of total sea exports). Clearly mineral fuel and oil exports (HS27) dropped out of the top 15 exported products, from the second biggest item in 2011 (11.8 percent) to 2.6 percent in 2016. FIGURE 3.4. Top 15 products exported through sea gateways (2011 and 2016) Figure 3.4.A: Top 15 products exported Figure 3.4.B: Top 15 products exported through sea gateways (2011) through sea gateways (2016) HS85 Electronics... 13.4% HS85 Electronics... 12.7% HS27 Mineral fuels & oil 11% HS62 Clothing... 9.5% HS62 Clothing... 7.2% HS61 Clothing... 8.5% HS64 Shoes, sandals... 7.0% HS64 Shoes, sandals, gaiters... 8.4% HS61 Clothing... 6.2% HS84 Machines... 5.9% HS03 Fish and other... 5.1% HS94 Furniture... 5.3% HS84 Machines... 4.4% HS03 Fish... 5.2% HS40 Rubber 4.1% HS09 Co ee & tea 3.8% HS09 Co ee & tea 3.9% HS40 Rubber 3.1% HS10 Cereals 3.8% HS39 Plastic... 2.6% HS94 Furniture... 3.4% HS08 Nuts and Citrus fruits 2.5% HS71 Natural or cultured... 2.8% HS42 Leather products... 2.1% HS72 Iron and steel 1.9% HS16 Meat, sh 2.1% HS39 Plastic... 1.8% HS73 Products of iron or steel 1.8% HS08 Nuts... 1.8% HS87 Vehicles... 1.8% 0 5 10 15 0 5 10 15 US$ Billions US$ Billions Source: Customs data, calculation by authors. Compared to sea gateways, air gateways export fewer products with higher values, like electronics and electrical equipment (HS8517 – telephones, HS 8542 – electronic integrated circuits), natural and cultured pearls (HS71), optical, photographic (HS90), and machine equipment (HS84), etc. Figure 3.5 compares the structure of products exported through air gateways between 2011 and 2016. The share of exported electronics and electrical equipment increased considerably from 52.7 percent in 2011 to 76.5 percent in 2016. In contrast, the share of exported natural and cultured pearls dropped from 17.7 percent in 2011 to 1.6 percent in 2016. 34 Vietnam: Connecting value chains for trade competitiveness FIGURE 3.5. Products exported through air gateways 2011 2016 HS85 HS71 HS85 HS84 Machines equipment, Natural or cultured pearls, 7.1% 17.7% Optical, Optical, Machines photographic, photographic, equipment, 5.2% 5.5% 5.5% Cloth... Clo... HS62 HS61 Sho... G... and... and... Cloth... Cloth... and sand... a... Electronics and electrical and cloth... Pla... ... Nat... Sh... equipment (HS8517 Telephone; acce... ... Electronics and electrical equipment HS8542 Electronic integrated Mine... (HS8517 Telephone; ... ... circuits), 52.7% fuels,... ... ... HS8542 Electronic integrated circuits), 76.5% ... ... ... Source: Customs data, calculation by authors. Figure 3.6 shows the structure of top ten products imported through air gateways between 2011 and 2016. Similar to exports, the share of electronics and electrical equipment imports increased from 52.3 percent in 2011 to a dominant share of 75.9 percent in 2016. The share of all other imported products therefore reduced considerably. FIGURE 3.6. Products imported through air gateways 2011 2016 HS85 Electronics and... 52.3% HS85 Electronics and... 75.9% HS71 Natural or cultured... 15.5% HS90 Optical, photographic 6.8% HS84 Machines equipment 9.9% HS84 Machines equipment 6.2% HS90 Optical, photographic 6.4% HS30 Medicine 3.4% HS30 Medicine 5.8% HS39 Plastic 3.1% HS88 Flying vehicles 4.0% HS73 Products of iron or steel 1.4% HS39 Plastic 2.2% HS71  Natural or cultured... 1.3% HS73 Products of iron or steel 1.8% HS41 Hides and leather 0.9% HS87 Vehicles 1.1% HS82  Tools, utensils, knives,... 0.6% HS41 Hides and leather 1.1% HS38 Other chemical products 0.6% 0 4 8 0 20 40 US$ Billions US$ Billions Source: Customs data, calculation by authors. Chapter 3 – Efficient international trade gateways 35 From a trade and transport perspective these are different but equally important: while high-value electronics are increasingly important in people’s lives, they generate much smaller volumes in weight per cubic meter (m3) than other value chains. As such, they need less infrastructure per dollar of trade and a different type. The same applies in ports: a container of electronics may have 100 times the value of a container of plastic products, but both require a container of transport capacity. Hence large infrastructure planning requires close attention to trade volume as well as value. 3.3. Analysis of top trade gateways As mentioned in section 3.1, the top 12 (of 48 total) trade gateways by trade value are two airports, five seaports, and five land gateways. Together, they accounted for 85.6 percent of total trade in 2016. Figure 3.7 ranks trade value via the top 12 gateways between 2011 and 2016, which has changed overtime. For example, Noi Bai Airport moved up from third in 2011 to second in 2016. Similarly, Tan Son Nhat Airport upgraded from the fifth position in 2011 to the fourth position in 2016. Meanwhile, the rank of Hai Phong seaport complex has fallen from the second position in 2011 to the third position in 2016, and Cai Lan seaport ranked sixth in 2011, fell to eighth in 2016. In terms of land-gateways, Huu Nghi Lang Son moved up from eighth in 2011 to fifth in 2016. FIGURE 3.7. Top 12 gateways by trade value Trade 2011 Trade 2016 HCMC Seaport Complex 92.5 HCMC Seaport Complex 103.9 Hai Phong Seaport Complex 39.1 Noi Bai Airport - HN 76.7 Noi Bai Airport - HN 16.5 Hai Phong Seaport Complex 59.6 Vung Tau Seaport Complex 15.0 Tan Son Nhat Airport - HCMC 27.9 Tan Son Nhat Airport - HCMC 14.6 Huu Nghi - Lang Son 8.3 Cai Lan Seaport - Ha Long 6.6 Vung Tau Seaport Complex 6.1 Tien Sa Seaport - Da Nang 2.8 Tien Sa Seaport-Da Nang 4.9 Huu Nghi - Lang Son 1.9 Cai Lan Seaport - Ha Long 4.5 Mong Cai - Quang Ninh 1.9 Mong Cai - Quang Ninh 1.7 Lao Cai 1.5 Quang Binh - Quang Tri 1.3 Tay Ninh 0.7 Tay Ninh 1.2 Quang Binh - Quang Tri 0.5 Lao Cai 0.7 - 100 -40 60 US$ Billions US$ Billions Source: Customs data, calculation by authors. Changes in trade value through the top 12 gateways, especially in terms of ranking interexchange in transportation modes, reflect changes in the product mix structure of key value chains. Among the top 12 gateways, there are six that are most statistically significant: HCMC Seaport Complex, Hai Phong Seaport Complex (Hai Phong City), Vung Tau Seaport Complex (Ba Ria-Vung Tau Province), Tan Son Nhat Airport (HCMC), Noi Bai Airport (Hanoi City), and Huu Nghi land gateway (Lang Son Province). 36 Vietnam: Connecting value chains for trade competitiveness FIGURE 3.8. Six most significant gateways by trade value HCMC Seaport Complex: 29.6% of total trade (2016): Hai Phong Seaport Complex: 17.0% of total trade (2016): Saturation Sign of saturation 60 100 US$ Billions 25.3 26.8 US$ Billions 54.6 54.6 47.6 48.0 40 21.0 23.7 46.0 51.2 14.7 17.9 50 20 33.6 32.8 49.9 52.9 52.2 55.9 25.1 30.7 32.0 46.6 45.9 24.4 0 0 2011 2012 2013 2014 2015 2016 2011 2012 2013 2014 2015 2016 Vung Tau Seaport Complex: 1.7% of total trade (2016): Lang Son Land - Port: 2.4% of total trade (2016): Fluctuation Growing and change in structure 8 1.2 2.3 1.1 US$ Billions US$ Billions 10 8.7 10.0 8.7 0.9 3 5.5 6.8 6.0 4.1 0.4 0.7 3.9 3.0 1.6 1.8 6.3 6.2 6.3 1.8 4.2 2.5 3.1 2011 2012 2013 2014 2015 2016 0 -2 2011 2012 2013 2014 2015 2016 Noi Bai Airport: 21.9% of total trade (2016): Tan Son Nhat Airport: 8.0% of total trade (2016): Steadily growing Growing 80 30 US$ Billions 60 12.5 44.3 US$ Billions 39.5 20 10.5 40 30.8 26.8 7.8 10 6.6 7.3 7.1 20 15.2 28.8 32.4 12.62 15.40 8.3 22.5 23.3 9.41 8.3 15.5 7.99 7.71 7.26 0 0 2011 2012 2013 2014 2015 2016 2011 2012 2013 2014 2015 2016 Import Export Source: Customs data, calculation by authors. In short, data analyzed in this chapter shows trade value increased but unevenly across gateway type, leading to structural change in the relative importance of gateway type. For example, the share of trade via air gateways increased rapidly from 15.6 percent in 2011 to 39.5 percent in 2016, while the share of trade via sea gateways reduced dramatically from 78.8 percent in 2011 to 56.1 percent in 2016. This reflects the drastic shift in export structure from primary exports including crude oil and non-oil (coal, stone, sand, gravel, aluminum, copper, etc.) and resource-based exports (agriculture-based products) to high-tech exports (electronics, cell phones, incorporated circuits, etc.). It reflects the structural change in products: rapid increases in products with small volumes but high values, such as mobile phones, electronic components, high fashion exports, and high-value, processed agricultural products. Export structural changes bear heavily on investment allocation to international gateways according to appropriate logistics modes. This report reveals the current air gateway capacity would not be able to catch up to this shift in export structure (especially for Tan Son Nhat airport). The report recommends including value-chain analysis in developing a policy framework for international trade gateways (which is currently lacking) that would guide related master planning and prioritized investment. Additionally, the capacity of Ho Chi Minh City’s seaport complex is saturated and further capacity extension seems impossible (due to its limited area and urbanization). The report recommends enhancing the utilization of Vung Tau’s seaport complex and a substitute solution for the current situation in Ho Chi Minh City’s seaport complex. From the report’s findings, we recognize the importance of airports and seaports, but it is even more important to focus policy on the ease of access and egress to those gateways. Chapter 3 – Efficient international trade gateways 37 CHAPTER 4 Regional specialization and coordination This chapter recommends transport infrastructure investment for regional development take into account effective regional links and connectivity of value chains. Moreover, decisions on transport infrastructure investment should be based on a conducive environment for regional specialization and inter-regional cooperation (rather than unhealthy competition) for public investment sources. 38 Vietnam: Connecting value chains for trade competitiveness 4.1. Provincial specialization Provincial specialization in all 14 identified (see report section 2.1) value chains and their segments is defined based on Location Quotients (LQs) computed from the Enterprise Census 2011 and 2016. Map 4.1 shows provincial specialization in northern Vietnam for 2016 and Map 4.29 shows provincial specialization in southern Vietnam. In addition, comparing LQs for 2011 and 2016 allows us to take a dynamic perspective of specialization patterns in each of the provinces. Location quotients that increase over time (for a particular subsector) reflect stronger relative provincial specialization. In contrast, provinces with lower LQs reflect lower relative specialization for the selected subsectors. MAP 4.1. Locational distribution of provincial specialization in the North Dien Bien Lao Cai Ha Giang Cao Bang Bac Kan Tuyen Quang Lang Son Quang Ninh Ciment_Products Planting_vegetable Logging Other_garments Planting_Rice Aquaculture Seeding Mining_iron Planting_Rupber Rice Planting_vegetable Petro_hole_seller Logging Iron_Products Aquaculture Processing_Rubber Planting_foresting Processing_Co ee Mining_iron Planting_foresting Ciment_hole_seller Ciment_Products Sawmilling Planting_vegetable Fertilizers Planting_foresting Rice Planting_vegetable Sawmilling Leather Sawmilling Ciment_hole_seller Processing_Co ee Wood_products Processing_Co ee Clothing Planting_vegetable Aquaculture Mining_iron Mining_clay_sand Petro_hole_seller Mining_clay_sand Mining_clay_sand Mining_iron Petro_hole_seller Lai Chau Mining_clay_sand Producing_yarns Aquaculture Coal Yen Bai Iron_hole_seller Ciment_hole_seller Petro_hole_seller Ciment_hole_seller Planting_Rupber Phu Tho Bac Giang Iron_hole_seller Processing_Co ee Seeding Planting_Rice Rice Thai Nguyen Mining_clay_sand Ciment_hole_seller Aquaculture Planting_foresting Final_products Planting_foresting Iron_hole_seller Sawmilling Ciment_products Petro_hole_seller Petro_hole_seller Mining_clay_sand Coal Bac Ninh Sawmilling - E&E Final_products Son La Seedling Ciment_hole_seller Vinh Phuc Hai Duong Aquaculture Ciment_hole_seller Mining_iron Planting_vegetable Systems_modules Aqua_food Processing_Co ee E&E Hung Yen Iron_Products Iron_Products E&E Thai Binh Footwear Parts_components Hoa Binh Iron_Products E&E Planting_Rice Hanoi Petro_hole_seller Hai Phong Rice Other_garments Petro_products Aqua_food Planting_Rice Ninh Binh Ciment_products Logging Processing_Co ee Systems_modules Petro_hole_seller Petro_hole_seller Nam Dinh Iron_hole_seller Producting_fabric Parts_components Ha Nam Aquaculture Processing_vegetable Iron_hole_seller Parts_components Ciment_hole_seller Petro_hole_seller Fertilizers Mining_clay_sand Producing_yarns Petro_hole_seller Ciment_hole_seller E&E E&E Other_garments Final_products Ciment_products Cases Clothing Ciment_products Final_products Clothing Disclaimer: The boundaries, colors, denominations and other information shown on any map in this work do not imply any judgement on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Source: Enterprise Census 2016 calculation by authors. 9 Map 4.2 shows Ho Chi Minh City specializes in three sectors: fertilizers, iron, and petroleum. This may result from data constraint or the fact that headquarters of these sectors’ largest producers are located in HCMC, even though their production is elsewhere. Chapter 4 – Regional specialization and coordination 39 MAP 4.2. Locational distribution of provincial specialization in the South Long An Binh Duong Tay Ninh Binh Phuoc Ninh Thuan Petro_products Leather Ciment_Products Seedling Logging Fertilizers Processing_Co ee Aqua_food Planting_Rice Rice Rice Aqua_food Leather Sawmilling Mining_clay_sand Iron_Products Sawmilling Producing_fabric Producing_fabric Precessing_Co ee Binh Thuan Processing Footwear Producing_yarns Processing_vegetable Coal Fertilizers Producing_fabric Wood_products Footwear Footwear Other_garments Breeding Other_garments Processing Processing_Rubber Cases Breeding Clothing Clothing An Giang Dong Nai Processing_Rubber Logging Aquaculture Seedling Aqua_food Systems_modules Mining_clay_sand Dong Thap Leather Cases Seedling Sawmilling Ciment_Products Fertilizers Iron_Products Rice Petro_hole_seller Processing_Co ee Footwear Processing Wood_products Ba Ria - Vung Tau Can Tho Processing_Rubber Vinh Long Breeding Fertilizers Aquaculture Ho Chi Minh City Planting_vegetable Kien Giang Processing_vegetable Petro_products Fertilizers Petro_products Sawmilling Cases Rice Iron_hole_seller Seeding Ciment_Products Footwear Mining_clay_sand Petro_hole_seller Petro_hole_seller Mining_clay_sand Other_garments Fertilizers Petro_hole_seller Fertilizers Processing_vegetable Tien Giang Petro_hole_seller Other_garments Ben Tre Cases Producing_yarns Rice Ca Mau Fertilizers Footwear Cases Processing Logging Hau Giang Soc Trang Breeding Clothing Iron_Products Fishing Planting_vegetable Planting_vegetable Tra Vinh Aquaculture Aquaculture Bac Lieu Ciment_hole_seller Fertilizers Aquaculture Producing_yarns LQ_2016 (average) Oil_exploited Planting_Rice Seedling Processing_vegetable Planting_Rice Other_garments 20 to 30 Breeding Aquaculture Processing_vegetable Planting_Rice Petro_hole_seller Processing_Co ee 10 to 20 Fertilizers Coal Aqua_food Aquaculture Cases Cases 5 to 10 Processing Processing Footwear Processing Footwear Cothing 1 to 5 Disclaimer: The boundaries, colors, denominations and other information shown on any map in this work do not imply any judgement on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Source: Enterprise Census 2016 calculation by authors. Table 4.1 exemplifies the relative specialization of the Ca Mau Province in the South. In 2016, the Ca Mau Province specialized in aquaculture (including aqua-breeding, aqua-culturing, and aqua- processing), but also oil and gas exploitation, and fertilizers. The latter two value chains are interlinked because natural gas services input to fertilizer production. This regional specialization trend has been strengthened between 2011 and 2016. TABLE 4.1. Regional specialization of Ca Mau province Change in LQ Specialized Production Employment 2016 LQ 2011 LQ 2016 (2011 - 2016) Aqua-processing 18,622 26.3 27.7 1.4 Aqua-culturing 155 2 5.1 3.1 Aqua-breeding 472 13.2 20.3 7.1 Crude oil and gas extraction 186 0 6.3 6.3 Fertilizers 896 0.2 8.9 8.7 Source: Enterprise Census 2011 and 2016, calculation by authors. Figure 4.1 demonstrates that Ha Nam Province specializes in the subsectors plotted above the x-axis in the upper quadrants (with LQ 2016 > 1), including textile and garment, electronics, cement, auto parts, and animal food. Among those, the subsectors that are plotted in the upper- right quadrant (with higher LQ 2016 > LQ 2011) show stronger relative specialization, while 40 Vietnam: Connecting value chains for trade competitiveness those plotted in the left quadrant (showing LQ 2016 < LQ 2011) show weaker specialization. The textile and garment value-chain structure has changed over time: it is less relatively specialized in fabric and clothing, and more specialized in yarns. This change would require a shift in labor skills, and a conducive policy environment for increased competition in the labor force. The electronics specialization has emerged since 2011. Between 2011 and 2016, Ha Nam became more specialized in manufacturing final cement products, and relatively less specialized in stone and sand exploitation. This may reflect a shortage of local inputs or the provincial government’s increasing awareness of environmental protection. FIGURE 4.1. Regional specialization of Ha Nam province Upper quadrants: LQ 2016 >1; Upper Right Quadrant: LQ 2016 > LQ 2011 15 Stone, sand, gravel, and clay Cement manufacturing quarrying manufacturing 10 Animal food Yarn LQ 2016 5 Clothing Electronic components Fabric Final electronics products 0 Auto parts and components Other garments -5 -10 -5 0 5 10 Change of LQ between 2011 and 2016 circle size = emp2016 Source: Enterprise Census 2011 and 2016, calculation by authors. Regional specialization patterns as demonstrated by the Ca Mau and Ha Nam examples were performed for all 63 provinces. This regional specialization analysis can be used by stakeholders (government, development partners, and the private sector) for various purposes, including but not limited to regional planning and development (an integrated multisectoral approach), interregional coordination, job and skills development analyses, and particularly value-chain/ cluster-based connectivity and GVC integration. Data analysis in this report could be used to provide the LQ 2016 and the change since 2011 (LQ2016-LQ2011) for all 14 identified sectors across all 63 provinces. Changes in provincial specialization can create opportunities for lagging regions like poor and remote provinces. This report shows the garment segment of the textile and garment value chain has shifted from Red River Delta provinces (Hai Duong, Bac Ninh, Ha Nam) to a lagging province (Tuyen Quang) between 2011-2016. The shift in provincial specialization needs to be reviewed more closely in separate focused studies. Chapter 4 – Regional specialization and coordination 41 4.2. Aligning trade and growth strategies with regional specialization In Vietnam, a higher relative regional specialization in manufacturing activities (as measured by LQ) is associated with higher per capita income (and to a lesser extent exports) and lower regional poverty (Figures 4.2-4.4). FIGURE 4.2. Manufacturing agglomeration Understanding the geographic structure versus provincial income of value chains is important for formulating 2.50 y = 0.1207e 0.5317x policy on regional coordination, integrating Dong Nai R² = 0.1369 value-chain links, and for planning 2.00 Long An Binh Duong human capital in appropriate regions and Tay Ninh Bac Giang Bac Ninh areas accordingly. Changes in provincial Manufacturing LQ Ben Tre Hai Duong 1.50 Thai Nguyen Nam Dinh Quang Nam Hai Phong specialization can create opportunities for Binh Phuoc 1.00 Tuyen Quang Ca Mau Vung Tau lagging regions, and value-chain develop­ Ho Chi Minh Yen Bai Da Nang ment could help address regional inequality Lam Dong Ha Noi 0.50 Son La issues. The report therefore recommends Cao Bang Ha Giang Gia Lai sharing provincial specialization information Dien Bien Kon Tum 0.00 2 3 4 5 6 with all concerned parties including central and local governments, the private sector, Source: Enterprise Census 2011 and 2016, Customs data, calculation by authors and development partners. FIGURE 4.3. Manufacturing agglomeration The report also recommends that investment versus provincial trade in transport infrastructure for regional development planning and implementation 18.00 Thai Nguyen Bac Ninh Binh Duong consider effective regional links and 16.00 Ho Chi Minh Bac GiangDong Nai Ha Noi Vung Tau Hai Duong Quang Ninh Da Nang Ninh Binh Binh Phuoc Tay Ninh value-chain connectivity, not only public 14.00 Ca Mau Tien Giang Kon Tum Bac Lieu investment discretions. Public investment, Ln Export per capita Gia Nong Lai Tra Vinh 12.00 Dac Vinh Long Tuyen Quang 10.00 Son La Ninh Thuan especially for transport infrastructure, can be Dien Bien 8.00 Cao Bang optimal if it is not fragmented or duplicated Lai Chau 6.00 due to excessive regional competition, but 4.00 allocated for objective regional coordination Bac kan 2.00 based on real value-chain spatial structure 0.00 0.50 1.00 1.50 2.00 Manufacturing LQ and links. Source: Enterprise Census 2011 and 2016, Customs data, calculation by authors 42 Vietnam: Connecting value chains for trade competitiveness FIGURE 4.4. Manufacturing agglomeration Our evidence shows manufacturing density versus poverty and value-chain concentration have a positive interrelationship with income, 30 Lai Chau exports, and employment in local areas. Percentage of poverty household in Dien Bien 25 Ha Giang Cao Bang Regional specialization measured by LQ provincial population (%) 20 Son La Lao Cai Yen Bai is a dynamic metric, changing over time. Bac Kan 15 Kon Tum Lang Son Hoa Binh Provinces change their participation in Nghe An Quang… Tuyen Quang Thai Nguyen 10 Dak Nong Phu Yen Hau Giang Phu… Tra Vinh the value chain for various reasons. For Quang Tri Binh Phuoc Bac Giang 5 Dak Lak Ninh Thuan Ha Nam Tien Giang Long An example, emerging value-chain expansion Kien Giang Da Nang Nam Dinh 0 Ha Noi HCMC Hai Phong Dong Nai Tay... Binh Duong following a move of foreign-led firms to - 0.50 1.00 1.50 2.00 the locality as in the case of Samsung, Manufacturing LQ 2016 policy changes, increased skilled human Source: Enterprise Census 2011 and 2016, Customs data, calculation by authors resources, or internal labor movements. The information on local specialization is important for understanding the geographic structure of value chains. The government needs this information to formulate policy on regional coordination, integrate value-chain links, and for support human capital improvements nationally. 