PRODUCTIVITY and COMPETITIVENESS of VIET NAM’S ENTERPRISES Volume 1: Manufacturing

VIETNAM ACADEMY MINISTRY OF PLANNING UNITED NATIONS OF SOCIAL SCIENCES AND INVESTMENT DEVELOPMENT PROGRAMME

PRODUCTIVITY AND COMPETITIVENESS OF VIET NAM’S ENTERPRISES Volume 1: Manufacturing

Ha Noi – 4/2019

Productivity and Competitiveness of Viet Nam’s Enterprises

Volume 1: Manufacturing

Hanoi - March 2019 Acknowledgements

This study, commissioned by the United Nations Development Programme (UNDP), was conducted by the following group of experts. Dr. Nguyen Thang (Centre for Analysis and Forecast (CAF), Viet Nam Academy of Social Sciences (VASS)) was the team leader responsible for the study’s methodology, analytical framework and consolidation of team members’ inputs, while Dr. Le Van Hung (Viet Nam Economics Institute, VASS) and Dr. Nguyen Anh Tuan (Viet Nam Productivity Institute - VNPI, Ministry of Science and Technology - MOST) provided inputs to the manufacturing section. Dr. Nguyen Do Anh Tuan and Dr. Truong Thi Thu Trang, Ms. Nguyen Thi Thuy, Ms. Nguyen Thi Thuy An, Ms. Ngo Thi Mai (Institute of Policy and Strategy for Agricultural and Rural Development (IPSARD), Ministry of Agriculture and Rural Development (MARD)) - inputs to the agriculture section and Dr. Luong Van Khoi and Dr. Tran Toan Thang (National Centre for Socio-Economic Information and Forecast, NCIF), Ministry of Planning and Investment (MPl) compiled the service section.

This report utilized literature review inputs from Dr. Nguyen Ngoc Anh (Development and Policies Research Centre, DEPOCEN), an Enterprise Census data analysis by the NCIF team and Mr. Vu Hoang Dat (CAF). This report, “Productivity and Competitiveness of Viet Nam’s enterprises - Volume 1 on manufacturing”, was prepared by Dr. Nguyen Thang and Mr. Nguyen Tien Phong (UNDP) based on the draft report prepared by Dr. Le Van Hung and additional inputs from Mr. Vu Hoang Dat (CAF) and Dr. Cengiz Cihan (UNDP).

Notably, the research team(s) received valuable comments from Prof. Rajah Rasiah (University of Malaya), Mr. Kamal Malhotra (United Nations Resident Coordinator in Viet Nam) and Ms. Caitlin Wiesen (UNDP Resident Representative in Viet Nam).