4.3. Core versus lagging regions Several issues have been identified that require reform to overcome the constraints of low agglomeration and longer transport distances in Vietnam (Farole and Winkler, 2012): First, industrial land in Vietnam is cheap, while urban-residential land is expensive. Land policies have caused excessive conversion of agricultural to industrial land but prevented urbanization of people and jobs. Second, the current city classification system could be updated and refined. It currently promotes urban expansion and investment, but in an unsystematic way that does not take into account a specific city’s needs. It could also include a spatial development framework aligned with Vietnam’s economic development strategy. Third, there is lack of coordination, particularly of policies, at the regional and national level, especially regarding mass transit infrastructure development. Integrated transport and logistics platforms need to be mainstreamed. Key supply-chain bottlenecks include roads to big ports, and important road corridors and expressways. Fourth, the government should not neglect rural- urban links to strengthen the competitiveness of the agribusiness sector. They should focus on connecting regional cities to major agricultural zones (for example the Mekong River Delta and Central Highlands) (World Bank, 2016). Finally, it is crucial to acknowledge that different types of lagging regions require different types of policies. Agglomeration research shows the likelihood of exporting is higher in core regions and this has implications for national and regional policies. There also needs to be the right balance between connectivity policies, particularly in lagging regions, and policies addressing other critical factors, or policies for attracting foreign direct investment (FDI) (see Box 4.1). Chapter 4 – Regional specialization and coordination 43 Box 4.1. Policies for different types of lagging regions Agglomeration research shows the likelihood of exporting is higher in core regions. Core regions are characterized by a higher density of firms and exporters in a specific sector (localization economies and export spillovers), but also by a significant sector diversity that allows for resource sharing including specialized suppliers and labor (urbanization economies). Both help overcome the fixed-entry costs to exporting. As a result, firms located in core regions tend to export more compared to firms located in noncore (lagging) regions. They also import more, underlining the importance of imports for increased export competitiveness in the context of GVCs (import-to-export). The findings of this agglomeration research and the likelihood of exporting has implications for national and regional policies. It is widely recognized that past interventions that specifically aim to lower spatial inequalities within countries, ranging from infrastructure investments, wage policies, deregulation, promotion of clusters, development of industrial parks and economic zones, and fiscal incentives to attract investment have been unsuccessful. In a world where import and export times are critical, particularly in manufacturing, attracting investment to noncore regions can seriously impede firms’ overall competitiveness. But where the opportunity to attract investment to noncore regions is realistic, it is crucial to identify and address location-specific barriers to importing and exporting. This is more likely to be the case for peripheral regions that have a larger economic mass and greater opportunity to link to regional and global value chains, whereas other peripheral regions are better suited to serve domestic markets (see framework table below). Suitable policies should focus on two objectives: (i) increasing the competitiveness of the region and its firms, and (ii) improving its connectivity with domestic and international markets. This requires countrywide growth and trade strategies to be aligned with regional comparative advantage. A framework for competitiveness policies in different types of lagging regions: Region type Nature of policies Near the core - Many traditional regional policies may be effective, including investment incentives and export-oriented incentives - Promotion and facilitation of agglomeration, including industrial parks/SEZs and cluster policies - Investment climate reforms Peripheral but with economic - Targeted FDI attraction (following comparative advantage and mass industry lifecycles) - Support for competitiveness of existing industry clusters - Transport connectivity and infrastructure - Investment climate reforms - Firm-level competitiveness interventions (training, finance, etc.) - Critical importance of governance Peripheral and without density - Limited prospects for export-oriented investment – focus on endowment-based opportunities if applicable (such as mining, agriculture, tourism) - Focus on social infrastructure and connectivity - Firm-level competitiveness interventions 44 Vietnam: Connecting value chains for trade competitiveness Better connectivity represents a “two-way road”. On one hand, it attracts new investors. But if other critical bottlenecks are not addressed at the same time, the increased competition might force local firms and resources to leave the region (brain drain). It is also important to consider the nature of FDI when formulating policy. Connectivity is much more important for efficiency-seeking FDI than for market-seeking FDI, since accessibility to core regions and to international trade gateways are more relevant for the former. Source: Farole and Winkler (2012), Farole (2013). Chapter 4 – Regional specialization and coordination 45 CHAPTER 5 Economic zones and value chains This chapter emphasizes the necessity of rethinking and modernizing industrial and economic zones to make them best support domestic supply chains links and connectivity for better GVC integration. 46 Vietnam: Connecting value chains for trade competitiveness 5.1. Industrial agglomeration/concentration through economic zone development and value-chain development Vietnam established its first economic zone in Tan Thuan in Ho Chi Minh City (HCMC) in 1991. Since then six models of economic and industrial concentrated zones have been formed and developed in Vietnam10. These zones have defined geographic boundaries, often in geographically advantageous locations, and operate based on specific preferential policies separated from the rest of the economy to promote exports, attract Foreign Direct Investment (FDI), and create jobs, among others. To date, Vietnam has 18 coastal economic zones, 26 border economic zones, and 328 industrial parks as shown in Map 5.1.A. The economic and industrial zones have cumulatively attracted 52 percent of the total FDI in Vietnam, and have accounted for 42 percent of the total industrial output and 52 percent total exports nationwide (MPI, 2018). Despite considerable achievement over the past 25 years, these models have had mixed results in meeting their desired objectives. The existing zone model faces several constraints, namely the lack of firm links within and outside zones, mostly between lead FDI firms and domestic firms in domestic supply chains, leading to weak spillover effects. In addition, the inconsistent quality and availability of infrastructure inside and outside these zones limits efficient connectivity and locational advantage from a supply-chain perspective. Moreover, the acceleration of planning and establishing economic and industrial zones in the absence of a clear strategic vision, clear demand, and appropriate master planning create unnecessary competition among provinces for limited infrastructure development resources. Overall, the zone model faces many challenges, including that the zone planning process—still mostly driven by local government real estate development initiatives—does not match industrial demand (the average occupancy rate of zones in Vietnam is about 40 percent), and industrial and trade- related stakeholders (like the Ministry of Industry and Trade) have not been involved in the process of building a nationwide industrial strategy. These challenges lead to fragmentation and inconsistency. Zones are heavily reliant on fiscal incentives (most zones are competing for generous incentives and as a result are less focused on the overall business environment) and the demand for both skilled and unskilled labor in the zones outnumbers the supply. Furthermore, there is no monitoring and evaluation framework and most zones do not have a solid system to regularly collect and analyze relevant data or help to gauge their progress. The characteristics and spatial structure of industrial parks and economic zones are far different from those of value chains. Map 5.1 compares the spatial structure of industrial zones to the spatial structure of the textile and garment value chain. Although industrial and economic zones are formed on a small, delimited area designed for multidisciplinary industries with preferential policies, the spatial structure of a value chain opens up in larger areas, sometimes much larger, with preferential policies not applying to the whole value chain. 10 Namely, Export Processing Zones (EPZs), Industrial Parks (IPs), High-Tech Parks (HTPs), Economic Zones (EZs), Concentrated Information Technology Zones (CITZs), and High-Tech Agriculture Parks (HTAPs). Chapter 5 – Economic zones and value chains 47 MAP 5.1. Spatial structure of industrial zones versus the textile and garment value chain MAP 5.1.A: Industrial zone spatial stucture MAP 5.1.B: The textile and garment value chain spatial stucture Industrial Zones by Area Central Coast North Central Highlands Southeast Mekong Delta Red River Delta Disclaimer: The boundaries, colors, denominations and other information shown on any map in this work do not imply any judgement on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Source: MPI and other sources. Source: I/O Table 2016, Enterprise Census 2011 and 2016. International experience shows (Zeng, 2010) that the difference in spatial structure and the policy disparity inside and outside the fenced zones generally prevents or constrains links throughout the entire value chain. Depending on the characteristics of each chain, the effect of interaction differs. The majority of FDI firms get involved at the final processing stage of the production for export, and are located inside zones. Firms at the upstream segments of the value chains—often domestic, private enterprises—are mostly located outside zones (Table 5.1). 48 Vietnam: Connecting value chains for trade competitiveness TABLE 5.1. Share of establishments, employment, and revenue of firms in zones Firms in Firms’ Firms’ zones (%) employment revenue in in zones (%) zones (%) 1. Aquaculture - Animal food 18.8 59.7 63.4 - Processing 16.5 44.3 46.9 2. Textile and garment - Producing yarns 30.9 72.3 83.9 - Producing fabric (weaving, knitting, finishing) 19.5 56.5 67.1 - Clothing 7.4 36.2 36.7 - Other garments 9.2 33.7 41.1 3. Leather and footwear - Leather 50.9 82.0 87.0 - Leather products 14.3 8.9 35.7 - Cases and bags 8.8 47.6 44.7 - Footwear 18.5 47.4 52.0 4. 3C electronics (consumer, communication, computer) - Electronic components 48.6 78.5 78.6 - Subassemblies - Final products 43.4 90.0 97.0 5. Automotive vehicles - Parts and components (PTLK) 50.7 87.1 89.9 - Systems - modules 21.7 64.4 73.8 - Final assembly 32.3 40.8 42.2 6. Wood products - Sawmilling 5.1 12.2 16.6 - Wood products 5.4 15.3 23.0 7. Rice - Seedling 1.3 14.2 25.9 - Planting 0.6 0.4 0.2 - Rice 6.0 14.5 12.5 8. Coffee - Processing 5.5 29.5 69.8 9. Rubber - Processing 7.6 8.1 45.6 10. Vegetables and fruit - Processing 6.4 15.4 17.7 Source: Enterprise Census 2016. The aquaculture value chain analysis in the report shows that only 16.6 percent of aqua-processing firms are located in economic zones, but account for 44.3 percent of sectoral employment and generate 46.9 percent of sectoral revenue, showing economic zones are hosting the anchor firms in this particular value chain (see Table 5.2). Chapter 5 – Economic zones and value chains 49 TABLE 5.2. The aquaculture value chain and related industrial and economic zones Aquaculture value Firms Employment Revenue (Billion VND) chain Total In IPs % Total In IPs % Total In IPs & % & EZs & EZs EZs Animal food 776 146 18.8 73,652 43,985 59.7 255,633 162,055 63.4 Aqua-breeding 508 0 0 6,819 0 0 2,366 0 0 Farming (aqua-culturing) 574 0 0 8,981 0 0 3,376 0 0 Nonfarming (fishing) 826 0 0 36,259 0 0 10,265 0 0 Aqua-processing 1,264 208 16.5 197,171 87,442 44.3 228,042 106,849 46.9 Source: Enterprise Census 2016. Map 5.2 shows the distribution and colocation of aqua-processing and animal food segments, and their firms in the top ten (with industrial density measured by district-level LQs) in relation to industrial parks and economic zones, whose location is marked by stars. MAP 5.2. The aquaculture value chain and economic zones Disclaimer: The boundaries, colors, denominations and other information shown on any map in this work do not imply any judgement on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Source: Enterprise Census 2016, calculation by authors. 5.2. Economic zones and clusters An industrial cluster is generally defined as a geographic concentration of interconnected firms in a particular field with links to related institutions. Although clusters come in various forms and several scholars have tried different typologies, all clusters share one commonality: each comprises numerous different size firms belonging to one branch of industry. Box 5.1 presents China’s experience in developing cluster-based economic zones and facilitating localized cluster growth. 50 Vietnam: Connecting value chains for trade competitiveness Box 5.1. China’s experience with economic zones and clusters: “top-down” versus “bottom-up” While SEZs are normally constructed through a “top-down” approach by government policies, most clusters are formed in an organic way through a “bottom-up” process. Some clusters, however, have emerged from or within industrial parks or export-processing zones over time, but not often in low- and middle-income countries. Because cluster formation takes time and requires an ecosystem based on market forces, the purely top-down approach to cluster creation should be exercised with caution, especially in low-capacity countries, where many such efforts have failed. The challenges, however, should not necessarily prevent governments from facilitating the formation, growth, or scale-up of emerging clusters. Inevitably, it is easier to devise policies for a functioning cluster and devilishly hard to call a cluster into existence. In this sense, a mixture of bottom-up and top-down approaches to cluster development is possible. Some cases in China perfectly illustrate this “mixed” approach. Despite that governments can have more control in developing SEZs than industrial clusters, an SEZ is not necessarily easier to develop, and many SEZ initiatives have failed. The success of SEZs requires a capable government and a well-functioning market system, at least within the zone or park. Designing an SEZ using a purely cluster approach might be possible but there is a risk of failure unless the market signals are clear, and the government has a perfect understanding of the domestic comparative advantages, and market situations (both domestic and international) that are often beyond their capacity. In China, while market forces are usually responsible for the initial inception of local industrial clusters, the government supports or facilitates them in various ways, including setting up an industrial park based on an existing cluster, like the Wenzhou (Zhejiang province) footwear cluster, and the Foshan (Guangdong province) home appliances cluster, etc. Meanwhile, after decades of development, some clusters have begun to grow out of certain SEZs, like the information and communication technology clusters in Zhongguancun (Beijing) and Shenzhen, the electronics and biotech clusters in Pudong (Shanghai), the software cluster in Dalian, and the optoelectronics cluster in Wuhan. The emergence of these clusters actually hinges on the success of the SEZs, which serve as their “greenhouse,” and on market forces over time. Furthermore, in recent years, some cities have set up cluster-type industrial parks, or “specialized industrial parks,” such as the liquid crystal display (LCD) high- tech park in Kunshan, and the Wuxi Wind Power Science and Technology Park and the Photovoltaic Industry Park in Jiangsu Province. In these examples, two different models are converging. However, despite the fact that in recent years SEZs and clusters in China have overlapped to some extent, in most cases their origins, development trajectories, market segments, firm compositions, operation levels, and success factors are different. The key factors of China’s success in developing SEZ and industrial cluster include: - Strong governmental commitment and support to pilot market-oriented economic reforms. This determination ensures a stable and supportive macroenvironment for reform and new policies. - Institutional autonomy. Local governments were given the authority to develop and manage zones or clusters, which allowed them more freedom in pursuing new policies and development measures deemed necessary to vitalize the economy. - Technology learning, innovation, upgrading, and strong links with the domestic economy. It is important to emphasize technology learning and innovation, and foster links between zones, domestic enterprises, and industrial clusters through supply chains or value chains. - Clear objectives, benchmarks, and competition. Coupled with their autonomy, SEZs or industrial clusters are closely measured and monitored against their objectives, and competition also helps to maintain good performance. - More support for local clusters. While the government put a lot of effort into SEZs, they also gave more support to developing and upscaling local clusters. Source: Zeng (2010, 2015). Chapter 5 – Economic zones and value chains 51 5.3. How economic and industrial zones can contribute to value- chain creation and development The first policy objective for leveraging economic and industrial zones to create and develop value chains is that these zones should promote value-chain links and connectivity. This requires a clear strategic vision and appropriate master plans. Second, zone policies also need to be revised and supplemented to promote cluster development. Cluster links, while they may not fully reflect the entire value chain, include important links of one or more value chain that require spatial and policy priorities to facilitate input-output links of the broader chain. Both policy objectives similarly promote links and connectivity. Third, policies should also address the impacts of urbanization and spontaneous zone development along the main transport corridors. The following policy recommendations target these three broad objectives. We acknowledge it is not always easy or even practical to stick industrial zones with policy objectives of creating cluster links. Often the “beggar can’t be the chooser”, especially for poorer and peripheral provinces. The successful model of the Nomura Industrial Zone in Hai Phong was so costly that even Ho Chi Minh City, Hanoi, and Danang—which are in a much better position to select and direct FDI projects—failed to duplicate it due to immediate pressure to attract investors. This suggests that good master planning is not enough. Supportive measures to nurture and strengthen business links along the value chains to build up clusters are more crucial to achieve the master plans’ goals. Vietnam’s recently launched supplier development program is a supportive measure that developed to ensure firms can respond to investor demand, build small and medium enterprise (SME) capacity, and embed a comprehensive set of policies in links. The program was launched by the Industry Agency under the Ministry of Industry and Trade in collaboration with IFC in 2018. Policies should also address the impacts of spontaneous zone development along the main transport corridors. In Vietnam, there seem to be issues with over or under capacity of infrastructure (for example ports) between the North and South. Policies should promote an integrated approach to infrastructure upgrading within transport corridors, particularly those connected to trade gateways, to avoid independent upgrading (for example of a port) and bottlenecks in connecting infrastructures (for example connector roads). It is also important to ensure SEZs that aim to attract FDI do not penalize local procurement, but instead, that such policies support links to local suppliers (see Box 5.2). 