4 Productivity and Competitiveness of Viet Nam’s Enterprises Acknowledgements List of Abbreviations

This study, commissioned by the United Nations Development Programme (UNDP), was conducted by ASEAN Association of Southeast Asian Nations the following group of experts. Dr. Nguyen Thang (Centre for Analysis and Forecast (CAF), Viet Nam CAF Centre for Analysis and Forecast Academy of Social Sciences (VASS)) was the team leader responsible for the study’s methodology, analytical framework and consolidation of team members’ inputs, while Dr. Le Van Hung (Viet Nam CIEM Central Institute for Economic Management Economics Institute, VASS) and Dr. Nguyen Anh Tuan (Viet Nam Productivity Institute - VNPI, Ministry CIP Competitive Industrial Performance of Science and Technology - MOST) provided inputs to the manufacturing section. Dr. Nguyen Do Anh CPI Consumer Price Index Tuan and Dr. Truong Thi Thu Trang, Ms. Nguyen Thi Thuy, Ms. Nguyen Thi Thuy An, Ms. Ngo Thi Mai CPTTP Comprehensive and Progressive Agreement for Trans-Pacific Partnership (Institute of Policy and Strategy for Agricultural and Rural Development (IPSARD), Ministry of Agriculture and Rural Development (MARD)) - inputs to the agriculture section and Dr. Luong Van Khoi and Dr. Tran CRS Constant Returns to Scale Toan Thang (National Centre for Socio-Economic Information and Forecast, NCIF), Ministry of Planning EC Enterprise Census and Investment (MPl) compiled the service section. EU European Union This report utilized literature review inputs from Dr. Nguyen Ngoc Anh (Development and Policies FAEC Factor Allocative Efficiency Change Research Centre, DEPOCEN), an Enterprise Census data analysis by the NCIF team and Mr. Vu Hoang FDI Foreign Direct Investment Dat (CAF). This report, “Productivity and Competitiveness of Viet Nam’s enterprises - Volume 1 on manufacturing”, was prepared by Dr. Nguyen Thang and Mr. Nguyen Tien Phong (UNDP) based on the FTE Full Time Equivalent draft report prepared by Dr. Le Van Hung and additional inputs from Mr. Vu Hoang Dat (CAF) and Dr. GDP Gross Domestic Product Cengiz Cihan (UNDP). GSO General Statistics Office Notably, the research team(s) received valuable comments from Prof. Rajah Rasiah (University of Malaya), GVC Global Value Chains Mr. Kamal Malhotra (United Nations Resident Coordinator in Viet Nam) and Ms. Caitlin Wiesen (UNDP ICOR Incremental Capital Output Ratio Resident Representative in Viet Nam). ILO International Labour Organization IoT Internet of Things IPSARD Institute of Policy and Strategy for Agricultural and Rural Development IR4.0 Fourth Industrial Revolution LP Labour productivity MARD Ministry of Agriculture and Rural Development MHT Manufacturing High-Technology MOH Ministry of Health MOIT Ministry of Industry and Trade MOST Ministry of Science and Technology MPI Ministry of Planning and Investment MVA Manufacturing Value Added NCIF National Centre for Socio-Economic Information and Forecast ODA Official Development Assistance OECD Organization for Economic Co-operation and Development PPP Purchasing Power Parity RCA Revealed Comparative Advantage RoK Republic of Korea SDG Sustainable Development Goal

4 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 5 SEC Scale Efficiency Change SEDP Socio-Economic Development Plans SME Small- and Medium-sized Enterprise SOE State-owned Enterprise TE Technical Efficiency TFP Total factor Productivity TP Technical Progress UN United Nations UN Comtrade United Nations International Trade Statistics Database UNDP United Nations Development Programme UNIDO United Nations Industrial Development Organization USD United States Dollar US United States VA Value Added VASS Viet Nam Academy of Social Sciences VDMA Verband Deutscher Maschinen und Anlagenbau – (German Mechanical Engineering Industry Association) VND Viet Nam Dong VSIC Viet Nam Standard Industrial Classification WG Wage growth WTO World Trade Organization

6 Productivity and Competitiveness of Viet Nam’s Enterprises SEC Scale Efficiency Change Executive summary SEDP Socio-Economic Development Plans SME Small- and Medium-sized Enterprise SOE State-owned Enterprise TE Technical Efficiency TFP Total factor Productivity This report provides an assessment of Viet Nam’s manufacturing sector productivity and competitiveness as well as factors contributing to manufacturing labour productivity (LP) growth. It analyzed United TP Technical Progress Nations Industrial Development Organization (UNIDO) and General Statistics Office (GSO) Enterprise UN United Nations Census data using different indicators (revenue, employment, value added (VA), net exports, foreign UN Comtrade United Nations International Trade Statistics Database direct investment (FDI) backward-forward linkages with domestic firms) to describe key characteristics of manufacturing VSIC 2-digit sub-sectors, LP, Revealed Comparative Advantage (RCA), domestic UNDP United Nations Development Programme content of exports, VA-to-output ratios and wage growth to assess productivity and competitiveness UNIDO United Nations Industrial Development Organization of sub-sectors.