52 Vietnam: Connecting value chains for trade competitiveness Box 5.2. Policies supporting links and the role of economic zone Investment, incentives, and rigid requirements from foreign investors should not penalize local procurement. First, investment incentives like waiving import duties could disincentivize local sourcing, if value-added tax or other taxes still apply for domestic procurement. Other disincentives include restrictions on the flow of goods and labor in and out of SEZs, or barriers for trading between firms inside SEZs with firms outside. The reason for these practices could be a combination of export processing focus from foreign investors, but also the spatial and legal structures that govern SEZs thereby inhibiting integration of domestic actors outside the SEZs. Second, the government should encourage foreign investors to use flexible approaches to local sourcing to integrate domestic firms, especially SMEs, into their supply chains. Innovative approaches could include breaking procurement into smaller lots, establishing parameters for contracting with groups of smaller firms, or offering accelerated payment terms and upfront payment. Strategies and incentives policies to attract foreign investors can incorporate investment targets. These include access to fiscal incentives and privileged land and facilities (for example, in economic zones). For example, the government could require potential investors to prepare a local link and spillover strategy as part of their licensing application package. This will work well for large-scale FDI projects that are likely to deliver substantial long-term rents. In cases where potential investors are competing for access to a concession, an exploration license, or some other exclusive right, the submission of their local link and spillover strategy could be a component of the evaluation. In Australia’s mining sector, for instance, an Industry Participation Plan is required to access tariff concessions offered to investors in the sector. Criteria for assessing an Industry Participation Plan include, among others: employment creation, skills transfer, regional economic development, technology transfer and R&D, and “full, fair and reasonable opportunities” for suppliers to tender. Below is a summary of policy recommendations that can be considered when implementing link policy, including targeting, institutional arrangements, coordination, and monitoring: - Spillover policies should be integrated directly into national industrial policies. - Many low-and-middle-income-country governments will need to build capacity in their own institutions to implement spillover policy effectively. - Responsibility for delivering the spillover agenda should be held at a senior ministerial level rather than as an add-on activity for the investment promotion agency. - Given the huge potential base of beneficiaries in the domestic supply sector, targeting will be necessary. - Another approach for targeting and efficient delivery may be to implement some supply-side interventions through existing industry clusters. - Where technical support is provided to domestic SMEs, concurrent financing support should be included. - Finding sustainable funding for link and spillover programs should be a priority at the outset. - Matching grant programs can be another way to develop sustainable funding and can help crowd-in additional financing from foreign investors. - Establish sector forums for communication and coordination between government and the private sector around links and spillovers. - Multistakeholder partnerships can be effective in designing and delivering link and spillover programs. - Multistakeholder dialogues can be effective in managing expectations. - Monitoring is critical to ensure more effective policy and to encourage transparency and facilitate communications. Many country cases suggest policies fostering links as part of systematic supplier development programs, industrial upgrading programs, and spatial development programs, that also integrated investors and their requirements, and emphasized local capacity building were most successful. These are seen in Singapore, Malaysia, Chile, Costa Rica, and South Africa, compared to measures that were merely supply-driven, and neither Chapter 5 – Economic zones and value chains 53 comprehensive nor focused on active support for SME capacity building. One issue to consider is scale: many programs can only handle a limited number of suppliers, so including the private sector or establishing public- private partnerships is key. But case studies also show that programs targeting the most promising suppliers led to the best outcomes. The government should lead the planning on the comprehensive framework for link development and discuss with multinationals who take a lead in implementing these programs. This parallel approach will help leverage complementary activities and secure additional private-sector funding. It will also be effective in meeting investor demand. For example, foreign investors could provide operational assistance to suppliers, while government support programs focus on more general management and technical training. Or, multinationals could offer consultancy on quality improvement, while national quality bodies assist with certification. Or, investors might offer technical training, while government programs incentivize the use of new technology. Experience has shown that supplier development programs are instrumental in attracting and retaining additional FDI. This is because foreign investors that have developed strong links in the host economy are less likely to leave due to the high costs of building new supplier networks. Small development projects (SDPs) can also help attract additional FDI. Established local firms that are productive and supply multinational enterprises (MNEs) are a big draw for new investors looking to set up operations in a country. Source: Farole and Winkler (2014). While private firms can help train the local workforce, governments should take a proactive training approach to prepare their labor force and domestic firms to meet the needs of foreign operations. This should be part of the government’s investment promotion plan whereby they identify what type of FDI they would like to attract and then conduct a supply-side analysis to assess whether the current workforce has the necessary skills (vocational, managerial, etc.) and whether domestic firms can meet the required standards (for example labeling standards, sanitary, and phytosanitary standards, etc.). For example, in countries with plentiful natural resources, governments can set up centers for necessary vocational training and certification like for oil and gas technicians. Similar programs can be developed for the manufacturing, agriculture, and services sectors. The government can also work with education and training institutions and industrial associations to set up apprenticeship programs and provide incentives like training credits to encourage the private sector to collaborate on vocational training and certification. Box 5.3 describes an ongoing donor-funded initiative in Uganda (the E4D/SOGA program) that provides health, safety, and environment (HSE) and bid management training to local suppliers. 54 Vietnam: Connecting value chains for trade competitiveness Box 5.3. Employment and skills training in uganda (E4D/SOGA) The E4D/SOGA project promotes local skills development and enterprise capabilities to enable participation in natural resource-based industries. The project is jointly funded by BMZ, UKAID, NORAD, and Shell, and implemented by GIZ GmbH. E4D/SOGA supports local suppliers in Uganda’s oil industry through HSE training (phase I) and bid management training (phase II). Phase I was implemented in 2016–17 by E360 (a Ugandan firm specialized in HSE training), Astutis (a leading international HSE training provider), and the Association of Ugandan Gas and Oil Service Providers (AUGOS). This project phase was implemented in three stages to enable 30 local companies to successfully adopt industry- compliant HSE practices. In the first stage, the program selected 30 companies and 60 participants in consultation with AUGOS, and assessed firm-level training needs. In the following stages, participants completed two weeks of in-house, tailored HSE training, and two months of coaching at the firm level to implement company-specific HSE changes. Preliminary results suggest that 24 of the 30 companies believe they are better equipped to win tenders as a result of the project, and 128 supplier agreements have been made since. A total of 462 jobs have been created as a result of the increase in contracts awarded to these companies. Following the positive impact of phase I, E4D/SOGA (in conjunction with E360) is implementing a second phase to improve the competitiveness of Ugandan enterprises in bidding processes. Over 230 companies applied for the training, from which 40 companies and 80 decision-makers were selected through a competitive process. This ongoing initiative (launched in January 2018) plans to train participants in market research, developing sales master plans, pricing strategies, financial planning, and setting up strategic partnerships. Twenty of the participating enterprises will receive further support through individual mentoring and coaching. Source: Ritwika Sen, “Enhancing local content in Uganda’s Oil and Gas Industry.” UNU-WIDER working paper 2018/110. In South Korea, the Masan Free Trade Zone (FTZ) provides a good example for generating links between the zone and local firms, where Masan FTZ administration actively promoted the link between local industries and investors in the FTZ. FTZ firms have linked to the local economy through subcontracting and domestic purchases, and have performed positively in generating net exports and spillover effects. By doing so, the zone authority allowed preferential access to intermediate goods and raw materials to local companies supplying FTZ firms. In addition, the zone administration provided technical assistance to subcontracting firms. Granting ‘equal footing’ to local suppliers of capital and intermediate goods, and using subcontracting mechanisms from zone enterprises to local producers were among the most effective measures. These methods, combined with overall trade and investment reforms, fostered successful export-oriented zones and backward/forward links between the FTZ and the local economy (Jeong and Zeng, 2016). Chapter 5 – Economic zones and value chains 55 CHAPTER 6 Implementing trade-oriented connectivity and competitiveness policies This chapter highlights key policy recommendations. 56 Vietnam: Connecting value chains for trade competitiveness 6.1. Making connectivity policy and transport investment more robustly trade-oriented by integrating the comprehensive value- chain connectivity assessment and trade gateways analysis At present, the objectives of improving trade growth and trade competitiveness are not clearly linked with the objectives of developing connectivity policies and investment transport infrastructure. Trade information, especially on value chains, is rarely used in policy formulation and implementation. There remains a lack of in-depth analyses on spatial structure and connective propensity along various linked segments of value chains to inform relevant policies and investment for transport infrastructure development. This chapter suggests related policies, transport master plans, and investment priorities should be formed and implemented to support trade more strongly. Chapter two provided a new four-step methodology for a comprehensive value-chain connectivity and competitiveness assessment that identifies corridors and gateways critical for key domestic export-oriented value chains. These corridors are defined based on spatial structure of input-output links, industrial concentration, and hierarchical connective propensity linking all segments of value chains with international trade gateways. This important information should guide related policies for and investment to transport infrastructure to effectively enhance trade competitiveness and improve GVC integration. The chapter identified key trade corridors and gateways for ten selected value chains (see section 2.1) of national comparative advantage, good trade performance, and governmental priority. These are the aquaculture, textile and garment, leather and footwear, electronics and electrical equipment, motor vehicles, wood products, rubber, rice, coffee, and fruits and vegetables value chains. Furthermore, chapter three scrutinized international trade gateways and their trade flows and structure showing the share of total trade through air gateways has increased rapidly from 15.6 percent in 2011 to 39.5 percent in 2016, while the share of trade via sea gateways plummeted from 78.8 percent in 2011 to 56.1 percent in 2016. This primarily reflects the drastic shift in export structure from primary exports including crude oil and non-oil (coal, stone, sand, gravel, aluminum, copper, etc.) and resource-based export (agriculture-based products) to high-tech exports (electronics, cell phones, incorporated circuits, etc.). This structural change in products –a rapid increase in small but valuable products like mobile phones, electronic components, high fashion exports, and high-value, processed agricultural products requires the transport system and its investments supporting exports to consider a shift from logistics perspectives, based not only on trade growth but also (and more important) on structural change and developing domestic value chains. We suggest formalizing the comprehensive value-chain connectivity and competitiveness assessment and trade gateway analysis into new transport and trade strategies. Actions would require authorities to formalize these analyses and appoint lead agency and research institutions to regularly conduct these studies, guide interagency coordination, and integrate their outputs and outcomes into trade policy, export-import strategy, and national and provincial socio-economic development strategies and master plans. Chapter 6 – Implementing trade-oriented connectivity and competitiveness policies 57 One of the proposed activities is to include the information and policy analyses on the trade flows and key value chains into the transport strategy for 2030. Trade-related indicators should be factored into renewed transport strategy to better benchmark Vietnam against international practice and to monitor policy implementation. Key trade indicators would include trade cost reduction, and Vietnam’s improved position in the interrelationship between efficient connectivity, measured by the quality of trade-related infrastructure, and trade development, measured by trade per capita, etc. In conjunction, the import-export strategy for 2030 should also be renewed to include trade-related infrastructure factors including policies concerning transport and logistics. Similarly, Vietnam should consider including infrastructure-related indicators such as trade-related transport capacity (road, airway, seaway and ports, railway) and logistics performance indicators into import-export strategy to promote this critical policy coordination. In selecting ten value chains to demonstrate the four-step methodology for value-chain-based connectivity and competitiveness analysis in chapter two, the report used existing datasets to produce empirical results. Looking forward, when scrutinizing structural changes in developing GVCs, policy makers should also account for mega trends that may disrupt GVCs, notably the acceleration of the digital transformation and associated de-globalization process. Over the medium to long term, GVCs will consolidate, with fewer countries and firms participating. Automation may result in reshoring manufacturing and therefore the comparative advantage in cheap labor enjoyed by low- and-middle-income countries like Vietnam may be quickly eroding. In other words, infrastructure investments should not only support current economic activities (and, therefore, inevitably reinforce the current economic structure), but also be forward-looking and consider emerging trends and future developments. The proposed methodology allows for close follow-up with dynamic change in spatial structure and connective propensity of existing and emerging value chains in Vietnam. Policy makers may face some trade-offs when using information on GVC-based connectivity for master planning given limited resources and capacity. For example, developing connective infrastructure and gateways to support electronics GVCs may come at the expense of aquaculture value chains. This is already happening in Vietnam, when infrastructure in the Mekong River Delta does not keep pace with rapidly rising demand, while in the North, activities along some highways are relatively low. 6.2. Establishing an efficient mechanism for coordinating trade and transport connectivity and GVC policies, and implementing policy recommendations It is vital to establish an effective interagency coordination mechanism to implement this recommendation for multisectoral policies and investment related to pro-trade transport infrastructure and GVC integration. This mechanism should be put in context for Vietnam to enhance its trade competitiveness. An institutional framework for implementing trade-oriented connectivity policy that supports enhanced national competitiveness and GVC participation is essential. Figure 6.1 presents the integrated four-pillar framework for trade facilitation and logistics (Pham and Oh, 2018). 58 Vietnam: Connecting value chains for trade competitiveness FIGURE 6.1. Integrated four-pillar framework for trade facilitation and logistics 1. Regulatory framework for trade facilitation and standards 2. Trade-related infrastructure and connectivity quality 3. Regulations for logistics services, logistics service providers/users - Actors: Customs, border management - Actors: Government (MOT, - Actors: Transport operators, and trade-related specialized agencies; MOIT, MOF, MPI) and the logistics service providers - Policy levers: (i) rationalize non-tari private sectors; and users; measures; (ii) simplify customs and - Policy levers: (i) master - Policy levers: (i) competitive other border management procedures; planning on trade growth logistics services; (ii) human (iii) apply risk-based inspections and demand; (ii) intermodal capacity. post-clearance audits; (iv) target coordination; (iii) investment Information Communication policy for e cient use with Technology (ICT) and single window; private sector participation. and (v) make regulations transparent. 4. Institutional framework for inter-agency coordination and implementation arrangements - Actors: Government (leaders, public agencies, the private sector, development partners); - Policy levers: (i) improved interagency coordination; (ii) institutional capacity building and change management; (iii) mechanism for enhancing consultation and partnership with the private sector; (iv) clear roles, responsibilities, and implementation arrangements; (v) practical results framework and regular performance monitoring. Source: Pham and Oh (2018). Pillar one covers issues related to the regulatory framework for trade facilitation and standards that involve customs, border management, and trade-related specialized agencies. Pillar two deals with trade-related infrastructure and connectivity quality, the main topics of this report. The main actors are line ministries like Ministry of Transport (MOT), Ministry of Industry and Trade (MOIT), Ministry of Finance (MOF), and Ministry of Planning and Investment (MPI), among others, and the private sector. Policy levers include those responsible for developing connectivity policies, transport master plans, and prioritized investments to support trade growth, for intermodal coordination, and for formulating investment policy for efficient use with private sector participation. Pillar three covers regulations for logistics services and providers including transport operators and logistics service providers and users, mostly in the private sector. Pillar four on consolidating the institutional framework for interagency coordination and implementation arrangements involves government leaders, actors from the three other pillars, the private sector, and development partners. Policy levers are those that (i) improve interagency coordination, (ii) build institutional change management capacity, (iii) establish mechanisms for enhancing consultation and partnership with the private sector, (iv) develop clear roles, responsi­ bilities, and implementation arrangements of key stakeholders, (v) and develop practical results frameworks and regularly monitor reform performance. The report recommends that the National Trade Facilitation Committee (NTFC) should take the lead in coordinating trade, trade-related transport, and GVC policies by providing strategic direction and guidance, and supervising related multisectoral policies, particularly for delivering Policy Recommendation One. This committee was established according to the Prime Minister’s Decision Chapter 6 – Implementing trade-oriented connectivity and competitiveness policies 59 1899/QD-TTg dated April 10, 2016, chaired by Deputy Prime Minister Vuong Dinh Hue, with senior representatives from 20 line ministries, primarily to comply with the WTO’s Trade Facilitation Agreement (TFA). More important, this committee has been assumed to coordinate multiagency efforts to facilitate trade, reduce trade costs, and improve trade competitiveness. In response to the World Bank’s policy recommendations (Pham and Oh, 2018), the Prime Minister issued the Decision 684/QD-TTg dated June 4, 2019 to revise and supplement the Decision 1899/QD-TTg by adding the role to coordinate interagency efforts on national logistics development. This additional function makes the upgraded NTFC a perfect body to coordinate multisectoral policies on trade, traded-related transport and connectivity, and GVC development for trade competitiveness as proposed in the four-pillar framework and all recommendations in this report. We further suggest strengthening this mechanism by recommending the committee appoint an interagency taskforce to assist managing this assumed task. 6.3. Securing firm-level data for qualified multisectoral policy analyses on trade, transport, and value chains Relevant data sets should be in place with appropriate and regularly updated statistical indicators on value chains and gateways to ensure reliable analysis informs connectivity policy and investment in trade-related infrastructure. Much of the information and data needed for such analyses is missing and/or difficult to collect. This is partly due to a new approach that requires complex datasets and time for statistical systems to respond, but more important due to strict regulations for disclosing raw and firm-level data. We propose issuing relevant regulations to make firm-level trade and transport data available for comprehensive value-chain connectivity assessment and trade gateways analysis, as well as establishing an effective mechanism for better data collection, processing, and coordination at national and sectoral statistics levels among the General Statistics Office (GSO), the Ministry of Transport, the General Department of Customs, and others to supplement the data. Innovative methodologies using big data for real-time analysis should be explored toward this modern policy formulation process. The availability of such information and analysis can help address so-called information externalities and agglomeration effects (Krugman, 1991). Greater competitiveness and effectiveness require specialization in areas where an integrated presence of related and supporting activities can support an optimum productivity level that any individual company would find hard to achieve. The analysis would use considerable firm-level data to address research questions and inform key findings. Disaggregated data would be combined from various sources, including: (i) input-output data for value-chain link identification, (ii) enterprise data (per industry, per province, per commodity, per industrial park, per industry, etc.) for capturing regional concentration of domestic supply chains, (iii) transportation data and origin-destination (OD) flows (both within supply-chain structure and between cluster locations and trade gateways), and (iv) border/port trade data (land gateway, seaport, and airport) with the Harmonized System (HS) code of export and import volume. 60 Vietnam: Connecting value chains for trade competitiveness Input-output table data for value-chain links identification: Input-output (I/O) tables focus “on the interrelationships between industries in an economy with respect to the production and uses of their products and products imported from abroad.” (UN, 1999). In Vietnam, an I/O table is a model that reflects inter-sectoral relationships in the whole process of production and usage of products for final consumption, asset accumulation, and export of goods and services for the entire economy. In other words, an I/O table indicates how many products of other sectors are needed to produce a final product for an industry and vice versa. It allows researchers to analyze inter-sectoral relationships, evaluate production efficiency, and calculate indicators and indexes for macroeconomic management, economic analysis, and forecast. Vietnam’s I/O table 2016, developed by the GSO, is the sixth edition of the I/O table, covering 164 industries. This study uses I/O tables for sectoral, value-chain mapping based on backward links. Starting from sectors producing final products, first-tier sectors are defined as inputs purchased directly by the initial sectors. Repeatedly, second-, third- or lower-tier supplying sectors are computed. Once the main source industries are identified, we can diagram the value chains. Diagrams show the respective inputs (including their sector classification codes) and the direction of flows. Enterprise data for capturing regional concentration of domestic supply chains: In Vietnam, the GSO has conducted Enterprise Censuses every five years since 1995 and sample surveys every year between the censuses. The Enterprise Censuses of 2011 and 2016 are the two most recent. The censuses and surveys collect basic information on enterprise business activities, labor, operating results, investments, etc. We use Enterprise Census data to calculate the location quotient (LQ) index for each district and province, which allows us to analyze industrial agglomeration at district and provincial levels across the country, and to map potential industrial clusters and value-chains links from the I/O table. Transportation data and origin-destination (OD) flows: This model utilizes disaggregated commodity flow data, for the most freight intensive activities, to estimate existing freight flows across the network by mode (road, rail, inland waterway, and air) and OD information on key clusters with domestic supply-chain structure of commodities. In Vietnam, the MOT has conducted some OD studies, but they are not necessarily linked nor regularly conducted. Commodities included in these studies were not matched with VSIC system. The Enterprise Census questions should be revised to collect missing data and information on transportation OD flows to create a standard model that—even if it were improved—could be used by provincial and central-level planners and policy makers. The output would be standardized, irrespective of various inputs, and regularly updated. Customs data for gateway analysis: A key for value-chain analysis and mapping is identifying the location and concentration of trade gateways. Customs data from all gateways, collected and managed by the Vietnam Customs Office, provides this information, including the location and trade value of imported materials and exported commodities at international gateways at HS-8-digit level. Together with these two data sources mentioned above (I/O tables and Census Data), data on trade gateways helps capture all commodities flows and economic activities in a value chain to fully map its connective model – 4. Chapter 6 – Implementing trade-oriented connectivity and competitiveness policies 61 Data on origin-destination flows of goods from the digital traceability system: Vietnam is relatively more open to trade than other low-and-middle-income countries, with 14 FTAs signed to date. A goods’ origin traceability system is crucial for trade facilitation and FTAs effective use. Furthermore, such a system would help not only improve supply-chain performance, but also enable key stakeholders including government agencies and the private sector to collect a comprehensive and reliable value-chain dataset for connectivity and competitiveness analysis and make relevant business decisions. It will also allow policy makers to trace products origins to avoid fraudulent trade and ensure rules of origin for clean production and export. Sooner or later Vietnam should think about developing a digital traceability system for key value chains and could consider block- chain technology as a platform. Because this value-chain analysis is critical for businesses and the private sector, the report recommends building an information point with convenient access to publicly available information about value- chain links and spatial structure, including but not limited to geographic location and value-chain links, provincial specialization, international gateways’ statistics, etc. To be sustainable, such a multidisciplinary, cluster-development data center would require strong interagency coordination and a government-private sector partnership. Optimally, this center would be managed by a government agency, strongly motivated to use the data (which could oversee development master planning, competitiveness enhancement, and connectivity policy and investments). For coordinating data inputs, this agency should be mandated to work with the various sources of the previously mentioned data sets (GSO, customs, transport, other development partners, etc.) and empowered to manage data sharing with the private sector. Preferably, the center would be overseen by the trade facilitation and logistics policy coordination mechanism of the NTFC proposed in (Section 6.1). Regular updates and visualization of indicators on industrial concentration and commodities flows is useful for all stakeholders, policy makers, academics, researchers, and businesses alike. Box 6.1 provides a good example from the United States, in pooling a comprehensive dataset and organizing it into key statistical indicators for policy makers and the private sector to use. The report recommends sharing the information on provincial specialization for all concerned parties including central and local governments, the private sector, and development partners. The information on value-chain links, spatial structure, and connectivity is important not only for policy formulation, but also for the private sector to proactively participate in domestic supply chains as a significant part of GVCs. Such information is crucial to inform the domestic private sector for effective participation into GVCs. This is especially essential in Vietnam where more than 90 percent of the domestic private sector are small firms who lack this information and have weak links to foreign invested firms. The information on value-chain links, spatial structure, and connectivity could be made available on a cluster-mapping website following the U.S. model, with information collected and analyzed via the comprehensive value-chain connectivity assessment and trade gateways analysis, on big data in real 62 Vietnam: Connecting value chains for trade competitiveness time. It would be developed and shared publicly, for both policy makers and the private sector not only to implement Policy Recommendation One but also to fulfill Vietnam’s e-government initiative. Vietnam should consider developing a similar cluster-mapping project as the United States, with data sources and operational organizations properly ascribed. In addition to the cluster website, online freight-flow modeling can be developed based on OD flow data. The website and online freight-flow modeling would provide dynamic, visual information for governments and businesses to understand and shape the competitive landscape for a range of industries. The website would also help local governments understand local specialization and regional comparative advantages to promote strategic investment and lay the groundwork for new industries. Box 6.1. Visualization of cluster mapping in the United States The U.S. cluster mapping website, www.clustermapping.us, is a national initiative that provides open data on regional clusters and economies to support U.S. business, innovation, and policy. Users find interactive, robust data and tools to understand clusters and regional business environments, improve institutions, and locate appropriate partners across the country. Launched in September 2014, the website is a collaboration between the U.S. Commerce Department’s Economic Development Administration (EDA) and the Institute for Strategy and Competitiveness at Harvard Business School to generate practical and user-friendly cluster tools. Cluster mapping is designed to enable systematic comparison across regions. Strong clusters are those where their location quotient (the cluster’s relative specialization) puts them in the top 25% among U.S. regions in their respective category. The main underlying data source for generating benchmark cluster definitions is the U.S. Census Bureau’s County Business Patterns on employment, establishments, and wages by six-digit NAICS code (North American Industry Classification System). These data are collected at state and local level for economic areas, metropolitan and micropolitan statistical areas, and counties. The cluster data on the website is refreshed when new underlying industrial data becomes available, which is typically each year in June or July. The cluster-mapping website is valuable because it defines a set of standardized national clusters, which allow objective identification of a cluster’s regional competitiveness. It also enables comparisons and relative performance measurements between any regions in the United States, which are crucial for understanding and improving regional economic competitiveness. Cluster patterns across the country may reveal unique advantages, disadvantages, and opportunities in a region. Against these national standards, regions can dig deeper using local knowledge to identify their “region-specific clusters”. The mapping helps provide information on these region-specific clusters, as many of the initiatives focus on these more specific sectors. The U.S. cluster mapping website is used by governments, economic developers, and businesses to understand and shape the competitive landscape for a range of industries. Local officials are using the data to make strategic investments, recruit new companies, and lay the groundwork for new industries. Across the United States, cluster-mapping tools enable users to reinvent and modernize economic development strategies – all driven by open data. Source: http://www.clustermapping.us Chapter 6 – Implementing trade-oriented connectivity and competitiveness policies 63 Annex 1 Analysis of the textile and garment value chain A1.1. Industry overview FIGURE A1.1. Employment in textile and The textile and garment (T&G) industry has garment been a key Vietnamese industry for some decades, since the opening of the economy 1,800,000 26% Number of employees (person) Share in total mnf employees (%) 1,600,000 in the late 1980s. As a labor-intensive sector, 26% 1,400,000 it generated 25 percent of jobs in the 25% 1,200,000 manufacturing sector in 2016. As shown in 1,000,000 25% Figure A1.1, from 2010 to 2016, employment 800,000 24% 600,000 in T&G increased from one million people to 24% 400,000 1.6 million, thus on average, about a hundred 200,000 23% thousand new jobs were absorbed by the - 23% 2010 2011 2012 2013 2014 2015 2016 sector each year. Although the absolute Textile Garment Share number of employees increased, the sector’s share in total manufacturing employees Source: Statistical yearbook 2015, 2017. decreased slightly from 25.4 percent to 25.2 percent, implying faster employment growth FIGURE A1.2. Textile and garment exports in other manufacturing sectors. 35,000 T&G has been a major exporting sector for 30,000 decades. Figure A1.2 shows export value Export value (US$ mil.) 25,000 grew an average of 14 percent per year, 20,000 from US$ 8 billion in 2007 to US$ 31 billion in 15,000 10,000 2017. Clothing is a major export product of 5,000 the sector, with more than 80 percent share, - while exports of upstream products (yarns 2010 2011 2012 2013 2014 2015 2016 2017 and fabrics) contributed 11 and 4 percent Yarns Fabrics Garment respectively. Source: ITC Trademap, 2017. 64 Vietnam: Connecting value chains for trade competitiveness FIGURE A1.3. Decomposition of the T&G Figure A1.3 illustrates the decomposition of export gross exports in the sector. The foreign value- added component in Vietnam’s T&G export 100% increased slightly from 42 percent in 2005 80% to 45 percent in 2016. In the same period, 60% direct domestic value addition dropped from 40% 48 percent to 42 percent. This decrease was 20% compensated by an increase of 3 percent 0% in indirect domestic value addition, from 10 percent to 13 percent. Clearly, the T&G sector has not changed their structure in value Direct domestic Indirect domestic Reimported domestic Foreign addition in the last decade. However, even though the value addition shares did not Source: OECD, 2017. change dramatically, the level of domestic value addition grew substantially. FIGURE A1.4. Trade balance of up- and mid- The situation of the T&G sector is reaffirmed stream T&G segments by the trade balance of upstream segments. As shown in the Figure A1.4, Vietnam gained a 4000 trade surplus in yarns from 2008, but had trade 2000 deficits in raw materials and fabrics – inputs 0 and outputs of yarns. This is to be expected 20 1 20 2 20 3 20 4 20 5 20 6 20 7 20 8 20 9 20 0 20 1 20 2 20 3 20 4 20 5 20 6 17 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 -2000 20 given the country imports upstream inputs -4000 for downstream production and export. Trade -6000 balance data show Vietnam’s T&G sector relies Materials -8000 Yarns heavily on imported inputs. However, the -10000 Fabrics sector may find it difficult to satisfy the rules of Source: ITC Trademap, 2017. origin of the Comprehensive and Progressive Agreement for Trans-Pacific Partnership (CPTPP) and the European Union Vietnam Free Trade Agreement (EVFTA) and enjoy their tariff preferences. A1.2. Value-chain links The T&G value-chain structure obtained from the 2016 input-output (I/O) tables is shown in Figure A 1.5. The analysis applies a sourcing intensity of 2% as threshold. That is, those supplying sectors with sourcing intensities below 2% are not considered in the graph. In addition, the analysis excludes services inputs and activities (for example agricultural services) as well as capital goods (for example machinery) and also does not consider forward links to consuming sectors (for example wholesale/ retail trade). In 2016, the top four first-tier supplying sectors to wearing apparel were textiles (59.3%), other textiles (14.3%), wearing apparel (7.4%) and manufactured goods not elsewhere classified (n.e.c). (3.1%), Annex 1 – Analysis of the textile and garment value chain 65 representing 84 percent of total inputs to the sector. For the subsequent analysis the exact underlying percentages are not critical; the main goal is to identify the T&G value-chain links and structure, the location of actors involved in certain value-chain segments, and their connectivity. The main suppliers to the textile sector include inputs from the textiles sector itself (65.2 percent of all inputs supplied to the textiles sector), but also plastics and synthetic rubber in primary forms (10.6%), other chemical products, man-made fibers (5.4%), other nonperennial crops (3.9%) and leather and fur (2.5%). The largest supplying sectors to other textiles are almost identical and include the textiles sector (46.4 percent of all inputs to other textiles), followed by plastics and synthetic rubber in primary forms (11.3%), other textiles (10.2%), other chemical products, man-made fibers (8.3%), and plastic products (3%). FIGURE A1.5. T&G value-chain links Final Wearing apparel (IO 53) product Manufactured goods Textiles (IO 51) Other textiles (IO 52) Wearing apparel (IO 53) n.e.c. (IO 97) First tier SI = 59.3% SI = 14.3% SI = 7.4% SI = 3.1% Textiles (IO 51) Textiles (IO 51) Second tier SI = 65.2% SI = 46.4% Plastic and synthenic rubber Plastic and synthenic rubber in primary forms (IO 64) in primary forms (IO 64) SI = 10.6% SI = 11.3% Other chemical products, Other textiles (IO 52) man-made bres (IO 66) SI = 10.2% SI = 5.4% Other non-perennial crops Other chemical products, (IO 8) man-made bres (IO 66) SI = 3.9% SI = 8.3% Tanned and dressed leather, dressed and dyed fur, and Plastic products (IO 69) related products (IO 54) SI = 3.0% SI = 2.5% Source: I/O Table 2016, authors. Comparing the value-chain structure against existing value-chain maps confirms the most relevant inputs to apparel production are natural and synthetic fibers as well as yarn and fabrics. Industry experts suggest the refined T&G value-chain links depicted in Figure A1.6. The T&G value chain consists of five segments: (i) Producing materials (growing fiber crops and plastics and synthetic 66 Vietnam: Connecting value chains for trade competitiveness rubber in primary form), (ii) Producing yarns (spinning), (iii) Producing fabric (weaving, knitting, finishing), (iv) Garments (designing, cutting, sewing, buttonholing, ironing, packing), and (v) Other garments (designing, cutting, sewing, buttonholing, ironing, packing). FIGURE A1.6. T&G value-chain segments Key segments T&G MATERIALS TEXTILE GARMENTS Designing Cutting Sewing Clothing Weaving Buttonholing (VSIC1410, 1430) Growing of ber Spinning Knitting Ironing crops (VSIC01160) Fabric Packing Yarn Finishing (VSIC13120, (VSIC13110, 13130, Plastic and synthetic 20300) 13210) Other garments rubber in primary form (VSIC13920, 13930, (VSIC2013) 13940, 13990) Source: Authors. A1.3. Spatial structure and value-chain mapping This section identifies geographic locations of clusters linked to the five segments of the T&G value chain. Scatterplots show the location quotients (LQs) by province for each segment. Figure A1.7 shows the spatial structure of yarn production, the second segment of the T&G value chain. The provinces above the horizontal y = 1 line are those for which the LQ exceeds 1, in other words whose relative specialization in yarn production is greater than the national average, based on employment data. Those provinces in the upper-right quadrant additionally increased their LQ between 2011 and 2016, which can be interpreted as stronger relative specialization in yarn spinning. The circle size denotes actual employment in the province in 2016. Specialization patterns by province suggest Tay Ninh has the highest relative specialization in yarn production and also strengthened its specialization over time. Thua Thien-Hue has the second largest relative specialization by LQ, but their specialization declined from 2011. Dong Nai has the third largest relative specialization and employs many more people in yarn production. However, its relative specialization declined over the previous five years. Quang Ninh and Long An are also important provinces in their specialization in yarn production, but only the former became more specialized between 2011 and 2016. Annex 1 – Analysis of the textile and garment value chain 67 FIGURE A1.7. Locational distribution of the yarn segment Thua Thien Hue, 6.4 Dong Nai, 4.7 Long An, 4.0 Tay Ninh, 7.3 8 Ha Nam, 3.0 Nam Dinh, 2.7 Ba Ria - Vung Tau, 2.3 Quang Ninh, 4.4 LQ2016 5 Thai Binh, 3.1 Ben Tre, 1.5 Lam Dong, 2.2 Quang Ngai, 1.9 2 Khanh Hoa, 2.1 -4 -1 2 5 8 Phu Tho, 1.1 -1 Circle size = number of employee 2016, Changing in LQ between 2011 - 2016 Min Ben Tre = 700; Max Dong Nai = 25 800 Source: Enterprise Census 2011 and 2016, calculation by authors. Shifting to the fabric segment, Figure A1.8 shows the growing importance of Tay Ninh not only in yarn production, but also in weaving, knitting, and finishing fabrics. The province had the highest LQ and greatest expansion of the segment since 2011. Other provinces specialized in yarn spinning also specialized in fabric weaving, knitting, and finishing, including Nam Dinh, Long An, Phu Tho, Dong Nai, and Lam Dong. Binh Puoc and Long An both had a relatively high specialization in 2016 and a strong increase since 2011 (upper right quadrant). And, while Ho Chi Minh City had a low and relatively constant LQ, it employed the largest number of workers in the fabric segment. Other provinces reduced their relative specialization in fabric production, in particular Nam Dinh, Phu Tho, Ha Nam, and Binh Duong, among others (upper left quadrant). FIGURE A1.8. Locational distribution of the fabric segment 8 Nam Dinh, 5.6 Tay Ninh, 6.7 Phu Tho, 4.3 6 Binh Phuoc, 4.5 Long An, 4.0 4 Lam Dong, 2.4 Dong Nai, 2.5 LQ2016 Binh Duong, 1.6 Ho Chi Minh City, 1.2 2 Ninh Binh, 1.1 -6 -3 0 3 6 Ha Nam, 1.8 0 Ha Noi, 0.4 Circle size = number of employee 2016, Changing in LQ between 2016 - 2011 -2 Min Lam Dong = 700, Max HCMC = 17.850 Source: Enterprise Census 2011 and 2016, calculation by authors. 68 Vietnam: Connecting value chains for trade competitiveness The clothing segment involves a wider range of activities, from designing, to cutting, sewing, buttonholing, ironing, and packing, reflected by more provinces specialized in this segment (Figure A1.9). This segment also employed the largest number of workers in the T&G value chain. Scatterplots show several interesting trends. First, the provinces that employed a larger absolute number of workers in the clothing segment (like Binh Duong and Dong Nai) and provinces with larger LQs (like Hai Duong, Thai Binh, Bac Giang, and Hung Yen, among others) had a lower relative specialization over time (upper left quadrant). Second, several provinces increased in relative specialization in clothing (upper right quadrant), including Thua Thien Hue, Tien Giang, Quang Nam, and Thanh Hoa, which employed the greatest number of clothing workers among these provinces. FIGURE A1.9. Locational distribution of the clothing segment Phu Tho, 2.2 5 Nam Dinh, 3.1 Quang Nam, 2.1 Hai Duong, 2.7 Ben Tre, 2.2 Thai Binh, 3.2 Tien Giang, 2.3 LQ2016 Bac Giang, 3.2 Thua Thien Hue, 2.8 3 Thanh Hoa, 1.9 Hung Yen, 2.4 Hoa Binh, 1.9 Tay Ninh, 1.8 Tuyen Quang, 1.7 Ha Nam, 1.6 Binh Thuan, 1.4 1 -2.0 0.0 Phu Yen, 1.1 2.0 Vinh Phuc, 1.2 Vinh Long, 1.2 Binh Duong, 1.3 Ninh Thuan, 1.0 Ninh Binh, 1.3 Dong Nai, 1.0 -1 Circle size = number of employee 2016, Changing in LQ between 2016 - 2011 Min Ninh Thuan= 2500, Max Binh Duong = 140 700 Source: Enterprise Census 2011 and 2016, calculation by authors. Figure A1.10 outlines Thai Binh and Long An were dynamic provinces with increased specialization in other garments activities, employing many workers in this segment. Additionally, provinces like Quang Nam, Ben Tre, Ba Ria–Vung Tau, Can Tho, and Binh Dinh increased specialization in this segment but employed fewer workers (upper-right quadrant). Among the provinces with declined specialization (upper-left quadrant) we find again Binh Duong and Dong Nai, which employed much of the workforce in the segment. The declined specialization in these two provinces follows the trend in other segments of the T&G value chain. Annex 1 – Analysis of the textile and garment value chain 69 FIGURE A1.10. Locational distribution of the other garment segment 16 Thai Binh, 13.4 Binh Duong, 1.7 12 Dong Nai, 1.1 Long An, 5.2 LQ2016 Hai Phong, 1.1 8 Ben Tre, 3.3 Tay Ninh, 1.4 Tien Giang, 1.3 Nam Dinh, 2.2 4 Can Tho, 1.5 Ninh Thuan, 4.6 VinhPhuc, 1.3 Quang Nam, 2.8 Ba Ria - Vung Tau, 1.1 0 -3 -1 1 3 5 7 Binh Dinh, 1.2 -4 Circle size = number of employee 2016, Changing in LQ between 2016 - 2011 Min Ninh Thuan = 900; Max Thai Binh = 19500 Source: Enterprise Census 2011 and 2016, calculation by authors. Map A1.1 summarizes the provincial specialization patterns depicted in the previous scatterplots. The five segments of the T&G value chain are shown in different colors: producing materials in green, yarn in turquoise, fabric in bright green, clothing in dark blue, and other garments in red. Darker shades indicate a higher LQ or relative specialization, while lighter shades suggest a lower provincial specialization. 70 Vietnam: Connecting value chains for trade competitiveness MAP A1.1. Geographic distribution of the T&G value chain Disclaimer: The boundaries, colors, denominations and other information shown on any map in this work do not imply any judgement on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. LQ Other garments 1.2 - 2.0 2.1 - 5.0 5.1 - 10.0 10.1 - 13.4 LQ Clothing 1.1 - 2.0 2.1 - 3.2 LQ Fabric 1.2 - 2.0 2.1 - 5.0 5.1 - 5.6 LQ Yarns 1.9 - 2.0 2.1 - 5.0 5.1 - 7.3 Top 10 Textile firms (income) 1,000,000 10,000,000 Source: I/O Table 2016, Enterprise Census 2011 and 2016, calculation by authors. Annex 1 – Analysis of the textile and garment value chain 71 A1.4. Value-chain-based connectivity and key corridors Map A1.2 highlights connective propensity of the T&G value chain. Each segment in the value chain uses different transportation corridors. - From materials to yarns: NR22, NR22B, NR1 (Hue – HCMC), NR51, NR56, NR20, NR24B, Phap Van – Cau Gie expressway, NR5, NR18, NR4B, NR1 (Lang Son – Ha Noi) - From yarns to fabrics: NR22B, NR13, NR14, NR28, NR51, NR56, NR20, NR27, NR1 (Khanh Hoa – Ninh Thuan, Quang Tri – Quang Ngai), NR24B, NR9, NR18, Noi Bai – Lao Cai expressway (to Phu Tho), NR32, Phap Van – Cau Gie – Ninh Binh expressway (to Ha Nam), NR21A (Ha Nam – Nam Dinh) - From fabrics to apparels: HCMC - Trung Luong expressway, NR1 (HCMC – Vinh Long, Lang Son – Quang Ngai), NR62, NRN2, NR22, NR13, NR14, AH17, NR26, NR19C, NR20, NR55, NR51, NR56, HCM – Long Thanh – Dau Giay expressway, NR9, NR12B, NR45 (Thanh Hoa), NR217, Phap Van – Cau Gie – Ninh Binh expressway (Ha Noi - Ha Nam), NR21A (Ha Nam – Nam Dinh), NR38, NR39, Ha Noi – Bac Giang expressway, NR2 (Phu Tho – Tuyen Quang), Noi Bai – Lao Cai expressway (to Phu Tho), NR32 - For apparel exports: NR2 (Phu Tho – Tuyen Quang), Noi Bai – Lao Cai expressway (Phu Tho), NR1 (Lang Son – Phu Yen, Binh Thuan – Vinh Long), NR18, AH14, NR21A (Ha Nam – Nam Dinh), Phap Van – Cau Gie – Ninh Binh expressway (Ha Noi - Ha Nam), Ha Noi – Bac Giang expressway, NR47, NR15, HCM road (Hoa Binh - Thanh Hoa), AH13, NR19C, HCMC - Trung Luong expressway, HCM – Long Thanh – Dau Giay expressway, NR51, NR56, NR22 72 Vietnam: Connecting value chains for trade competitiveness MAP A1.2. Connective propensity of the T&G value chain LQ Other garments 1.1 - 2.0 2.1 - 5.0 5.1 - 10.0 10.1 - 13.4 LQ Clothing 1.1 - 2.0 2.1 - 3.2 LQ Fabric 1.2 - 2.0 2.1 - 5.0 5.1 - 5.6 LQ Yarns 1.9 - 2.0 2.1 - 5.0 5.1 - 7.3 Top 10 Textile firms (income) 1,000,000 10,000,000 Disclaimer: The boundaries, colors, denominations and other information shown on Airport any map in this work do not imply any judgement Border gate on the part of The World Bank concerning the legal Seaport status of any territory or the endorsement or acceptance of such Clothing(2IV) - Export boundaries. OtherGarments(2V) - Clothing(2IV) Fabric(2III) - Clothing(2IV) Yarns(2II) - Fabric(2III) Materials(2I) - Yarns(2II) Source: I/O Table 2016, Enterprise Census 2011 and 2016, Customs data, calculation by authors. Annex 1 – Analysis of the textile and garment value chain 73 Annex 2 Analysis of the leather and footwear value chain A2.1. Industry overview FIGURE A2.1. Employment in leather and In addition to the textile and garment sector, footwear leather and footwear is another labor- intensive and export-oriented Vietnamese 1,400,000 19% sector. Figure A2. 