USD United States Dollar Based on the comprehensive analysis of manufacturing and its sub-sectors’ productivity and US United States competitiveness performances, this report recommends Viet Nam prioritize the enhancement of VA Value Added domestic private enterprises’ productivity and competitiveness during its next development stage and as integral parts of public investment, State-owned enterprise (SOE) reform and FDI policies. This VASS Viet Nam Academy of Social Sciences report also makes several specific recommendations to elevate the productivity and competitiveness of VDMA Verband Deutscher Maschinen und Anlagenbau – different sub-sectors tailored to their specific characteristics and past performances. (German Mechanical Engineering Industry Association) While the manufacturing sector experienced remarkable productivity and VND Viet Nam Dong competitiveness improvements in recent years, gaps to middle-income and VSIC Viet Nam Standard Industrial Classification developed comparator countries remain large WG Wage growth In recent years, Viet Nam’s industrial competitiveness index, manufacturing exports and RCA have WTO World Trade Organization continuously improved compared to other regional countries. In some indicators (VA-to-output ratio and RCA) Viet Nam outperformed India and Indonesia. In other manufacturing performance indicators, especially LP, Viet Nam lagged behind comparator countries with narrowing gaps still remaining large between Viet Nam and middle-income countries in the region (China, Indonesia and Malaysia) and very large to industrialized countries (Japan and the Republic of Korea (RoK)). While IR4.0 accelerates and leaves simple skilled and repetitive manufacturing jobs at risk to automation, the majority of Viet Nam’s manufacturing firms have low levels of IR4.0 readiness. Key factors contributing to higher labor productivity and IR 4.0 readiness include: firm size, capital concentration, labour skills and geographical concentration of the industries. Manufacturing sub-sectors differ by importance to the economy

Food and beverages, and furniture (medium-tech), textiles and wearing apparel, leather-footwear (low- tech) and electronics (high-tech) are sub-sectors with large economies of scale (except beverages and furniture that are medium-sized) making important contributions to the manufacturing sector and economy in terms of job creation, revenue, value addition and exports. Wood (excluding furniture), printing and tobacco as small-sized (low-tech) sub-sectors, other vehicles (high-tech) as a medium- sized sub-sector and non-metallic mineral products as a (medium-tech) large-sized sub-sector also contributed to manufacturing exports. A group of sub-sectors with high and medium technology, large and medium-sized, negative net exports (import substitution) included: (rubber-plastics, basic metal, fabricated metal), large-sized sub-sectors (chemicals, electrical equipment and motor vehicles), medium-sized sub-sectors (coke-petroleum-nuclear fuel, paper products, medical precision-optical equipment, machinery and equipment n.e.c.) and small-sized sub-sectors (repair and installation of machinery and other manufacturing).

6 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 7 FDI, SOEs and domestic private firms differ in sub-sector participation levels and sizes with limited linkages

FDI is the biggest player in manufacturing, dominating a majority of sub-sectors with large sizes and net export values, as well as in high and medium technology import substitution sub-sectors. FDI participation (measured by its VA share) in manufacturing was high in 16 out of 24 manufacturing sub-sectors and in four out of the top five manufacturing exporting sub-sectors (textiles and wearing apparel, leather-footwear, electronics and furniture). While FDI firms generate major employment, revenue and VA shares in many sub-sectors, they have weak linkages with domestic companies (especially in high- and medium-tech sub-sectors), while slightly stronger in resource-based sub- sectors.

SOEs are the smallest player in manufacturing, high profile in only two small-sized sub-sectors (coke-refined petroleum products-nuclear fuel and tobacco). Domestic private firms are the second biggest player in manufacturing, but commanding in only two sub-sectors with large positive net export values (food and beverages, and furniture). Domestic firms also dominate in large size non- metallic mineral products subsector with medium positive net export, small size wood and bamboo (excluding furniture) and printing subsectors with medium positive net export, but display medium- level participation in large size wearing apparel and textiles sectors which have positive net export. In the remaining large and medium-sized sub-sectors with positive net exports where FDI (and SOEs) dominate, domestic firms only have low participation levels. In contrast to often large-sized FDI and SOEs, domestic private firms tended to be SMEs. Productivity and competitiveness performances varied substantially across manufacturing sub-sectors Top exporting sub-sectors