1 shows from 2010 to 2016, Number of employees (person) 1,200,000 employment in the sector increased from 18% 1,000,000 700 thousand to 1.2 million people, and its 800,000 share in total manufacturing employees 17% 600,000 grew from 16 to 18.5 percent. From 2014 400,000 16% onward, leather and footwear employment 200,000 increased in absolute value but its share in - 15% total manufacturing employment decreased 2010 2011 2012 2013 2014 2015 2016 modestly from 18.4 to 17.9 percent. This Number of employees Share in total mnf employees suggests there has been faster employment growth in other sectors. Source: Statistical yearbook 2015, 2017. FIGURE A2.2. Leather and footwear exports Figure A2.2 shows export of leather and footwear products grew 17 percent between 20,000 2010 and 2017, from US$ 5.5 to 18 billion. 18,000 Footwear is a major export, accounting for Export value (mil. USD) 16,000 14,000 more than 80 percent of the sector export 12,000 value, followed by handbags with a share of 10,000 8,000 17 percent, and leather products with just 6,000 0.1 percent. 4,000 2,000 - 2010 2011 2012 2013 2014 2015 2016 2017 Leather Handbags Footwear Source: ITC Trademap, 2017. 74 Vietnam: Connecting value chains for trade competitiveness The export structure shows Vietnam’s leather and footwear sector seemed to focus heavily on final products but not upstream products. This can be confirmed by the sector’s import data. In 2017, the sector imported more than US$ 700 million. The top three imported products were parts of footwear (uppers and parts thereof, and outer soles and heels), and raw leather, which accounted for more than 60 percent of the sector’s total import value. A2.2. Value-chain links Inter-sectoral links have been identified from the I/O table 2016 for the leather and footwear sector and graphed in Figure A2.3. The links show footwear in Vietnam is mainly made from leather and textile materials, accounting for 34 and 29 percent respectively. Compared to 2012, the share of textile materials increased slightly from 25 percent to 29 percent, implying a trend toward more textile footwear. It is impossible to disaggregate the data to the sectoral level for precise textiles used in the garment or footwear segments. FIGURE A2.3. Leather and footwear value-chain links Final Footwear (IO 55) product Tanned and dressed Textiles Other Rubber Other chemical Footwear Paper and First leather, dressed and (IO 51) textiles products products, man- (IO 55) paper tier dyed fur, and related SI = 16.0% (IO 52) (IO 68) made bers SI = 4.9% products products (IO 54) SI = 13.0% SI = 6.3% (IO 66) (IO 57) SI = 34.7% SI = 5.7% SI = 2.5% Tanned and dressed Fabricated metal Textiles Other Plastic products Plastic and synthetic Second leather, dressed and products, except (IO 51) textiles (IO 69) rubber in primary tier dyed fur, and related machinary an SI = 7.7% (IO 52) SI = 3.4% forms (IO 64) products (IO 54) equipment (IO 76) SI = 4.8% SI = 2.1% SI = 52.4% SI = 9.6% Source: I/O table 2016, authors. Key segments in the leather and footwear sector are refined in Figure A2.4. Other related materials are still included in the figure, but only key segments were considered in analyzing the value-chain links, including leather (VSIC 15110), leather products (VSIC 14200), handbags (VSIC 15120), and leather footwear (VSIC15200). Annex 2 – Analysis of the leather and footwear value chain 75 FIGURE A2.4. Leather and footwear value-chain segments PRODUCTS MADE MATERIALS FROM LEATHER Animals Raw Leather Leather products for skins animals skins (VSIC15110) (VSIC14200) Textiles Luggage, handags (VSIC15120) Fabricated metal products Footwear Packaging (VSIC15200) Source: Authors. A2.3. Spatial structure and value-chain mapping Consistent with the findings that an upstream leather and footwear sector has not developed in Vietnam, Figure A2.5 shows very few provinces had a high specialization of leather production, barring Ba Ria - Vung Tau, Binh Duong, Dong Nai, and Tay Ninh. Another province, Lang Son, also appears in the map, but their employment was not significant despite having a high LQ which increased between 2012 and 2016. This may imply that leather production activity has emerged in Lang Son recently. FIGURE A2.5. Locational distribution of the leather segment 32 24 Dong Nai, 4.1 Tay Ninh, 14.3 16 LQ2016 8 Ba Ria - Vung Tau, 23.7 Binh Duong, 2.3 Lang Son, 8.4 0 -15 -10 -5 0 5 10 15 -8 Circle size = number of employee 2016, Changing in LQ between 2016 - 2011 Min Lang Son =130, Max Ba Ria Vung Tau = 2900 Source: Enterprise Census 2011 and 2016, calculation by authors. 76 Vietnam: Connecting value chains for trade competitiveness Figure A2.6 shows provinces with a high production concentration of handbags and other leather products. Ho Chi Minh City (HCMC), Binh Duong, and Dong Nai provinces had high employment but low LQs, which decreased between 2012 and 2016. This implies the sector structure in these provinces has been shifting to other segments. In contrast, the figure shows the segment is emerging in other provinces with a high LQ, and positive change between 2012 and 2016, including Long An, Ben Tre, Tien Giang, and Tra Vinh. Ba Ria - Vung Tau, Nam Dinh, Vinh Long, and An Giang provinces had no significant concentration, with low LQs and low employment. FIGURE A2.6. Locational distribution of the handbags and other leather segments 16 Tien Giang, 9.6 Ben Tre, 11.0 12 Long An, 7.9 Dong Nai, 1.6 LQ2016 8 Ba Ria - Vung Tau, 1.8 Binh Duong, 2.2 Tra Vinh, 7.7 An Giang, 1.3 4 Tay Ninh, 1.7 Nam Dinh, 2.8 -4 -1 0 2 5 8 Ho Chi Minh city, 1.0 Vinh Long, 2.7 -4 Changing in LQ between 2016 - 2011 Circle size = number of employee 2016, Min An Giang=900, Max HCMC = 31300 Source: Enterprise Census 2011 and 2016, calculation by authors. There were more provinces with a high concentration of footwear production activity as shown in Figure A2.7. Tra Vinh, Vinh Long, Tay Ninh, Thanh Hoa, Hau Giang, and Binh Phuoc provinces had high LQs that increased between 2012 and 2016. Binh Duong and Dong Nai are the largest hubs but decreased in LQ between 2012 and 2016. Other provinces with a negative change in LQ were Hai Phong, Hai Duong, Quang Nam, and Long An. Annex 2 – Analysis of the leather and footwear value chain 77 FIGURE A2.7. Locational distribution of the footwear segment Binh Duong, 2.0 9 Tra Vinh, 6.1 Vinh Long, 4.6 Dong Nai, 3.4 Tay Ninh, 4.2 6 Hai Phong, 2.1 Thanh Hoa, 2.9 Hau Giang, 3.1 Long An, 2.7 3 Binh Phuoc, 2.6 Quang Nam, 1.6 An Giang, 1.1 -2 0 2 4 Hai Duong, 1.1 0 Tien Giang, 2.4 -3 Changing in LQ between 2016 - 2011 Circle size = number of employee 2016, Min An Giang = 5600, Max Dong Nai = 207,000 Source: Enterprise Census 2011 and 2016, calculation by authors. Combining findings from the three figures above, it appears all segments of the value chain have existed in Binh Duong and Dong Nai, which are considered the hubs of leather and footwear in Vietnam. However, decreases in LQ between 2012 and 2016 in all segments in these provinces imply these provinces might be downsizing in this sector or seeing faster growth of other sectors. The data also indicate northern provinces are not competitive in leather and footwear, with few provinces appearing in the figures, and none having a significant LQ index. Map A2.1 demonstrates the provincial specialization patterns of the leather and footwear value chain. The three segments are shown in different colors: producing leather in green, handbags and other leather products in turquoise, and footwear in bright green. Darker shades denote a higher LQ or relative specialization, while lighter shades indicate a lower specialization. Red circles denote the largest final production firms in the sector. There is obvious leather and footwear specialization in the South, where all segments of the value chain exist, and the top five largest firms are located. In the North, materials are mainly imported from China through Huu Nghi (Lang Son province) and Mong Cai (Quang Ninh province) gateways, while footwear with upper leather (HS6403) and textile (HS6404) are mainly exported through the Dinh Vu seaport (Hai Phong province). Handbags and other leather products are exported equally via air (Noi Bai Airport) and sea (Dinh Vu). Trade data from the Central Region are not significant. Major products are footwear exported through Tien Sa seaport in Da Nang. Trade data from the South indicate the region is the most dynamic area of the sector. Besides footwear, handbags and other leather products are also major exports of the region. Most of these products are imported and exported via Cat Lai seaport, some through Cai Mep seaport, and very few via Dong Nai port. Dong Nai and Binh Duong are the hubs of the leather and footwear industry but import and export mainly through ports in Ho Chi Minh City, requiring well-organized logistics and transportation between Dong Nai and Binh Duong to the ports in Ho Chi Minh City. 78 Vietnam: Connecting value chains for trade competitiveness MAP A2.1. Geographic distribution of the leather and footwear value chain Disclaimer: The boundaries, colors, denominations and other information shown on any map in this work do not imply any judgement on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. LQ Footwear 1.1 - 2.0 2.1 - 4.6 LQ Cases&bags 1.0 - 2.0 2.1 - 5.0 5.1 - 10.0 10.1 - 11.0 LQ Leather 2.3 - 5.0 5.1 - 10.0 10.1 - 23.7 Top 10 Leather firms (income) 20,000,000 Source: I/O Table 2016, Enterprise Census 2011 and 2016, Customs, and calculation by authors. Annex 2 – Analysis of the leather and footwear value chain 79 A2.4. Value-chain-based connectivity and key corridors Map A.2.2 demonstrates the main transportation corridors of the leather and footwear value chain. Each segment uses different corridors as follows: - From leather to cases and bags: NR91, NR1 (Dong Nai – Can Tho, Lang Son – Ha Noi), HCMC - Trung Luong – My Thuan expressway, NR60, NR54, NRN2, NR22, NR22B, NR13, NR51, NR56, Phap Van – Cau Gie – Ninh Binh expressway (to Ha Nam), NR21A (Nam Dinh), NR18, NR5 - From leather to footwear: NR1 (Dong Nai – Hau Giang, Lang Son – Thanh Hoa)), HCMC - Trung Luong – My Thuan expressway, NR61, NR60, NR54, NRN2, NR22, NR22B, NR13, NR51, NR56, QL14 (Binh Phuoc), HCM Road (Hoa Binh - Thanh Hoa), NR18, NR5 - For cases and bags exports: NR91, NR1 (HCMC – Can Tho, Lang Son – Da Nang), NR51, HCMC - Trung Luong – My Thuan expressway, NR60, NR54, NRN2, NR22, HCM – Long Thanh expressway, Phap Van – Cau Gie – Ninh Binh expressway, NR21A (Ha Nam – Nam Dinh), NR18, AH14, NR10 - For footwear exports: NR91, NR1 (Dong Nai – Can Tho, Quang Tri – Quang Nam, Lang Son – Ha Noi), HCMC - Trung Luong – My Thuan expressway, NR61, NR60, NR54, NRN2, NR22, NR22B, NR13, HCM – Long Thanh expressway, NR14E, NR9, Noi Bai – Lao Cai expressway, NR47, HCM road (Hoa Binh - Thanh Hoa), NR10, AH14, NR18 (Bac Ninh), NR37 (Hai Duong) 80 Vietnam: Connecting value chains for trade competitiveness MAP A2.2. Connective propensity of the leather and footwear value chain LQ Footwear 1.1 - 2.0 2.1 - 4.6 LQ Cases&bags 1.0 - 2.0 2.1 - 5.0 5.1 - 10.0 10.1 - 11.0 LQ Leather 2.3 - 5.0 5.1 - 10.0 10.1 - 23.7 Top 10 Leather firms (income) Disclaimer: The boundaries, colors, 20,000,000 denominations and other information shown on any map in this work do Airport not imply any judgement on the part of The World Bank concerning the legal Border gate status of any territory or the endorsement Seaport or acceptance of such boundaries. Footwear(3III) - Export Cases&bags(3II) - Export Leather(3I) - Footwear(3III) Leather(3I) - Cases&bags(3II) Source: I/O Table 2016, Enterprise Census 2011 and 2016, Customs data, calculation by authors. Annex 2 – Analysis of the leather and footwear value chain 81 Annex 3 Analysis of the electronics value chain A3.1. Industry overview FIGURE A3.1. Employment in the electronics The electronics sector emerged in Vietnam sector in the last decade. The number of enterprises in this sector doubled from 613 in 2010 to 700,000 8% 1399 in 2016, which supported jobs for 160 Number of employees (person) Share in mnf employees (%) 600,000 7% thousand employees in 2010 and more than 6% 500,000 600 thousand in 2016. Figure A3.1 shows a 5% 400,000 sharp increase in electronics employment and 4% 300,000 3% its share in the manufacturing sector. 200,000 2% 100,000 1% Figure A3.2 shows electronics exports grew - 0% 2010 2011 2012 2013 2014 2015 2016 drastically since 2010, from US$ 6 billion to US$ Number of employees Share 70 billion in 2017, jumping from the second to the first largest export sector, leaving others Source: INDSTAT, 2016. far behind. More than half of these exports FIGURE A3.2. Electronics exports compared were telephone sets. The electronics export to other sectors growth rate between 2010 and 2017 was 40 percent, while the growth rates of T&G and 80,000 footwear were 13 and 16 percent respectively. 60,000 Figure A3.3 shows the decomposition of gross 40,000 exports in the electronics sector into two 20,000 portions: the foreign value addition (FVA), - and the domestic value addition (DVA). The 2010 2011 2012 2013 2014 2015 2016 2017 DVA consists of three components: (i) the Electronics T&G Footwear Furniture Fish Fruit direct value-added (DDC) contribution within Co ee the electronics sector, (ii) the indirect (IDC) contribution of upstream sectors supplying Source: ITC Trademap, 2017. the electronics sector, and (iii) reimported intermediates (RIM). As illustrated in the figure, Vietnam’s electronics export increasingly 82 Vietnam: Connecting value chains for trade competitiveness FIGURE A3.3. Decomposition of the relied on the FVA. Its share fluctuated around electronics gross export 50 percent from 2005 to 2010, then gradually increased to 63 percent in 2016. As a result, 100% DVA diminished its contribution in the value 80% chain, with direct value addition dropping 60% from 35 percent in 2005 to 28 percent in 2016, 40% and indirect value addition from 18 percent 20% to 9 percent in the same period. However, 0% in GVCs, countries import to export and by 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 doing so, they grow their exports, as well as EXGR_DDC EXGR_IDC EXGR_RIM EXGR_FVA the DVA embodied in them. It is implausible Source: OECD, 2017. that Vietnam would have been able to grow its gross exports so dramatically if the sector did not import foreign parts and components. A3.2. Value-chain links The electronics value chain is graphed in Figure A3.4 based on inter-sectoral links obtained from the I/O table 2016. The links show different value-chain tiers from up, middle, and downstream, with lower tiers providing inputs to produce goods in upper tiers. FIGURE A3.4. Electronics value-chain links Final Computer, electronic products (IO 77) products Computer, electronic Fabricated metal Other electrical Wiring and wiring Manufactured First products (IO 77) products etc. (IO 76) equipment (IO 86) devices (IO 83) goods n.e.c (IO 97) tier SI = 33.9% SI = 11.9% SI = 11.7% SI = 6.4% SI = 5.9% Products of iron and steel Colour and precious metals, Second (IO 74) casting services of metals (IO 75) tier SI = 40.3% SI = 32.6% Fabricated metal Wiring and wiring devices (IO 83) products etc. (IO 76) SI = 31.3% SI = 26.1% Fabricated metal products, Colour and precious metals, ect. (IO 76) casting services of metal (IO 75) SI = 6.7% SI = 8.6% Plastic and synthetic rubber Other chemical products, in primary forms (IO 64) man-made bres (IO 66) SI = 4.0% SI = 3.0% Rubber products (IO 68) Other electrical equpiment SI = 3.8% (IO 86) SI = 2.9% Products of iron and steel (IO 74) Basic chemicals (IO 62) SI = 3.7% SI = 2.0% Other electrical equipment (IO 86) SI = 2.3% Source: I/O Table 2016, authors. Annex 3 – Analysis of the electronics value chain 83 Since sectors in the I/O Table are more aggregated than the 4- and 5-digit VSIC codes, the links in Figure A3.4 show the main suppliers to the electronic sector include inputs from the electronic sector itself (more than one third). Therefore, to reflect the reality of the sector’s value chain, the links in the initial graph are refined in the Figure A3.5. into three segments. The upstream segment provides inputs (materials – metals, plastics, rubbers, chemicals, etc.) to produce intermediate goods (electronic and electric components and subassemblies) in the middle segment, which are then used to assemble final products downstream (computers, communications, consumer electronic products (3C) electronics). Each segment is associated with corresponding VSIC codes provided in the figure. Electronic and electric (E&E) components and subassemblies belong to the same 5-digit group, and some specific parts and components are grouped with the final products, and are only broken down at product levels (6 and 7-digit groups). Therefore, at the VSIC 5-digit level, the electronics value chain can only be divided into two segments: E&E components and final products. However, when discussing trading of electronic products, SITC and HS systems allow separating into three segments: components, subassemblies, and final products. FIGURE A3.5. Electronics value-chain segments Indirect segments Key segments MATERIALS COMPONENT SUBASSEMBLIES FINAL PRODUCTS Integrated Consumer electronics Semiconductor circuits (VSIC2640) Metal (VSIC24100, wafer Active 24200, 24310) discrete Product speci c Communications PCBA parts (VSIC2630) Electronic PCB (VSIC2610, Plastics Enclosure/ Computer/ 2680) Passives Display (VSIC20132) housing Storage/O ce (VSIC2620) Chemicals Cables (VSIC20110, 20221) Electrical (VSIC27330) Batteries Source: I/O Table 2016, authors. A3.3. Spatial structure and value-chain mapping Provinces that had high specialization of electronic component manufacturing appear in Figure A3.6. The Y axis represents a province’s LQ in 2016, and the X axis shows the change in LQ between 2011 and 2016. The circle size represents the number of employees in this segment in a province. Provinces in the upper-right quadrant increased participation in the electronics value chain, with high agglomeration and increased LQs between 2011 and2016. They include Vinh Phuc, Hoa Binh, Hai Duong, Hung Yen, Ha Nam, Phu Tho, and Quang Nam. Provinces in the upper left had decreased LQs. Their concentration was still higher than the national average but less agglomerative than in 2011. They are Bac Giang, Bac Ninh, and Binh Duong. Each had more than 10 thousand employees in the segment, therefore the decrease in LQ may not signify shrinkage in electronic component 84 Vietnam: Connecting value chains for trade competitiveness manufacturing but rather an expansion of other economic activities in these provinces, or because other provinces are emerging in this activity. FIGURE A3.6. Locational distribution of the electronics components segment 20 Bac Ninh, 3.9 Hai Duong, 3.4 Vinh Phuc, 16.5 15 Binh Duong, 1.8 Hung Yen, 3.0 10 Bac Giang, 64 Ha Nam, 1.8 Hoa Binh, 10.3 LQ2016 5 Phu Tho, 1.4 0 -15 -10 -5 0 5 10 15 20 Quang Nam, 1.5 -5 Circle size = number of emplyee 2016, Changing in LQ between 2016 - 2011 Min Phu Tho= 2600, Max Vinh Phuc = 35200 Source: Enterprise Census 2011 and 2016, calculation by authors. Figure A3.7 shows provinces that had high concentration of final electronic product manufacturing. Thai Nguyen was the most dynamic province with the second highest employment in the segment, second highest LQ in 2016, and the largest increase in LQ between 2011 and 2016. Bac Ninh province had the highest LQ in 2016 and the highest employment, but its LQ in 2016 was lower than 2011. The changes in these two provinces are linked closely with Samsung’s investment movement. Other provinces in the upper-right quadrant are Bac Giang, Ha Nam, Ninh Binh, and Quang Ngai, and in the upper left, Da Nang and Hai Duong. FIGURE A3.7. Locational distribution of the final 3c products segment Bac Ninh, 15.0 Thai Nguyen, 14.8 18 Bac Giang, 7.9 13 Da Nang, 1.2 Quang Ngai, 1.3 LQ2016 8 Ha Nam, 2.7 Hai Duong, 1.3 3 Ninh Binh, 1.1 -12 -7 -2 3 8 13 -2 Circle size = number of employee 2016, Changing in LQ between 2016 - 2011 Min Quang Ngai = 2500, Max Bac Ninh = 144000 Source: Enterprise Census 2011 and 2016, calculation by authors. Annex 3 – Analysis of the electronics value chain 85 From Figure A3.6 and Figure A3.7, we can see there was a specialization trend in electronics sector in northern provinces. It is not surprising to see the high LQs in Bac Ninh, Bac Giang, Vinh Phuc and Thai Nguyen, due to the presence of Samsung and its suppliers, while in other provinces, high LQs likely resulted from medium and large multinational, electronics companies like Brother in Hai Duong and Mabuchi Motor in Da Nang. Some provinces hosting giant electronic companies, for example Ha Noi with Panasonic and Canon, Hai Phong with LG, and Ho Chi Minh City with Intel and Samsung, do not appear in either figure. This may be because these companies hired fewer employees compared to other companies with similar revenues. The geographic distribution of the electronics value chain is summarized in Map 3.1, confirming the electronics sector is highly concentrated in the northern provinces. The darker the color, the higher the LQ, and the greater the provincial agglomeration. Among the ten biggest companies in terms of revenue, most are located in the North. Different colors represent different segments in the value chain thus we can see both segments are highly concentrated in the North and can assume the electronics value chain has emerged in the northern provinces. 86 Vietnam: Connecting value chains for trade competitiveness MAP A3.1. Geographic distribution of the electronics value chain Disclaimer: The boundaries, colors, denominations and other information shown on any map in this work do not imply any judgement on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. LQ Final Product 1.1 - 2.0 2.1 - 5.0 5.1 - 10.0 10.1 - 15.0 LQ EE Components 1.4 - 2.0 2.1 - 5.0 5.1 - 10.0 10.1 - 16.5 Top 10 Electronic firms (income) 10,000,000 100,000,000 Source: I/O Table 2016, Enterprise Census 2011 and 2016, Customs, and calculation by authors. Annex 3 – Analysis of the electronics value chain 87 A3.4. Value-chain-based connectivity and key corridors Map A3.2 demonstrates the electronics value-chain connective propensity. Pie-charts demonstrate the trade value of electronics components and final products, showing the dominance of air transportation with 57 percent imported through Noi Bai airport, 21 percent through Tan Son Nhat airport, and only 8.3 and 7.6 percent through Hai Phong seaport and HCMC seaport, respectively. Most electronics products are also exported by air with more than 70 percent going through Noi Bai airport in the North, and 14 percent through Tan Son Nhat airport in the South. Only 10 percent are exported through the Hai Phong seaport, and 3.8 percent through the HCMC seaport. The main corridors for transportation of each segment in the electronics value chain are: - For electronic parts and component exports: NR13, NR22, HCM – Long Thanh expressway, NR51, NR1 (Quang Tri – Quang Nam, Lang Son – Ha Nam), NR14E, NR9, AH13, NR32, Noi Bai – Lao Cai expressway (Phu Tho), Phap Van – Cau Gie – Ninh Binh expressway (to Ha Nam), NR21A (Ha Nam – Nam Dinh), NR10, AH14, NR18, NR31, NR38 - For electronic final product exports: NR22, HCM – Long Thanh expressway, NR51, NR1 (Quang Tri – Quang Ngai, Lang Son – Ninh Binh), NR24B, NR9, Ha Noi – Thai Nguyen expressway, Ha Noi – Bac Giang expressway, Phap Van – Cau Gie – Ninh Binh expressway, NR21A (Ha Nam – Nam Dinh), NR10, AH14, NR18, NR31, NR38 88 Vietnam: Connecting value chains for trade competitiveness MAP A3.2. Connective propensity of the electronics value chain LQ Final Product 1.1 - 2.0 2.1 - 5.0 5.1 - 10.0 10.1 - 15.0 LQ EE Components 1.4 - 2.0 2.1 - 5.0 5.1 - 10.0 10.1 - 16.5 Top 10 Electronic firms Disclaimer: The boundaries, colors, (income) denominations and other information shown on 10,000,000 any map in this work do not imply any judgement 100,000,000 on the part of The World Bank concerning the legal status of any territory Airport or the endorsement or acceptance of such Border gate boundaries. Seaport FinalProducts(4III) - Export EEComponents(4I) - Export Source: I/O Table 2016, Enterprise Census 2011 and 2016, Customs, and calculation by authors. Annex 3 – Analysis of the electronics value chain 89 Annex 4 Analysis of the automotive value chain A4.1. Industry overview The automotive value chain is not young in Vietnam. It has developed for more than twenty years, with the establishment of vehicle assembly plants by multinational companies, like Toyota, Honda, Ford, GM, and Mercedes-Benz etc. So far, the industry’s development has heavily relied on high tariff barriers on imported vehicles and the GVC allocation of those multinationals companies. Vietnam is one of five ASEAN countries with an automotive industry. Compared to the other four ASEAN countries, Vietnam has the smallest market size and production capacity. However, a high sector growth rate and big market potential of 100 million population are advantages of Vietnam’s automotive industry. New entries of local auto makers recently, namely Truong Hai Auto Corporation (Thaco) and Vinfast, suggest the attractiveness of the sector in the near future. FIGURE A4.1. Employment in the automotive The sector is characterized as a heavy industry industry by its high sophistication, and technological 140,000 2.0% and capital intensiveness. However, thanks Number of employees (person) 120,000 to the high sector growth between 2010 and 100,000 1.5% 2016, sector employment increased from 70 80,000 60,000 1.0% thousand to about 130 thousand people, and 40,000 0.5% its share in total manufacturing employment 20,000 grew slightly from 1.6 to 1.9 percent as shown - 0.0% 2010 2011 2012 2013 2014 2015 2016 in Figure A4.1. Number of employees A single car has about 30,000 parts, counting Share in total mnf employees Source: GSO Statistical Yearbook, 2017. every part down to the smallest screws. Some parts are made in-house by auto makers, but many are made by suppliers hierarchized in different tiers. The 30,000 parts are made from different raw materials (metal, rubber, plastics, glasses) and through different manufacturing processes. As such, proper sector development will have spillover effects to markets and create other sectors. Because of the small local market size and weak supporting industry, the localization ratio of Vietnam’s automotive industry is still low compared to other ASEAN countries, and the sector naturally relies on imported parts and components. However, as shown by the trading data in Figure A4.2, Vietnam has a trade surplus in auto parts thanks to foreign export-oriented auto-parts makers locating in Vietnam, like Denso, Bosch, Yazaki, etc. 90 Vietnam: Connecting value chains for trade competitiveness FIGURE A4.2. Automotive sector trade value 6,000 5,000 4,000 Trade value (mil. USD) 3,000 2,000 1,000 - 2010 2011 2012 2013 2014 2015 2016 2017 2018 (1,000) (2,000) (3,000) Parts export Parts import Parts trade balance CBU import Total trade balance Source: Customs. FIGURE A4.3. Decomposition of the automotive gross export Figure A4.3 shows the decomposition of 100% gross exports in the automotive sector. As 80% 60% illustrated, more than 50 percent value-added 40% embedded in Vietnam’s automotive exports 20% (most are auto parts) derive from foreign 0% sources. About 30 percent are directly from 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 auto assemblers, and the remaining are from Direct domestic Indirect domestic auto-parts suppliers. Foreign Reimported domestic Source: OECD. A4.2. Value-chain links As mentioned, a single car is sophisticated and assembled from more than 30,000 parts made from different sectors by different manufacturing processes. As such, identifying inter-sectoral links for the automotive value chain is challenging. The value chain obtained from the I/O table 2016 is graphed in Figure A4.4, which shows that first-tier suppliers provide trailers and semi-trailers, seats (grouped in the furniture sector), iron, steel, and metal products, while second-tier suppliers provide wiring, metal and plastic products, casting services, and so on. Annex 4 – Analysis of the automotive value chain 91 FIGURE A4.4. Automotive value-chain links Motor vehicles (IO 89) Final Trailers and Fabricated metal Products of iron Furniture (IO 94) First semi-trailers (IO 90) products etc. (IO 76) and steel (IO 74) SI = 2.3% tier SI = 60.9% SI = 12.4% SI = 2.6% Wiring and wiring Products of iron Products of iron Fabricated metal Second devices (IO 83) and steel (IO 74) and steel (IO 74) products, etc. (IO 76) tier SI = 33.0% SI = 40.3% SI = 70.3% SI = 6.8% Fabricated metal Fabricated metal Metal ores (IO 31) Other chemical products etc. (IO 76) products etc. (IO 76) SI = 5.7% products, man-made SI = 18.6% SI = 26.1% bres (IO 66) Fabricated metal SI = 4.3% Colour and precious products, etc. (IO 76) Other textiles (IO 52) SI = 5.4% metals, casting services SI = 11.4% of metals (IO 75) SI = 8.6% Products of iron and steel (IO 74) Other chemical SI = 6.9% products, man-made bres (IO 66) Colour and precious SI = 3.0% metals, casting services of metals (IO 75) Other electrical SI = 4.1% equipment (IO 86) SI = 2.9% Plastic products (IO 69) Basic chemicals (IO 62) SI = 2.8% SI = 2.6% Source: I/O Table 2016, authors. Because a single vehicle is assembled from many parts, the I/O table cannot describe the value chain in detail. To reflect the reality of the value-chain hierarchy, sector links are refined in Figure A4.7 into three segments, including (i) parts and components, (ii) modules, and (iii) final assembly. In Vietnam, due to the small local market size, most modules are imported or produced in-house by the auto makers (final assembly), thus the local value chain has been refined into two segments: parts and components (VSIC293), and modules and final assembly (VSIC292 and 291). FIGURE A4.5. Automotive value-chain segments PARTS AND COMPONENTS SYSTEMS: MODULES FINAL ASSEMBLY MATERIALS (VSIC293) (VSIC292) (VSIC291) Iron & Steel Electronic components Interior system: seat Trailer (VSIC3=24100; 259999) interior trim, cockpit Mechanical components module Passenger vehicles Plastics & Rubbers (VSIC22209) Composite components Body system: skin, nish, trim, doors Buses Metal products Wiring (VSIC24310) E&E system: ignition, Trucks Aluminum components chassis electronic, interior Chemicals electronics (VSIC20110) Rubber components Chassis system: drive train, Software rolling chassis, front and rear end modules Source: Authors. 92 Vietnam: Connecting value chains for trade competitiveness A4.3. Spatial structure and value-chain mapping Figure A4.8 illustrates local distribution of auto parts and components. Provinces that had higher concentration of auto parts and components were plotted in the upper-right quadrant, including Hai Phong, Hung Yen, Quang Nam, and Ha Nam. Provinces that had high concentration (high LQ index) but decrease in LQ between 2011 and 2016 were plotted in upper-left quadrant, including Ben Tre, Hai Duong, Thai Nguyen, Dong Nai, Da Nang, Phu Tho, and Vinh Phuc. FIGURE A4.6. Locational distribution of the auto parts and components segment 12 Ben Tre, 9.4 Hai Phong, 7.5 8 Hai Duong, 5.2 Hung Yen, 3.8 LQ2016 4 Thai Nguyen, 2.2 Da Nang, 1.2 Quang Nam, 3.1 -8 -4 0 0 4 Dong Nai, 2.2 Ha Nam, 1.5 Phu Tho, 1.2 Vinh Phuc, 1.0 -4 Circle size = number of employee 2016, Changing in LQ between 2016 - 2011 Min Phu Tho= 1300, Max Hai Phong = 25000 Source: Enterprise Census 2011 and 2016, calculation by authors. Figure A4.9 shows the provinces that had high concentration of final assembly. The highest LQ and increase in LQ between 2011 and 2016 was in Quang Nam, because the Truong Hai Auto Corporation is located in the Chu Lai Economic Zone. Vinh Phuc is home to Toyota and Honda, while Dong Nam has Mercedes-Benz Vietnam and Isuzu. The automotive industry is a sophisticated sector that normally requires suppliers to set-up near assemblers to easily implement just-in-time (JIT) manufacturing. However, few provinces appeared in both figures suggesting links of the automotive value chain are still weak in Vietnam. This is because parts and components produced in Vietnam are mostly for export, while auto makers import parts and components to assemble in Vietnam, thus links between parts producers and assemblers are not strong. Annex 4 – Analysis of the automotive value chain 93 FIGURE A4.7. Locational distribution of the modules and final assembly segment 19 Hai Duong, 1.8 Thua Thien Hue, 1.4 Quang Nam, 16.8 14 Quang Ninh, 1.4 Vinh Phuc, 9.9 9 LQ2016 Thanh Hoa, 1.9 Ninh Binh, 3.8 4 Hung Yen, 2.9 Dong Nai, 1.6 -3 -1 1 3 5 -1 -6 Circle size = number of employee 2016, Changing inLQ between 2016 - 2011 Min Quang Ninh = 450, Max Quang Nam = 3200 Source: Enterprise Census 2011 and 2016, calculation by authors. Map A4.1 shows the automotive value-chain spatial structure. The auto parts and components segment is shown in turquoise and the final assembly segment is in bright green. The top five largest automotive companies are shown in red circles, located in Vinh Phuc, Hai Duong, and Quang Nam. In the North, Vietnam imports completed vehicles (CBUs), including trucks (HS8704), passenger vehicles (HS8703), and special vehicles (HS8705) from China through Huu Nghi and Mong Cai land gateways, and from other countries through the Dinh Vu port. In central Vietnam, CBUs including trucks and passenger vehicles are mainly imported through Tien Sa port in Da Nang. Southern Vietnam imports trucks and passenger vehicles mainly through Cat Lai port. 94 Vietnam: Connecting value chains for trade competitiveness MAP A4.1. Geographic distribution of the automotive value chain Disclaimer: The boundaries, colors, denominations and other information shown on any map in this work do not imply any judgement on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. LQ Systems&Modules 1.4 - 2.0 2.1 - 5.0 5.1 - 10.0 10.1 - 16.8 LQ Parts&Components 1.2 - 2.0 2.1 - 5.0 5.1 - 9.4 Top 10 Automotive firms (income) 10,000,000 Source: I/O Table 2016, Enterprise Census 2011 and 2016, Customs, and calculation by authors. Annex 4 – Analysis of the automotive value chain 95 A4.4. Value-chain-based connectivity and key corridors Map A4.2 demonstrates the connective propensity of the automotive value chain. The two pie-charts show the importance of two seaports in Hai Phong and HCMC, for the automotive trade. These two seaports contributed more than 70 percent of total sector imports and 80 percent of total exports. Provinces should pay attention to local and regional automotive industry development to ensure there are no policy barriers to agglomeration of auto parts and components producers, and to ensure access for domestic and foreign suppliers and buyers. MAP A4.2. Connective propensity of the automotive value chain LQ Systems&Modules 1.4 - 2.0 2.1 - 5.0 5.1 - 10.0 10.1 - 16.8 LQ Parts&Components 1.2 - 2.0 2.1 - 5.0 5.1 - 9.4 Top 10 Automotive firms (income) 10,000,000 Airport Disclaimer: The boundaries, colors, denominations and other Border gate information shown on any map in this work do Seaport not imply any judgement on the part of The World Bank concerning the legal Sys&Modules(5II) - Export status of any territory or the endorsement PartsComps(5I) - Sys&Modules(5II) or acceptance of such boundaries. Source: I/O Table 2016, Enterprise Census 2011 and 2016, Customs, and calculation by author. 96 Vietnam: Connecting value chains for trade competitiveness Annex 5 Analysis of the wood products value chain A5.1. Industry overview FIGURE A5.1. Employment in wood products Wood products flourished in the past few manufacturing years, with more than 8,000 wood processing and furniture manufacturing enterprises. 600,000 10% However, employment in the sector did not Number of employees (person) 500,000 8% change significantly. In 2010, wood processing 400,000 6% and furniture manufacturing enterprises 300,000 4% attracted nearly 400,000 workers, and 200,000 2% about 100 thousand more workers by 2016, 100,000 0% accounting for about 7 percent of the total - 2010 2011 2012 2013 2014 2015 2016 labor of the manufacturing sector. Although it Furniture Wood processing increased in quantity, the proportion of wood Share in total mnf employees products labor in the manufacturing sector Source: Statistical Yearbook 2017. decreased from 9 percent in 2010 to 7 percent in 2016, proving strong labor development in FIGURE A5.2. Rapid export growth in wood other sectors as shown in Figure A5.1. products The wood sector is a key export industry of 5,000 4,500 Vietnam, ranking fifth after electronics, textiles, footwear and machinery. In 2010, exports Export value (mil. USD) 4,000 3,500 3,000 of wood products reached US$ 3 billion, up 2,500 to US$ 6 billion in 2017, and US$ 8 billion in 2,000 2018. The wood sector is targeted to reach 1,500 1,000 over US$ 10 billion in exports by 2020. The 500 main wood exports were furniture and fuel - wood. At present, domestic material timber 06 07 08 09 10 11 12 13 14 15 16 17 20 20 20 20 20 20 20 20 20 20 20 20 does not meet the demand of wood industry. Wooden furniture Articles of wood Each year, Vietnam imports about US$ 2 billion Source: ITC Trademap. in wood materials to meet production needs for domestic and export markets. Figure A5.2 shows a high growth rate for exports of wooden furniture and wood articles over the past decade. Annex 5 – Analysis of the wood products value chain 97 FIGURE A5.3. Decomposition of wood Figure A5.3 shows the value-added structure products’ gross export of Vietnam’s wood exports. Generally, in the last 10 years, this structure did not change 100% much. About 50 percent of the sector’s 80% added value was based on imports, over 60% 30 percent from supporting industries, and 40% nearly 20 percent created within the industry. 20% Compared to other export-oriented sectors, the domestic value-addition of wood was 0% quite high and stable. With the plan of planting 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 and developing material forests, in the future, Direct domestic Indirect domestic Reimported domestic Foreign the value-added structure could continue to increase its domestic value addition. Source: OECD. A5.2. Value-chain links Inter-sectoral links for the wood products value chain are identified and graphed in Figure A5.4. The links show more than 70 percent of furniture were from standing timber and wood products. Because sectors in the I/O table are not disaggregated at 4- or 5-digit VSIC codes, the links show standing timber and wood products are inputs to its own sector. Therefore, the sector’s value chain is refined in Figure A5.5 into four segments: planting and foresting, sawmilling, wood products (VSIC 1621m, 1622, 1623, and 1629), and furniture (VSIC 31001). FIGURE A5.4. Wood products value-chain links Final Furniture (IO 94) products Standing Wood, products Fabricated Other chemical Paper and Furniture timber of wood metal products products, paper products (IO 94) First (IO 23) and cork, etc (IO 76) man-made (IO 57) SI = 2.2% tier SI = 40.3% (IO 56) SI = 6.8% bres (IO 74) SI = 2.3% SI = 31.7% SI = 4.3% Live forest Standing Wood, products tree plants, timber of wood Second seeds (IO 22) (IO 23) and cork, etc tier SI = 42.0% SI = 33.7% (IO 56) SI = 15.8% Source: I/O Table 2016, authors calculation. 98 Vietnam: Connecting value chains for trade competitiveness FIGURE A5.5. Wood products value-chain segments Wood products (VSIC1621, 1622, 1623, 1629) Planting & foresting Sawmilling (VSIC0125, 0210) (VSIC161) Furniture (VSIC31001) Source: Authors. A5.3. Spatial structure and value-chain mapping Figure A5.6 shows planting and foresting are concentrated mainly in western provinces. Gia Lai, Binh Phuoc, and Binh Duong provinces employ over 10,000 in the sector. From 2011 to 2016, planting and foresting were more concentrated in the provinces in the upper-right quadrant, including Gia Lai, Kon Tum, Lai Chau, and Quang Nam. These provinces are sources of materials for wood processing and furniture manufacturers. FIGURE A5.6. Locational distribution of the planting and foresting segment 60 Bac Kan, 1.8 Quang Tri, 4.9 Dak Lak, 11.8 Dak Nong, 6.1 50 Quang Binh, 2.3 Kon Tum, 25.9 Binh Phuoc, 23.6 40 Tuyen Quang, 1.5 Ha Tinh, 3.1 Gia Lai, 34.5 Son La, 12.4 30 Nghe An, 2.1 LQ2016 Lang Son, 1.4 Quang Nam, 4.9 20 Tay Ninh, 4.2 Lai Chau, 14.4 10 Dien Bien, 2.6 Dong Nai, 1.0 Yen Bai, 2.8 0 -13 -8 -3 2 7 12 17 22 27 Ba Ria - Vung Tau, 2.0 Lam Dong, 1.5 Ha Giang, 2.1 -10 Binh Thuan, 1.2 Binh Duong, 1.4 Circle size = number of eployee 2016, Changing in LQ between 2016 - 2011 Min Bac Kan = 100, Max Binh Phuoc = 22,200 Source: Enterprise Census 2011 and 2016, calculation by authors. Annex 5 – Analysis of the wood products value chain 99 As seen in Figure A5.7., sawmilling was most concentrated in Binh Duong, with more than 8,000 employees. Binh Dinh also had high sawmilling agglomeration, with an LQ of 8.4 and more than 3,000 workers. Dong Nai also employed many sawmilling workers. Phu Tho, Tuyen Quang, Quang Binh, Binh Dinh, and Kon Tum provinces are plotted in the upper-right quadrant with high LQs in 2016, and greater concentration than in 2011. FIGURE A5.7. Locational distribution of the sawmilling segment 19 Quang Binh, 4.7 Phu Tho, 4.3 Quang Nam, 3.4 15 Quang Tri, 10.4 Bac Kan, 8.9 Quang Ngai, 9.9 Binh Dinh, 8.5 Yen Bai, 4.9 11 Tuyen Quang, 9.4 Gia Lai, 6.0 7 Binh Phuoc, 4.9 Lam Dong, 4.6 LQ2016 Kon Tum, 3.7 Thua Thien Hue, 3.7 3 Bac Giang, 1.6 Dak Lak, 2.0 -10 -7 -4 -1 2 5 8 -1 Thanh Hoa, 1.7 Ha Tinh, 1.8 Lang Son, 1.6 Nghe An, 2.0 -5 Kien Giang, 1.6 Khanh Hoa, 1.0 Quang Ninh, 1.3 -9 Binh Duong, 2.7 Dong Nai, 1.8 Phu Yen, 2.4 Circle size = Number of employee 2016, Changing in LQ between 2016 - 2011 Min Lang Son = 120, Max Binh Duong = 8000 Source: Enterprise Census 2011 and 2016, calculation by authors. The number of provinces highly concentrated in wood products manufacturing were much fewer than other segments. Figure 5.8 shows only a few provinces had high wood processing agglomeration, namely Binh Duong, Dong Nai, Binh Dinh, Gia Lai, Phu Yen, and Yen Bai. Notably, Binh Duong and Dong Nai have a large labor force, with 180,000 and 65,000 employees respectively. The 2016 LQs for these two provinces were also high and increased from 2011. The remaining provinces had small labor scales, and decreasing LQs between 2011 and 2016. 100 Vietnam: Connecting value chains for trade competitiveness FIGURE A5.8. Locational distribution of the wood products and furniture segment 10.0 Binh Duong, 6.5 8.0 Binh Dinh, 5.8 6.0 LQ 2016 Dong Nai, 3.0 4.0 Gia Lai, 1.2 2.0 Yen Bai, 1.1 Phu Yen, 1.2 0.0 -5.0 -4.0 -3.0 -2.0 -1.0 0.0 1.0 2.0 -2.0 Circle size = number of emplyee 2016, Changing in LQ between 2016 - 2011 Min Yen Bai = 1000, Max Binh Duong = 183500 Source: Enterprise Census 2011 and 2016, calculation by authors. Map A5.1 shows the spatial structure of the wood products value chain. Planting and foresting segments were mainly concentrated in the Northwest and Central Highlands provinces, sawmilling activities concentration spanned the country, while production of wood products was highly concentrated in Binh Duong and Dong Nai. These two provinces are home to the five largest wood processing enterprises. Wood imports are mainly through Dinh Vu port in the North, Le Thanh gateway in Central Vietnam, and Cat Lai port in the South. The high volume of exported wood products through Mong Cai gateway suggests China may be a big market for Vietnamese wood products. Annex 5 – Analysis of the wood products value chain 101 MAP A5.1. Geographic distribution of the wood processing value chain Disclaimer: The boundaries, colors, denominations and other information shown on any map in this work do not imply any judgement on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. LQ Wood product 3.0 6.5 LQ Sawmilling 1.0 - 2.0 2.1 - 5.0 5.1 - 10.0 10.1 - 10.4 LQ Planting 1.2 - 2.0 2.1 - 5.0 5.1 - 10.0 10.1 - 30.0 30.1 - 34.5 Top 10 Wood firms (income) 2,000,000 Source: I/O Table 2016, Enterprise Census 2011 and 2016, Customs, and calculation by authors. 102 Vietnam: Connecting value chains for trade competitiveness A5.4. Value-chain-based connectivity and key corridors Map A5.2 shows connective propensity of wood products. Along the value chain, timber products are transferred from the Northwest and Central Highlands provinces to the sawmilling and wood processing provinces. As shown in the map, Binh Duong, Dong Nai, and neighboring provinces like Lam Dong, Binh Phuoc, and Tay Ninh have formed a wood processing cluster, with the presence of leading enterprises in the sector. In the central region, connecting the Central Highlands to coastal provinces from Quang Nam to Binh Dinh could also form a wood processing cluster. In the northern region, wood processing is not really agglomerated and there are fewer leading enterprises than in the South. Strengthening the wood products value chain depends on connecting northwest provinces to midland provinces and the Red River Delta region. Below are the main transportation corridors for wood: - From planting to logging: NR22, NR22B, NR51, HCM – Long Thanh – Dau Giay expressway, NR56, NR1 (Lang Son – Can Tho), NR80, HCMC - Trung Luong – My Thuan expressway, NR20, NR13, NR14, AH17, NR28, NR27, NR26, NR29, NR25, NR19, NR24, NR24B, NR14E, NR15, AH13, NR37, NR21A, NR32, NR279 (Dien Bien, Tuyen Quang), Noi Bai – Lao Cai expressway, NR4D, NR2, NR34, NR1B, Ha Noi – Thai Nguyen expressway, Ha Noi – Bac Giang expressway, NR38, AH14, NR4B, NR18, NR10 - From sawmilling to wood production: NR1 (Lang Son – Can Tho), NR80, NR13, NR14, HCM – Long Thanh – Dau Giay expressway, HCMC - Trung Luong – My Thuan expressway, NR51, NR20, AH17, NR26, NR29, NR25, NR19, NR24, NR24B, NR14E, NR21A, NR32, NR279 (Tuyen Quang), Ha Noi – Thai Nguyen expressway, Ha Noi – Bac Giang expressway, NR38, AH14, NR18, NR10 - For exports of wood products: NR22, NR51, NR13, NR14, AH17, HCM – Long Thanh – Dau Giay expressway, NR19C, NR19, NR1 (Quang Tri – Binh Dinh, Lang Son – Ha Noi), NR9, AH14, NR18, Noi Bai – Lao Cai expressway (to Yen Bai) Annex 5 – Analysis of the wood products value chain 103 MAP A5.2. Connective propensity of the wood products value chain LQ Wood product 3.0 6.5 LQ Sawmilling 1.0 - 2.0 2.1 - 5.0 5.1 - 10.0 10.1 - 10.4 LQ Planting 1.2 - 2.0 2.1 - 5.0 5.1 - 10.0 10.1 - 30.0 30.1 - 34.5 Top 10 Wood firms Disclaimer: (income) The boundaries, colors, denominations and other 2,000,000 information shown on any map in this work do not imply any judgement on the part of The World Airport Bank concerning the legal status of any territory Border gate or the endorsement or acceptance of such Seaport boundaries. WoodProduct(6IV) - Export Sawmilling(6III) - WoodProduct(6IV) Planting(6I) - Logging(6II) Source: I/O Table 2016, Enterprise Census 2011 and 2016, Customs, and calculation by authors. 