Electronics: This manufacturing sub-sector had the highest VA and revenue shares, large employment share and largest positive net exports among manufacturing sub-sectors, with a high RCA. Its LP was assessed as “within reach” to comparator countries. Dominated by FDI (with a VA share of more than 98 percent), it was assessed as “competitive” in the stage of final product assembling with a “promising” increase in domestic suppliers of components as long as: (i) foreign firms maintained competitiveness of products, (ii) Viet Nam’s sub-sector LP and wages remained competitive amid the high risk of losing repetitive assembling jobs to automation and (iii) Viet Nam’s domestic firms accelerated participation as major suppliers in domestic and global value chains (GVCs). Electronics sub-sector industries that focused on assembling electronic home appliances for the domestic market (to substitute imports) also faced risks of FDI firms moving assembly plants to other countries if no longer competitive due to trade agreements. Looking forward, strengthened backward-forward linkages between FDI and domestic firms, solidified domestic firms’ linkages in GVCs and movement to higher value chain stages will need to be prioritized.

Leather-footwear, textiles and wearing apparel: Major manufacturing sub-sectors in terms of exports, employment, revenue and VA share. These subsectors are dominated by FDI (while domestic firms have significant VA and employment shares in wearing apparel) and had the lowest LP among Viet Nam’s manufacturing sub-sectors with LP gaps with comparator countries either widening or narrowing very slowly. The performance of these sub-sectors, mainly focused on the final stage of physical production in value chains (producing final products based on foreign firms’ orders), can be assessed as rather “competitive” with weakening competitiveness (also evidenced by the leather- footwear and wearing apparel subsectors’ wage growth higher than their LP growth, and especially wearing apparel’s damaging competitiveness wage growth in addition to its long struggle to move vertically up value chains). Future competitiveness depends on several factors: (i) foreign firms’ ability to maintain competitiveness of products, (ii) Viet Nam’s sub-sector LP and VA-to-output ratio’s ability to continue growing and (iii) how the high risk of losing repetitive jobs to automation will unfold. Given