104 Vietnam: Connecting value chains for trade competitiveness Annex 6 Analysis of the rubber value chain A6.1. Industry overview FIGURE A6.1. Employment in the rubber Rubber is an input material for final products industry and components manufacturing, appearing in most manufacturing industries like Labors in rubber industries 40000 electronics, machinery, and vehicles. Between 35000 30000 2011 and 2016, rubber manufacturing (persons) 25000 enterprises created nearly 20,000 new jobs, 20000 with employment increasing from 45,587 15000 10000 to 65,232 (accounting for 0.5 percent of the 5000 nation’s labor force), and the labor growth rate 0 Rubbers Tires Other rubber was 7.5 percent, higher than the national labor products 2011 2016 growth rate of 5.6 percent. Source: GSO Enterprises surveys, 2011, 2016. Recently, Vietnam exported more than US$ 3 FIGURE A6.2. Rubber sector exports billion in rubber and rubber products annually. More than half of Vietnam’s rubber export 3,500,000 turnover is from primary rubber. Vietnam 3,000,000 currently ranks fourth in the world in natural Export value (1,000 USD) 2,500,000 rubber exports, after Thailand, Indonesia, and 2,000,000 Malaysia. Vietnam’s rubber industry depends heavily on the export market, accounting for 1,500,000 over 80 percent of its output. Between 2011 1,000,000 and 2015, there was a sharp decline in global 500,000 rubber prices, so although rubber exports - increased in quantity they decreased in value. Since 2016, the global rubber price has 20 6 20 7 20 8 20 9 20 0 20 1 20 2 20 3 20 4 20 5 16 17 0 0 0 0 1 1 1 1 1 1 20 20 Rubbers Tyres Others gradually recovered, helping the domestic rubber export grow again. Source: ITC Trademap. Annex 6 – Analysis of the rubber value chain 105 A6.2. Value-chain links Inter-sectoral links for the rubber value chain are graphed in Figure A6.3. The links show the three main inputs for rubber products are rubber products (29%), rubber in primary forms (19%), and natural rubber (15%). Because sectors in the I/O table are not disaggregated at 4- or 5-digit VSIC codes, the table shows rubber products are inputs to its own sector. Therefore, the sector’s value chain is refined in Figure A6.4 into three segments: planting (VSIC 0125), production of rubber in primary forms (VSIC 20132), and final rubber products (VSIC 22110, 22120). FIGURE A6.3. Rubber value chain links Final Rubber products (IO 68) products Rubber Plastic and Natural Basic Fabricated Products of Other chemical First products synthetic rubber chemicals metal iron and products, tier (IO 68) rubber in (IO 12) (IO 12) products steel man-made SI = 29.2% primary forms SI = 15.0% SI = 4.6% (IO 76) (IO 74) bres (IO 64) SI = 4.4% SI = 3.3% (IO 66) SI = 19.2% SI = 3.0% Plastic and Basic Fertilizers Re ned Bricks, Wood, Pesticides Second synthetic chemicals and nitrogen petroleum blocks, products of (IO 65) tier rubber in (IO 12) products products tiles etc. wood & cork SI = 4.0% primary forms SI = 22.0% (IO 63) (IO 60) (IO 71) (IO 56) (IO 64) SI = 32.9% SI = 5.8% SI = 4.8% SI = 4.6% SI = 34.3% Source: I/O Table 2016, authors calculation. FIGURE A6.4. Rubber value-chain segments Re ned petroleum products (VSIC19200) Rubber processing Final rubber products (VSIC20132) (VSIC22110, 22120) Planting Fertilizer (VSIC0125) Source: Authors. A6.3. Spatial structure and value-chain mapping Figure A6.5 shows the provincial concentration of rubber planting. The circle size indicates the province’s labor force in the sector. The horizontal axis indicates the LQ change between 2011 and 2016 and the vertical axis is the 2016 LQ index. Binh Phuoc, Gia Lai, and Binh Duong had a large labor force, with over 10,000 workers. In terms of concentration expressed by 2016 LQ, Binh Phuoc, Gia Lai, and Kon Tum were highly concentrated in rubber tree planting, especially in Kon Tum and Gia Lai, where the 2016 LQ increased sharply from 2011, while the Binh Phuoc LQ decreased slightly. In addition to these three provinces, rubber planting was also highly concentrated in other provinces in mountainous, highland, and southwestern regions. 106 Vietnam: Connecting value chains for trade competitiveness FIGURE A6.5. Locational distribution of the rubber planting segment 42 37 Binh Phuoc, 26.9 Gia Lai, 39.1 32 Son La, 13.8 Kon Tum, 27.1 27 LQ2016 Quang Tri, 3.6 22 Dak Lak, 12.4 Tay Ninh, 4.8 17 Dien Bien, 2.7 12 Ba Ria - Vung Tau, 2.2 Quang Nam, 5.5 7 Lai Chau, 1.0 Dak Nong, 2.3 2 -12 -7 -2 3 8 13 18 23 -3 Dong Nai, 1.1 Nghe An, 2.0 Binh Duong, 1.6 Ha Tinh, 3.3 Yen Bai, 1.6 Circle size = number of employee, Changing in LQ between 2016 - 2011 Min Lai Chau = 100, Max Binh Phuoc = 22 200 Source: Enterprise Census 2011 and 2016, calculation by authors. Figure A6.6 shows rubber processing concentration across the country. Data from the Enterprise Censuses 2011 and 2016 show rubber processing was mainly concentrated in Binh Thuan and Quang Binh. Between 2011 and 2016, although the LQs of these two provinces were high, their trends were opposite; In Binh Thuan the 2016 LQ increased sharply from 2011, but in Quang Binh it fell sharply. FIGURE A6.6. Locational distribution of the rubber processing segment 100 Quang Binh, 81.2 Binh Thuan, 79.4 80 60 Thua Thien Hue, 1.6 Dak Lak, 2.3 LQ2016 40 Nghe An, 3.3 Ba Ria - Vung Tau, 2.6 20 Quang Ninh, 3.8 0 -80 -60 -40 -20 0 20 40 60 80 100 An Giang, 1.8 -20 Changing in LQ between 2016 - 2011 Circle size = number of employee 2016, Min An Giang = 50, Max Binh Thuan = 2,300 Source: Enterprise Census 2011 and 2016, calculation by authors. Annex 6 – Analysis of the rubber value chain 107 Production of rubber products (tires, tubes, and other rubber products) was highly concentrated in Tay Ninh, Hai Phong, Dong Nai, and Binh Duong provinces. These provinces had a large labor force, with over 5,000 employees working in rubber companies. Long An, Da Nang, and Hung Yen had 2016 LQs greater than 1, and a labor size of over 1,000 employees. Notably, the 2016 LQ decreased compared to 2011 in all provinces, yet Tay Ninh, with the highest 2016 LQ, had the largest decrease. FIGURE A6.7. Locational distribution of the final rubber products segment 12 Tay Ninh, 9.6 Binh Duong, 2.3 Dong Nai, 2.8 8 Hung Yen, 1.4 Hai Phong, 3.6 LQ2016 4 Dak Nong, 2.9 Quang Tri, 2.8 Da Nang, 1.3 Long An, 1.2 0 -7 -4 -1 2 5 8 -4 Changing in between LQ2016 & 2011 Circle size = number of employment 2016, Min Dak Nong = 230, Max Binh Duong = 10,400 Source: Enterprise Census 2011 and 2016, calculation by authors. Map A6.1. shows the geographic distribution of each segment in the rubber value chain. Rubber trees are planted mainly in the northwest mountainous provinces of Son La, Dien Bien, Lai Chau, and Yen Bai, and in the highland provinces of Binh Phuoc, Dak Lak, Gia Lai, and Kon Tum. There were fewer provinces concentrated in rubber processing (Binh Thuan, Quang Binh, Nghe An, and Quang Ninh) compared to other segments. Rubber products manufacturing were mainly concentrated in the southern provinces of Tay Ninh, Dong Nai, and Binh Duong. Two provinces in the North, Hai Phong and Hung Yen, also had high concentration in rubber manufacturing. Most of the big rubber firms in the top ten by revenue are based in the South and only one firm is located in Hai Phong province in the North. A6.4. Value-chain-based connectivity and key corridors Map A6.2. demonstrates the connective propensity of the rubber value chain. Each segment uses different corridors for their product transportation. Main corridors of each segment are highlighted below. - From planting to processing: NR91 (An Giang), NR1 (Ha Noi – Binh Dinh, Binh Thuan – Can Tho), HCMC - Trung Luong – My Thuan expressway, NR51, NR56, HCM – Long Thanh – Dau Giay 108 Vietnam: Connecting value chains for trade competitiveness expressway, NR55, NR28, NR22B, NR22, NR13, NR14, AH17, NR19, PR616 (Kon Tum), PR614 (Kon Tum), NR14E, NR15, HCM road (Hoa Binh – Thanh Hoa), AH13, NR10, NR18, Noi Bai – Lao Cai expressway, NR4D - For exports of rubbers: NR91 (An Giang), NR1 (Quang Binh – Da Nang, Binh Thuan – Can Tho), HCMC - Trung Luong – My Thuan expressway, NR51, NR56, HCM – Long Thanh – Dau Giay expressway, NR55, NR13, NR14, NR9, NR36, NR10, NR18 (Quang Ninh) MAP A6.1. Geographic distribution of the rubber value chain Disclaimer: The boundaries, colors, denominations and other information shown on any map in this work do not imply any judgement on the part of The World Bank concerning the legal LQ Final Products status of any territory or the endorsement 1.2 - 2.0 or acceptance of such 2.1 - 5.0 boundaries. 5.1 - 10.0 LQ Processing 1.6 - 2.0 2.1 - 5.0 5.1 - 10.0 10.1 - 30.0 30.1 - 81.2 LQ Planting 1.0 - 2.0 2.1 - 5.0 5.1 - 10.0 10.1 - 30.0 30.1 - 39.1 Top 10 Rubber firms (income) 100,000 1,000,000 Source: I/O Table 2016, Enterprise Census 2011 and 2016, Customs, and calculation by authors. Annex 6 – Analysis of the rubber value chain 109 MAP A6.2. Connective propensity of the rubber value chain LQ Final Products 1.2 - 2.0 2.1 - 5.0 5.1 - 10.0 LQ Processing 1.6 - 2.0 2.1 - 5.0 5.1 - 10.0 10.1 - 30.0 30.1 - 81.2 LQ Planting 1.0 - 2.0 2.1 - 5.0 5.1 - 10.0 10.1 - 30.0 30.1 - 39.1 Disclaimer: Top 10 Rubber firms The boundaries, colors, (income) denominations and other information shown on 100,000 any map in this work do not imply any judgement 1,000,000 on the part of The World Bank concerning the legal status of any territory Airport or the endorsement or acceptance of such Border gate boundaries. Seaport Processing(9II) - Export Planting(9I) - Processing(9II) Source: I/O Table 2016, Enterprise Census 2011 and 2016, Customs, and calculation by authors. 110 Vietnam: Connecting value chains for trade competitiveness Annex 7 Analysis of the rice value chain A7.1. Industry overview FIGURE A7.1. Employment in the rice sector In the rice sector, Enterprise data only cover employment in registered enterprises, not 40,000 farmers, and households. In 2016, only 200 Number of employees (persons) 35,000 enterprises registered rice planting activities, 30,000 employing nearly 4,000 workers; meanwhile, 25,000 rice processing had more than 1,000 20,000 15,000 enterprises, and employed more than 36,000 10,000 workers. In 2016, more than 100 enterprises 5,000 participated in rice planting creating about - 2,500 more jobs than in 2011. However, Rice planting Rice processing 2011 2016 rice processing decreased slightly in both the numbers of enterprises and employees. Source: GSO Enterprises Census, 2011, 2016. This probably reflects the trend toward FIGURE A7.2. Rice exports industrialization in agricultural, and a labor shift away from agriculture to other economic 4,000,000 activities. 3,500,000 Export value (1,000 USD) 3,000,000 Currently, Vietnam ranks third in rice exports, 2,500,000 2,000,000 after India and Thailand. The main rice export 1,500,000 is semi-milled and milled rice. After a period of 1,000,000 continuous growth from 2006 to 2012, export 500,000 values reached over US$ 3.6 billion in 2013. - From 2013 to 2017, export turnover declined 06 07 08 09 10 11 12 13 14 15 16 17 sharply due to the decline in global rice prices 20 20 20 20 20 20 20 20 20 20 20 20 Semi-milled and milled rice Broken rice and demand. Some traditional rice import Brown rice Paddy markets like Bangladesh and Indonesia have Source: ITC Trademap. pursued self-sufficiency policies limiting the amount of imported rice. China, the world’s largest import market, raised technical barriers and set stricter requirements on product quality, which also caused a decline in Vietnamese domestic rice exports. In 2016, Vietnam’s export turnover was US$ 2.1 billion, which increased slightly to US$ 2.3 billion in 2017. Annex 7 – Analysis of the rice value chain 111 A7.2. Value-chain links Inter-sectoral links for the rice value chain are identified from the 2016 I/O table, and graphed in Figure A7.3. The links show a simple ecosystem in rice production with three main input sources for rice planting including fertilizers and nitrogen products (31%), paddy (14%), and agrichemicals (14%). Because sectors in the I/O table are not disaggregated at 4- or 5-digit VSIC codes, the links show grain mill products are inputs for itself (34%), beside the main input of paddy (56%). Therefore, the sector’s value chain is refined simply in Figure A7.4. into two segments: planting (VSIC 0111) and final rice products (VSIC 1061). FIGURE A7.3. Rice value-chain links Final Grain mill products, starches and starch products (IO 40) product Paddy (IO 1) Grain mill products, starches and starch products (IO 40) First SI = 56.3% SI = 34.2% tier Fertilizers and nitrogen Paddy (IO 1) Pesticides and other Second products (IO 63) SI = 14.1% agrochemical products (IO 65) tier SI = 31.2% SI = 8.4% Source: I/O Table 2016, authors calculation. FIGURE A7.4. Rice processing value-chain segments Seeding (VSIC 0130) Planting Rice (VSIC 0111) (VSIC 1061) Fertilizers & Pesticides (VSIC 20120, 20210) Source: Authors. A7.3. Spatial structure and value-chain mapping As an agricultural country, rice-growing has been a key economic activity in Vietnam for a long time, especially in rural areas. Enterprise data showed rice planting was highly concentrated in Soc Trang, Bac Lieu, Binh Phuoc, Tra Vinh, and Dien Bien, with high 2016 LQs. The increase in LQs between 2011 and 2016 shows rice cultivation was still the main economic activity in the Mekong Delta Region. As shown in Figure A7.5., Hanoi Province had the largest labor scale in rice cultivation in 2016, but its LQ in 2016 was small and did not change much from 2011. After expanding capital to neighboring provinces, Hanoi has a larger suburban area where the main activity is growing rice. Because the Enterprise data only consider labor in registered enterprises, and do not include farmers and households, it is more likely that Hanoi’s high LQ and labor scale over other provinces is attributed to the higher number of enterprises registered there. 112 Vietnam: Connecting value chains for trade competitiveness FIGURE A7.5. Locational distribution of the rice planting segment 40 Hung Yen, 1.3 Ha Noi, 1.1 Soc Trang, 27.7 30 Ninh Binh, 3.7 Ha Tinh, 2.7 Dien Bien, 14.8 Kien Giang, 8.1 20 Tra Vinh, 8.5 Bac Lieu, 22.7 Nam Dinh, 4.5 Binh Phuoc, 9.6 10 LQ2016 Lam Dong, 2.3 Bac Giang, 2.9 0 -40 -30 -20 -10 0 10 20 30 40 Bac Kan, 2.5 Quang Ngai, 1.2 -10 Hoa Binh, 1.2 Dak Lak, 1.9 Changing in LQ between 2016 - 2011 Circle size = number of employee 2016, Min Bac Kan = 5, Max Binh Phuoc = 310 Source: Enterprise Census 2011 and 2016 and calculation by authors. Figure A7.6 illustrates rice processing concentration across the country. This activity took place mainly in the southern provinces of Tien Giang, Dong Thap, An Giang, Can Tho, Long An, and Tay Ninh. The provinces in the upper-right quadrant had a high LQ in 2016, which increased from 2011, showing they are increasingly strong in rice processing. FIGURE A7.6. Locational distribution of the rice processing segment Dong Thap, 17.1 Long An, 4.6 18.0 Tien Giang, 14.0 Kon Tum, 5.9 An Giang, 11.7 Vinh Long, 3.0 13.0 Tay Ninh, 6.8 Can Tho, 11.7 Quang Tri, 9.6 Soc Trang, 3.6 LQ2016 Phu Yen, 2.5 Yen Bai, 8.0 Kien Giang, 2.9 8.0 Hau Giang, 3.5 Quang Ngai, 6.4 Tra Vinh, 3.7 Dak Lak, 3.9 3.0 Bac Lieu, 2.3 Ninh Thuan, 2.9 -12.0 Hoa Binh, 1.3 -7.0 -2.0 3.0 8.0 Ba Ria - Vung Tau, 1.9 Binh Thuan, 1.0 -2.0 Binh Dinh, 1.6 Quang Binh, 1.5 Binh Phuoc, 1.0 Nghe An, 1.2 Lao Cai, 1.6 Dak Nong, 1.2 Chaning in LQ between 2016 - 2011 Circle size = number of emplyee 2016, Min Dak Nong= 50, Max Tien Giang = 5900 Source: Enterprise Census 2011 and 2016, calculation by authors. Annex 7 – Analysis of the rice value chain 113 As the third largest exporter of rice globally, seeing rice activity throughout the country is expected. However, Map A7.1 shows seeding is concentrated in Hau Giang and Lam Dong, planting is concentrated in the Mekong and Red River Deltas, and, while rice processing is spreading in every province, it is still quite focused in the Mekong Delta provinces of Tien Giang, Dong Thap, Can Tho, and An Giang. MAP A7.1. Geographic distribution of the rice value chain Disclaimer: The boundaries, colors, denominations and other information shown on any map in this work do not imply any judgement on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such LQ Rice boundaries. 1.0 - 2.0 2.1 - 5.0 5.1 - 10.0 10.1 - 30.0 LQ Planting 1.1 - 2.0 2.1 - 5.0 5.1 - 10.0 10.1 - 30.0 LQ Seeding 1.2 - 2.0 2.1 - 5.0 5.1 - 10.0 10.1 - 30.0 30.1 - 74.6 Top 10 Rice firms (income) 3,000,000 Source: I/O Table 2016, Enterprise Census 2011 and 2016, Customs, and calculation by authors. 114 Vietnam: Connecting value chains for trade competitiveness A7.4. Value-chain-based connectivity and key corridors Map A7.2 demonstrates the connective propensity of the rice value chain. Each segment in the value chain utilizes different transportation corridors, which are highlighted below. - From seeding to planting: Quan Lo – Phung Hiep expressway, NR91, NR1 (Lang Son – Quang Ngai, HCM – Ca Mau), NR80, NR53, NR13, NR14, HCM – Long Thanh – Dau Giay expressway, HCMC - Trung Luong – My Thuan expressway, NR20, NR28, NR24, NR24B, NR12B, AH13, Phap Van – Cau Gie – Ninh Binh expressway, NR21A, NR3, Noi Bai – Lao Cai expressway (Phu Tho), NR32, Ha Noi – Bac Giang expressway, NR37, NR39A, NR4A, NR1B - From planting to rice processing: Quan Lo – Phung Hiep expressway, NR91, NR1 (Lang Son – Binh Dinh, HCM – Ca Mau), NR63, NR80, NR60, NR62, NRN2, NR22B, NR51, NR56, NR53, NR13, NR14, NR55, HCM – Long Thanh – Dau Giay expressway, HCMC - Trung Luong – My Thuan expressway, NR20, NR27, NR26, NR29, NR19C, NR19, NR9, HCM road (Hoa Binh – Thanh Hoa), NR12, NR4D, NR279 (Dien Bien), NR28, NR24, NR24B, NR12B, AH13, Phap Van – Cau Gie – Ninh Binh expressway, NR21A, NR3, Noi Bai – Lao Cai expressway (to Yen Bai), NR32, Ha Noi – Bac Giang expressway, NR37, NR39A, NR18 - For rice exports: Quan Lo – Phung Hiep expressway, NR80, NR91, NR1 (Ha Noi – Ca Mau), NR60, NR30, NRN2, NR22B, NR51, HCMC - Trung Luong – My Thuan expressway, HCM – Long Thanh – Dau Giay expressway, NR13, NR14, AH17, NR24, NR24B, NR9, NR15, NR10, AH14, Noi Bai – Lao Cai expressway, AH13 (Hoa Binh) Annex 7 – Analysis of the rice value chain 115 MAP A7.2. Connective propensity of the rice value chain LQ Rice 1.0 - 2.0 2.1 - 5.0 5.1 - 10.0 10.1 - 30.0 LQ Planting 1.1 - 2.0 2.1 - 5.0 5.1 - 10.0 10.1 - 30.0 LQ Seeding 1.2 - 2.0 2.1 - 5.0 5.1 - 10.0 10.1 - 30.0 30.1 - 74.6 Disclaimer: Top 10 Rice firms The boundaries, colors, (income) denominations and other information shown on 3,000,000 any map in this work do not imply any judgement on the part of The World Airport Bank concerning the legal status of any territory or the endorsement Border gate or acceptance of such boundaries. Seaport Rice(7III) - Export Planting(7II) - Rice(7III) Seeding(7I) - Planting(7II) Source: I/O Table 2016, Enterprise Census 2011 and 2016, Customs, and calculation by authors. 116 Vietnam: Connecting value chains for trade competitiveness Annex 8 Analysis of the coffee value chain A8.1. Industry overview FIGURE A8.1. Employment in the coffee Labor in the coffee sector includes workers industry employed in planting and processing enterprises as indicated in Figure 8.1. Between 60,000 2011 and 2016, employment increased in Number of employees 50,000 both activities, from 18,785 employees in (persons) 40,000 planting and 47,703 workers in processing in 30,000 2011 to 23,728 and 52,053 employees in 2016, 20,000 10,000 respectively. - Coffee planting Coffee processing Vietnam ranks second in the world for coffee 2011 2016 exports. Exported coffee items include Source: GSO Enterprises surveys, 2011, 2016. unprocessed coffee, decaffeinated coffee, roasted coffee, roasted decaffeinated coffee, FIGURE A8.2. Coffee exports and coffee husks and skins. The main coffee export of Vietnam is unprocessed coffee, 4,000,000 accounting for over 90 percent of the total Export value (1,000 USD) 3,500,000 3,000,000 coffee export turnover, as shown in Figure 8.2. 2,500,000 2,000,000 From 2006 to 2012, coffee exports increased 1,500,000 steadily, except in 2009. From 2012 to 2017, 1,000,000 export turnover fluctuated around US$ 3 500,000 billion, with a margin of about US$ 0.5 billion - mainly because of instability of global coffee 2006 2007 2008 2009 2011 2012 2013 2014 2015 2016 2010 2017 demand and prices. Non-processed coffee Decaffeinated Roasted Roasted, decaffeinated Coffee husks and skins Source: ITC Trademap. Annex 8 – Analysis of the coffee value chain 117 A8.2. Value-chain links Figure A8.3 shows inter-sectoral links identified from the I/O Table 2016. The links determine the inputs used to produce the final coffee products. For instance, instant coffee is produced mainly from coffee beans (75%), dairy products (6.3%), sugar (2.4%), and packaging (2.9%). Inputs for coffee beans include fertilizers (60%), coffee seeds (8.4%), and gasoline (9.8%). The I/O table data do not indicate which source is a direct input (like coffee raw material), and which source is an indirect input (like fertilizer, gasoline). Therefore, the coffee value chain is refined in Figure A8.4 into two segments: coffee beans and coffee processing. FIGURE A8.3. Coffee value-chain links Final Co ee (IO 43) products Co ee beans (IO 13) Dairy products (IO 39) Plastic products (IO 69) Sugar and molasses (IO 41) First SI = 75.0% SI = 6.3% SI = 2.9% SI = 2.4% tier Fertilizers and Re ned petroleum Co ee beans Plastic and synthetic Sugar cane nitrogen products products (IO 13) rubber in primary (IO 5) Second (IO 63) (IO 60) SI = 8.4% form (IO 64) SI = 34.9% tier SI = 2.4% SI = 9.8% SI = 45.5% Source: I/O Table 2016, calculation by authors. FIGURE A8.4. Coffee processing value-chain segments Fertilizers Coffee planting Coffee beans (VSIC20120) (VSIC0126) (VSIC01260) Sugar cane Diary & Sugar Coffee processing (VSIC01140) (VSIC10500, 10720) (VSIC10790) Plastics Packaging (VSIC20131) (VSIC22201) Source: Authors. A8.3. Spatial structure and value-chain mapping Figure A8.5 shows coffee growing in Vietnam, was mainly in the Central Highlands provinces, including Dak Lak, Gia Lai, Kon Tum, Dak Nong, and Lam Dong, and the concentration in these provinces was quite high. Between 2011 and 2016, coffee growing strengthened in Dak Lak and Gia Lai, which had the highest 2016 LQs, increased from their 2011 LQs. In Kon Tum and Dak Nong, although their 2016 LQs were high, they decreased from 2011. Possibly new economic activities emerged recently in these provinces, attracting labor from coffee growing, resulting in the sharp decline in LQs between 2011 and 2016. 118 Vietnam: Connecting value chains for trade competitiveness FIGURE A8.5. Locational distribution of the coffee planting segment 120 Dak Lak, 93.2 90 60 LQ2016 Kon Tum, 35.6 Gia Lai, 41.6 30 Dak Nong, 21.1 Lam Dong, 1.6 0 -50 -40 -30 -20 -10 0 10 20 30 40 -30 Changing in LQ between 2016 - 2011 Circle size = number of empoyee 2016, Min Lam Dong=150, Max Dak Lak = 19,000 Source: Enterprise Census 2011 and 2016 and calculation by authors. Figure A8.6 shows coffee processing concentration. In addition to the Central Highlands, northern mountainous provinces like Son La, Lao Cai, Bac Kan, Phu Tho, Yen Bai, Tuyen Quang, Hoa Binh, and Ha Giang have a tradition of growing and processing arabica coffee from the French colonial period, but on a smaller scale. In addition, Binh Duong, Dong Nai, and Ben Tre are not suitable for coffee growing but employ many in processing activities. Provinces in the upper-right quadrant, including Son La, Lai Chau, and Ben Tre had high processing concentration, and this activity is on the rise, reflected by the higher concentration compared to 2011. FIGURE A8.6. Locational distribution of the coffee processing segment 20 Lam Dong, 11.5 Son La, 14.4 Dong Nai, 3.3 Yen Bai, 10.6 15 Lai Chau, 10.8 Phu Tho, 11.3 Ben Tre, 10.0 Ha Giang, 5.2 10 Ninh Thuan, 4.4 LQ2016 Tuyen Quang, 8.4 Gia Lai, 3.4 Dak Lak, 3.5 5 Bac Kan, 3.1 Dak Nong, 1.3 Hoa Binh, 1.6 -10 -5 0 0 5 10 15 Ha Tinh, 2.0 Binh Duong, 1.0 Nghe An, 1.7 Lao Cai, 1.5 -5 Changing in LQ between 2016 - 2011 Circle seize = numberoff employee 2016, Min Dak Nong = 100, Max Dong Nai = 11,000 Source: Enterprise Census 2011 and 2016 and calculation by authors. Annex 8 – Analysis of the coffee value chain 119 The coffee value chain is simple with only two segments: planting and processing. Coffee tree planting was mainly in the highlands: Kon Tum, Gia Lai, Dak Lak, and Dak Nong. Coffee processing was also concentrated in the highlands, in Lam Dong. Some northern provinces concentrated in both (Lai Chau, Son La, Yen Bai, Phu Tho) as well as some southern provinces (Ben Tre, Dong Nai, Binh Duong, Ninh Thuan), but all big coffee processing firms are located in the South. A8.4. Value-chain-based connectivity and key corridors Map A7.2 demonstrates the connective propensity of the coffee value chain. Pie-charts show the HCMC seaport complex is the main gate for coffee trading, which covers more than 50 percent of coffee imports and 90 percent of coffee exports. Considering inland coffee transportation, the main corridors used for each segment are highlighted as follows: - From planting to processing: NR13, NR14, AH17, NR19, NR28, NR26, NR1 (Ha Noi – Binh Dinh, Khanh Hoa – Ninh Thuan), PR616 (Kon Tum), NR9, AH13 (Hoa Binh – Son La), NR12B, NR10, Phap Van – Cau Gie – Ninh Binh expressway, NR32, Noi Bai – Lao Cai expressway, NR4D, NR2 - For coffee exports: NR60 (Ben Tre), NR1 (Ninh Thuan – Tien Giang), HCMC - Trung Luong expressway, NR20, AH13, AH14, NR32, Noi Bai – Lao Cai expressway, NR4D, NR2 120 Vietnam: Connecting value chains for trade competitiveness MAP A8.1. Geographic distribution of the coffee value chain Disclaimer: The boundaries, colors, denominations and other information shown on any map in this work do not imply any judgement on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. LQ Processing 1.0 - 2.0 2.1 - 5.0 5.1 - 10.0 10.1 - 14.4 LQ Planting 21.1 - 30.0 30.1 - 93.2 Top 10 Coffee firms (income) 1,000,000 10,000,000 Source: I/O Table 2016, Enterprise Census 2011 and 2016, Customs, and calculation by authors. Annex 8 – Analysis of the coffee value chain 121 MAP A8.2. Connective propensity of the coffee value chain LQ Processing 1.0 - 2.0 2.1 - 5.0 5.1 - 10.0 10.1 - 14.4 LQ Planting 21.1 - 30.0 30.1 - 93.2 Disclaimer: Top 10 Coffee firms The boundaries, colors, (income) denominations and other information shown on 1,000,000 any map in this work do 10,000,000 not imply any judgement on the part of The World Bank concerning the legal status of any territory Airport or the endorsement or acceptance of such boundaries. Border gate Seaport Processing(8II) - Export Planting(8I) - Processing(8II) Source: I/O Table 2016, Enterprise Census 2011 and 2016, Customs, and calculation by authors. 122 Vietnam: Connecting value chains for trade competitiveness Annex 9 Analysis of the fruit and vegetable value chain A9.1. Industry overview FIGURE A9.1. Employment in the fruit and Figure A9.1 shows the number of employees vegetable sector in enterprises in fruit and vegetable (F&V) cultivation and processing. Enterprise data only 80,000 Number of employees (persons) reflect employees in registered enterprises, 70,000 which is less common in agriculture. With 60,000 50,000 more than 40 percent working in the rural 40,000 sector, the actual number of practical workers 30,000 in farming should be very high. Employment 20,000 in F&V planting reflected in the Enterprise 10,000 data was just over 7,000 employees in 2011, - F&V planting F&V processing increasing to more than 16,000 in 2016. There 2011 2016 were more laborers in F&V processing, with Source: GSO Enterprises surveys, 2011, 2016. over 73,000 people in 2011 and nearly 60,000 in 2016. FIGURE A9.2. Fruit and vegetable exports 7,000,000 Fruit export turnover of Vietnam increased 6,000,000 continuously in recent years as indicated in Export value (1,000 USD) 5,000,000 Figure A9.2. During more than 10 years from 4,000,000 2006 to 2017, fruit export value increased 3,000,000 continuously, from US$ 0.5 billion in 2006, to more than US$ 6 billion in 2017. Vegetables 2,000,000 have not yet been an advantage for 1,000,000 Vietnamese exports, with an annual export of - just several hundred million (US$). 06 07 08 09 10 11 12 13 14 15 16 17 20 20 20 20 20 20 20 20 20 20 20 20 Fruit Vegetables Source: ITC Trademap. Annex 9 – Analysis of the fruit and vegetable value chain 123 A9.2. Value-chain links Inter-sectoral links for the F&V value chain are identified from the 2016 I/O table, and graphed in Figure A9.3. The links show sources for F&V processing with many inputs, including cashew nut (67%), fruits (23.2%), vegetable and bean (6.2%), etc. Because sectors in the I/O table are not disaggregated at 4- or 5-digit VSIC codes, the links show processed and preserved F&V inputs to its own sector (2.3%). The sector’s value chain is refined simply in Figure 9.4 into two segments: F&V planting (VSIC 0121, 0123, 01181, 01182) and processed and preserved products (VSIC 1030). FIGURE A9.3. Fruit and vegetable value-chain links Final Processed and preserved vegetables and fruits (IO 37) products Cashew nuts Fruits Vegetables and beans Processed and preserved Fabricated metal First (IO 10) (IO 9) of kinds (IO 6) vegetables and fruits (IO 37) products etc. (IO 76) tier SI = 67.7% SI = 23.2% SI = 6.2% SI = 2.3% SI = 2.1% Cashew Other non- Colour and Paper Fertilizers Other Fabricated Vegetables nuts perennial precious metals, and paper and nitrogen electrical metal and beans Second (IO 10) crops (IO 8) casting services products products equipment products of kinds SI = 76.7% SI = 50.1% of metals (IO 75) (IO 57) (IO 63) (IO 86) etc (IO 76) tier (IO 6) SI = 7.5% SI = 6.8% SI = 6.6% SI = 4.3% SI = 3.4% SI = 23.4% Fertilizers Fertilizers and and nitrogen nitrogen products products (IO 63) (IO 63) SI = 18.9% SI = 3.8% Manufactured goods n.e.c. (IO 97) Sl = 8.6% Pesticides and other agrochemical prod. (IO 65) Sl = 8.0% Source: I/O Table 2016, calculation by authors. 124 Vietnam: Connecting value chains for trade competitiveness FIGURE A9.4. Fruit and vegetable processing value-chain segments Cashew nuts (VSIC0123) ii. Cashew nuts (VSIC0123) Fertilizer (VSIC20120, 20210) Other non-perennial crops (VSIC0121) Fertilizer (VSIC20120, 20210) ii. Fruits iii. Processed and preserved (VSIC0121) V&F (VSIC1030) Equipment Packaging (VSIC17021) Veg. & beans of kinds (VSIC01181, 01182) ii. Veg. & beans (VSIC01181, 01182) Fertilizer (VSIC20120, 20210) Source: Authors. A9.3. Spatial structure and value-chain mapping As shown in Figure A9.5, F&V planting was most concentrated in Lam Dong and Nghe An with a labor size over 1,000 workers in those provinces. However, the concentration in Lam Dong decreased recently, reflected in the decreased LQ between 2011 and 2016. The provinces in the upper-right quadrant, including Son La, Ha Tinh, and Lang Son. had a high concentration of F&V planting and a greater LQ in 2016 than 2011. FIGURE A9.5. Locational distribution of the fruit and vegetable planting segment 70 Nghe An, 7.8 Lam Dong, 50.3 Quang Ninh, 1.7 50 Dak Lak, 3.3 Ha Giang, 3.9 Son La, 22.6 30 Bac Kan, 4.3 Dak Nong, 1.3 LQ2016 Lang Son, 6.9 Ba Ria - Vung Tau, 1.0 Ha Tinh, 11.4 10 Hoa Binh, 6.4 Can Tho, 2.2 -30 -25 -20 -15 -10 -5 0 5 10 15 20 25 -10 Cao Bang, 1.3 Kon Tum, 1.9 Soc Trang, 1.4 Lao Cai, 1.2 Hau Giang, 1.2 -30 Changing in LQ between 2016 - 2011 Circle size = number of employee 2016, Min Dak Nong = 30, Max Lam Dong = 3,800 Source: Enterprise Census 2011 and 2016 and calculation by authors. Annex 9 – Analysis of the fruit and vegetable value chain 125 Figure A9.6 shows the provincial concentration of F&V processing. Binh Phuoc had the largest labor force of more than 21,000 employees. This province also had the highest LQ in 2016, increased from 2011. Besides Binh Phuoc, other provinces with high 2016 LQs, like Lam Dong, Soc Trang, Can Tho, and Hau Giang are located in tropical climate regions, which is an advantage in F&V planting and processing. FIGURE A9.6. Locational distribution of the fruit and vegetable processing segment Hoa Binh, 1.9 70 Tra Vinh, 1.1 Hau Giang, 3.5 Tay Ninh, 2.9 50 Ninh Binh, 2.1 Phu Yen, 17.7 Long An, 5.1 Binh Phuoc, 43.1 30 LQ2016 Tien Giang, 1.7 Gia Lai, 2.5 Lam Dong, 6.5 10 Ninh Thuan, 5.8 Vinh Long, 3.2 -25 -15 -5 5 15 25 -10 Dak Nong, 2.1 An Giang, 1.7 Can Tho, 2.3 Binh Thuan, 1.4 Khanh Hoa, 1.2 Soc Trang, 1.1 -30 Changing in LQ between 2016 - 2011 Circle size = number of employee 2016, Min Soc Trang = 160, Max BinhPhuoc = 21,700 Source: Enterprise Census 2011 and 2016 and calculation by authors. Vietnam’s tropical climate produces various fruits and vegetables. Map A9.1 demonstrates the geographic distribution of the F&V value chain. As shown, F&V planting spans the country, but concentration still emerged in the North (Son La, Hoa Binh, Lang Son, Nghe An, and Ha Tinh) and Highlands (Lam Dong, Dak Lak), while F&V processing was concentrated more in the South (Binh Phuoc, Long An, Tay Ninh, Phu Yen, and Ninh Thuan, etc.). A9.4. Value-chain-based connectivity and key corridors Map A9.2. demonstrates the connective propensity of the F&V value chain. Main transportation corridors for F&V include NR91, NR1 (Lang Son – Da Nang, Khanh Hoa – Hau Giang), NR54, NR60, NR54, NR62, NRN2, NR22, NR22B, NR13, NR14, AH17, NR51, NR56, HCM – Long Thanh – Dau Giay expressway, ), HCMC - Trung Luong – My Thuan expressway, NR20, NR55, NR28, NR27, NR26, NR29, NR9, NR36, HCM road (Hoa Binh – Thanh Hoa), NR12B, AH13, Noi Bai – Lao Cai expressway, NR32, NR2, NR3, Ha Noi – Thai Nguyen expressway, AH14, NR18, NR4A, NR4B. 126 Vietnam: Connecting value chains for trade competitiveness MAP A9.1. Geographic distribution of the fruit and vegetable value chain Disclaimer: The boundaries, colors, denominations and other information shown on any map in this work do not imply any judgement on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. LQ Processing 1.0 - 2.0 2.1 - 5.0 5.1 - 10.0 10.1 - 30.0 30.1 - 43.1 LQ Planting 1.0 - 2.0 2.1 - 5.0 5.1 - 10.0 10.1 - 30.0 30.1 - 50.3 Top 10 Vegetable and fruit firms (income) 2,000,000 Source: I/O Table 2016, Enterprise Census 2011 and 2016, Customs, and calculation by authors. Annex 9 – Analysis of the fruit and vegetable value chain 127 MAP A9.2. Connective propensity of the fruit and vegetable value chain LQ Processing 1.0 - 2.0 2.1 - 5.0 5.1 - 10.0 10.1 - 30.0 30.1 - 43.1 LQ Planting 1.0 - 2.0 2.1 - 5.0 5.1 - 10.0 10.1 - 30.0 Disclaimer: 30.1 - 50.3 The boundaries, colors, denominations and other Top 10 Vegetable firms information shown on (income) any map in this work do not imply any judgement 2,000,000 on the part of The World Bank concerning the legal status of any territory or the endorsement Airport or acceptance of such boundaries. Border gate Seaport Planting(10I) - Processing(10II) Source: I/O Table 2016, Enterprise Census 2011 and 2016, Customs, and calculation by authors. 128 Vietnam: Connecting value chains for trade competitiveness Annex 10 Selected value chains, their segments, and industrial codes * Purple rows are indirect segments, white rows are direct segments in value chain. HS2007 Production chain VSIC 2018 (2-digits) (4-digits) (6-digits) (8-digits) Aquaculture Growing maize and other 01120 cereals Manufacturing starches 10612, 10620 and starch products i. Animal food i. 108003 i. 23099032 ii. Breeding ii. 03230 iii. Farming (aqua- iii. 03210, culturing) 03221, 03222 iv. Nonfarming (fishing) iv. 03110, 03121, 03122 v. Processing v. 1020 v. 0301-09, 1604, 1605 Textile and garment (T&G) Manufacturing plastics and 2013 synthetic rubber in primary forms Manufacturing of other 20290 chemical products i. Fibers (materials) i. 01160 i. 5001-03, 5101- i. 140420 05, 5201-03, 5301- 05, 5501-07 ii. Yarns ii. 13110, ii. 5004-06, 5106- 20300 10, 5205-07, 5306-08, 5402-06, 5509-11 iii. Fabric processing iii. 13120, iii. 5007, 5111-13, (weaving, knitting, 13130, 13910 5208-12, 5309-11, finishing) 5407-08, 5512-16, 6001-06 iv. Clothing iv. 1410, 1430 iv. 61, 62 v. Other garments v. 13920, v. 63 13930, 13940, 13990 Annex 10 – Selected value chains, their segments, and industrial codes 129 HS2007 Production chain VSIC 2018 (2-digits) (4-digits) (6-digits) (8-digits) Leather and footwear i. Leather i. 15110 i. 4101-07, 4112- 4114, 4115 Textiles (fabrics) 13220, 13290 Fabricated metal products 25999 Plastics 22209 ii. Leather products ii. 14200 iii. Cases and bags iii. 15120 ii. 4201, 4202, 4205 iv. Footwear iv. 15200 iii. 6401-6 Electronics i. Electronic components i. 2610, 2680 i. 8532, 8533, 8534, 8540, 8541, 8542, 8523 i. Electrical components 27330 ii. Subassemblies -- ii. 8473, 8522, 8529 851770, 900691, 900699, 900890, 844399 iii. Final products iii. 2620, 2630, iii. 8469, 8470, 851810, 851821, 2640 8471, 8472, 8519, 851822, 851829, 8521, 8525, 8527, 851830, 851840, 8528, 8443 851850, 851711, 851712, 851718, 851761, 851762, 851769, 900610, 900630, 900640, 900651, 900652, 900653, 900659 Automotive Manufacturing iron and 24100 steel Casting iron and steel 24310 Manufacturing fabricated 25999 iron and steel Manufacturing wiring 27330 devices Batteries 27200 Manufacturing other 22209 plastics products Manufacturing basic 20110 chemicals 130 Vietnam: Connecting value chains for trade competitiveness HS2007 Production chain VSIC 2018 (2-digits) (4-digits) (6-digits) (8-digits) i. Parts and components i.293 i. 8507, 8511 i. 401110, 401211, 870830, 870870, 870880, 870894, 870710, 700711, 700721, 830230, 870810, 870891, 870892, 842139, 853910, 940120, 870821, 852721, 852729, 910400, 870829, 840991, 840999, 870840, 870850, 870893, 854430, 851220, 851230, 851240, 851290 ii. Systems - modules ii. 292 ii. 8706 ii. 840733, 840734, 840820 iii. Final assembly iii. 291 iii. 8702, 8703, 8704, 8705 Wood i. Planting and foresting i. 0125, 021 ii. Logging ii. 0221 iii. Sawmilling iii. 161 iii. 4403 iv. Wood products iv. 1621, 1622, iv. 4401, 4402, iv. 940330, 940340, 1623, 1629 4404-21 940350, 940360 v. Wooden furniture v. 31001 Rice Fertilizer, pesticide 20120, 20210 i. Seeding i. 0130 ii. Planting ii. 0111 ii. 100610 iii. Rice iii. 1061 iii. 100620, 100630, 100640 Coffee i. Planting i. 0126 i. 090111, 090112 Sugar, diary 10720 Packaging 10500 Fertilizer 22201 ii. Processing ii. 107901 ii. 090121, 090122, 210111, 210112 Annex 10 – Selected value chains, their segments, and industrial codes 131 HS2007 Production chain VSIC 2018 (2-digits) (4-digits) (6-digits) (8-digits) Rubber Fertilizer, pesticide 20120 20210 i. Planting i. 0125 i. 400110 ii. Rubber processing ii. 20132 ii. 400121, 400122 ii. Manufacturing refined 19200 petroleum products iii. Manufacturing rubber 22110, 22120 products Fruit and vegetable Fertilizer, pesticide 20120, 20210 i. Planting i. 01181, 01182, 0121, 01230 Packaging 17021 ii. Processing ii. 1030 ii. 07, 08 Cement i. Quarrying stone, sand, i. 0810 i. 2505, 2521, 2508 gravel and clay ii. Manufacturing cement ii. 23941 ii. 2523 iii. Wholesale of cement iii. 46632 Fertilizer i. Coal, gas, fertilizer i. 05100, 0520, i. 2701, 2702, 2703, minerals 06200, 08910, 2711, 2510 08920, 08930 ii. Fertilizers ii. 20120 ii. 3102, 3103, 3104, 3105 Steel i. Iron ores, coke i. 071, 191 i. 2601, 2704 ii. Manufacturing iron and ii. 241, 2431 ii. 72 steel iii. Wholesale of iron and iii. 46622 steel Oil and gas i. Extracting crude i. 061, 062 i. 2709 petroleum and natural gas ii. Manufacturing refined ii. 192 ii. 2710, 2711 petroleum products iii. Wholesale iii. 46612, 46613, 46614 132 Vietnam: Connecting value chains for trade competitiveness References Farole, Thomas and Deborah Winkler. 2013. “Policy implications” in Farole, Thomas and Deborah Winkler (eds.), “Making Foreign Direct Investment Work for Sub-Saharan Africa: Local Spillovers and Competitiveness in Global Value Chains.” World Bank, Washington, DC. Farole, Thomas. 2012. “Competitiveness and Connectivity: Integrating Lagging Regions in Global Markets.” The World Bank Economic Premise 93: 1-5. World Bank, Washington, DC. Farole, Thomas. 2013. “The internal geography of trade: Lagging regions and global markets.” World Bank, Washington, DC. Frederick, Stacey. 2010. “Development and Application of a Value Chain Research Approach to Understand and Evaluate Internal and External Factors and Relationships Affecting Economic Competitiveness in the Textile Value Chain.” PhD Dissertation, North Carolina State University, Raleigh, NC. https://repository.lib.ncsu.edu/handle/1840.16/6190 Handbook of Input-Output Table Compilation and Analysis. 1999. United Nations, New York. https:// digitallibrary.un.org/record/370160?ln=en Hausmann, Tran, and Butos. 2017. “The Economic Complexity of Vietnam.” Hollweg, Claire H., Tanya Smith, and Daria Taglioni, eds. 2017. “Vietnam at a Crossroads: Engaging in the Next Generation of Global Value Chains.” World Bank, Washington, DC. International Standard Industrial Classification (ISIC), United Nations document (2008). https://unstats. un.org/unsd/publication/seriesm/seriesm_4rev4e.pdf Jeong, Hyung-Gon and Zeng, Douglas Zhihua. 2016. “Promoting Dynamic & Innovative Growth in Asia: The Cases of Special Economic Zones and Business Hubs.” November 2, 2016. KIEP Research Paper Policy Analysis-16-01. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2884483 JICA. 2009. “The Comprehensive Study on the Sustainable Development of Transport System in Vitranss2.” JICA, Tokyo. Ketels, Christian. 2017. “Cluster Mapping as a Tool for Development.” Institute for Strategy and Competitiveness, Harvard Business School. https://www.hbs.edu/faculty/Pages/item. aspx?num=53385 References 133 Krugman, Paul R. 1991. “Geography and trade.” MIT press. Ministry of Planning and Investment. 2018. “Regional Zoning Study for Regional Master Planning for 2021-2030.” National Assembly of Vietnam. “Law on Master Planning.” Law 21/2017/QH14, November 24, 2017. OECD-WTO. 2016. “Trade in value-added database.” https://stats.oecd.org/index.aspx?queryid=75537 Pham, Duc Minh, Deepak Mishra, Kee-Cheok Cheong, John Arnold, Anh Minh Trinh, Huyen Thi Ngoc Ngo, and Hien Thi Phuong Nguyen. 2013. “Trade Facilitation, Value Creation, and Competitiveness: Policy Implications for Vietnam’s Economic Growth, Volume 1.” World Bank, Hanoi. http://documents.worldbank.org/curated/en/242601468337904353/Trade-facilitation- value-creation-and-competiveness-policy-implications-for-Vietnams-economic-growth Pham, Duc Minh, Fabio Artuso, and Brian Mtonya. Dec 2018. “Facilitating Trade by Streamlining and Improving the Transparency of Non-Tariff Measures.” Taking Stock Report Part 2. World Bank, Hanoi. Pham, Duc Minh, Jung Eun Oh. June 2018. “Reform Priorities for Reducing Trade Costs and Enhancing Competitiveness in Vietnam.” Taking Stock Report Part 2. World Bank, Hanoi. Porter, Michael E. 1998. “Clusters and the new economics of competition.” Vol. 76, no. 6. Boston: Harvard Business Review. Porter, Michael E. 2000. “Location, competition, and economic development: Local clusters in a global economy.” Economic development quarterly 14, no. 1: 15-34. Prime Minister of Vietnam. 2013. “Transport Development Strategy to 2020 and the Vision to 2030.” Prime Minister’s Decision 335-QD/TTg dated February 25, 2013. Prime Minister of Vietnam. 2014. “Approving Vietnam’s industrial development strategy through 2025, with a vision toward 2035.” 879-QD/TTg dated June 9, 2014. Prime Minister of Vietnam. 2015. “Approving an integrated program to develop and upgrade value chains and clusters that are considered to be of competitive advantage “Decision 32-QD/TTg dated January 13, 2015. Ritwika Sen. 2018. “Enhancing local content in Uganda’s Oil and Gas Industry.” UNU-WIDER, working paper 2018/110. https://www.wider.unu.edu/sites/default/files/Publications/Working-paper/ PDF/wp2018-110.pdf Shakya, Mallika. 2009. “Clusters for competitiveness: A practical guide and policy implications for developing cluster initiatives.” World Bank, Washington, DC. 134 Vietnam: Connecting value chains for trade competitiveness Staritz, Cornelia and Stacey Frederick. 2014. “Sector Case Study: Apparel.” in Farole, Thomas and Deborah Winkler (editors), “Making Foreign Direct Investment Work for Sub-Saharan Africa: Local Spillovers and Competitiveness in Global Value Chains.” Directions in Development, Washington, DC: World Bank, pp. 209-244. To be published: Vietnam Development Report 2019 (VDR): “Connecting Vietnam for Growth and Shared Prosperity”. World Bank, Hanoi. Vietnam Standard Industrial Classification 2018 (VISC 2018). https://incorporation.seoul.vn/ document_business_code.pdf World Bank, 2019. Doing Business report. https://www.doingbusiness.org/en/reports/global-reports/ doing-business-2019 World Bank. 2016. “Vietnam 2035: Toward prosperity, creativity, equity, and democracy.” World Bank, Hanoi. https://openknowledge.worldbank.org/handle/10986/23724 Zeng, Douglas Zhihua, ed. 2010. “Building engines for growth and competitiveness in China: Experience with special economic zones and industrial clusters.” World Bank, Washington, DC. http://documents.worldbank.org/curated/en/294021468213279589/Building-engines- for-growth-and-competitiveness-in-China-experience-with-special-economic-zones-and- industrial-clusters References 135 HONG DUC PUBLISHING HOUSE Vietnam Laywers Association Hong Duc Publishing House • Address: 65 Trang Thi Street, Hoan Kiem District, Hanoi • Email: nhaxuatbanhongduc@yahoo.com • Tel: (84 24) 3 9260024 – Fax: (84 24) 3 9260031 Publishing responsible: Bui Viet Bac, Director Editor: Nguyen Phuong Mai Design and layout: hoanghaivuong Print 300 copies in English, size 20.5cm x 28.5cm at INSAVINA Address: 22B Hai Ba Trung, Hoan Kiem, Hanoi Publishing permit: 1033/QĐ-NXBHĐ Publishing registration plan: 4270 - 2019/CXBIPH/01 - 74/HĐ ISBN: 978-604-951-451-7 Compeleted and archived in 2019 With support from: 8 Dao Tan Street, Ba Dinh District, Hanoi, Vietnam Telephone: +84 24 37740100 Facsimile: +84 24 37740111 Website: www.dfat.gov.au 8th Floor, 63 Ly Thai To Street, Hoan Kiem District, Hanoi, Vietnam Telephone: +84 24 39346600 Facsimile: +84 24 39346597 Website: www.worldbank.org/en/country/vietnam