8 Productivity and Competitiveness of Viet Nam’s Enterprises FDI, SOEs and domestic private firms differ in sub-sector participation levels and these sub-sectors’ large sizes and importance to Viet Nam’s economy in terms of GDP, exports and sizes with limited linkages employment, the impacts of not effectively increasing productivity and competitiveness as well as managing risks of losing jobs to automation will be highly significant for Viet Nam’s socio-economic FDI is the biggest player in manufacturing, dominating a majority of sub-sectors with large sizes and development. In the short term, however, the Comprehensive and Progressive Agreement for Trans- net export values, as well as in high and medium technology import substitution sub-sectors. FDI Pacific Partnership (CPTTP) will offer significant opportunities for leather-footwear, textiles and participation (measured by its VA share) in manufacturing was high in 16 out of 24 manufacturing wearing apparel sub-sectors for growth in the context of increased exports demand and development sub-sectors and in four out of the top five manufacturing exporting sub-sectors (textiles and wearing of local value chains, including local firms to climb on value chains. apparel, leather-footwear, electronics and furniture). While FDI firms generate major employment, revenue and VA shares in many sub-sectors, they have weak linkages with domestic companies Furniture, food and beverages: Are the fourth and fifth largest of Viet Nam’s manufacturing export (especially in high- and medium-tech sub-sectors), while slightly stronger in resource-based sub- sub-sectors, respectively. FDI firms have high participation levels in beverages and furniture with VA sectors. shares of 57 and 58.3 percent respectively, but low backward linkages with domestic firms. In these two sub-sectors, while SOEs have low participation levels, domestic private firms have medium VA SOEs are the smallest player in manufacturing, high profile in only two small-sized sub-sectors and high employment shares in beverages and rather high VA, employment and revenue shares in (coke-refined petroleum products-nuclear fuel and tobacco). Domestic private firms are the second furniture. Food processing is the only large-sized and large positive net export manufacturing sub- biggest player in manufacturing, but commanding in only two sub-sectors with large positive net sector in which domestic private firms dominate with a VA share of 62.2 percent (FDI participation is export values (food and beverages, and furniture). Domestic firms also dominate in large size non- medium level). metallic mineral products subsector with medium positive net export, small size wood and bamboo (excluding furniture) and printing subsectors with medium positive net export, but display medium- Labour productivity of food processing was at medium level, while high in beverages and low in furniture. level participation in large size wearing apparel and textiles sectors which have positive net export. The LP gaps with comparator countries were: (i) medium and narrowing in food processing, (ii) small and In the remaining large and medium-sized sub-sectors with positive net exports where FDI (and SOEs) narrowing in beverages and (iii) large and slowly narrowing in furniture. VA-to-output ratios were low and dominate, domestic firms only have low participation levels. In contrast to often large-sized FDI and gaps with comparator countries were large in food processing and beverages, while furniture featured a SOEs, domestic private firms tended to be SMEs. medium ratio and narrower gap. Wage growth was lower than LP growth in food processing, but higher (though remaining competitiveness enhancing) in beverages and furniture. Productivity and competitiveness performances varied substantially across manufacturing sub-sectors Overall, Viet Nam’s food processing sub-sector was assessed as “competitive” (comparative advantages of local agriculture commodities a key factor). Domestic private firms are big players in this Top exporting sub-sectors sub-sector, being export-oriented and aggressively expanding shares in global markets, diversifying Electronics: This manufacturing sub-sector had the highest VA and revenue shares, large employment export markets and moving up in the GVCs. To unlock further productivity and competitiveness, the share and largest positive net exports among manufacturing sub-sectors, with a high RCA. Its LP sub-sector must build capacity in branding, marketing and play leading roles in GVCs and especially was assessed as “within reach” to comparator countries. Dominated by FDI (with a VA share of more local value chains (increasing economy-of-scale for higher efficiency of agriculture production and than 98 percent), it was assessed as “competitive” in the stage of final product assembling with a ensure international quality, food safety and environmental standards, promote organic farming/ “promising” increase in domestic suppliers of components as long as: (i) foreign firms maintained green production methods and application of IR4.0 technologies). competitiveness of products, (ii) Viet Nam’s sub-sector LP and wages remained competitive amid Furniture and beverages sub-sectors were assessed as “competitive” with risks. The furniture the high risk of losing repetitive assembling jobs to automation and (iii) Viet Nam’s domestic firms sub-sector’s dependency on wood imports (while CPTPP requires higher export content from accelerated participation as major suppliers in domestic and global value chains (GVCs). Electronics origin countries) presents the key threat. FDI domination in this sub-sector, negative (though still sub-sector industries that focused on assembling electronic home appliances for the domestic competitiveness enhancing) wage-LP growth and the likelihood of simple-skilled jobs being swallowed market (to substitute imports) also faced risks of FDI firms moving assembly plants to other countries up by automation heighten the probability of FDI relocating production. In the beverages sub-sector if no longer competitive due to trade agreements. Looking forward, strengthened backward-forward FDI increased its participation level, while smaller-sized and fragmented domestic firms will face more linkages between FDI and domestic firms, solidified domestic firms’ linkages in GVCs and movement robust competition in the near future. To confront increased competition, domestic firms must build to higher value chain stages will need to be prioritized. linkages with local suppliers in domestic value chains to enhance productivity and competitiveness Leather-footwear, textiles and wearing apparel: Major manufacturing sub-sectors in terms of exports, within the next five years. employment, revenue and VA share. These subsectors are dominated by FDI (while domestic firms Other export-contributing sub-sectors have significant VA and employment shares in wearing apparel) and had the lowest LP among Viet Nam’s manufacturing sub-sectors with LP gaps with comparator countries either widening or Other vehicles: The only hi-tech (medium-sized in terms of employment, revenue and VA) sub-sector narrowing very slowly. The performance of these sub-sectors, mainly focused on the final stage of in this group is characterized by a large FDI footprint and very low backward-forward linkages with physical production in value chains (producing final products based on foreign firms’ orders), can be domestic firms. LP is high compared to other manufacturing sub-sectors in Viet Nam and “within assessed as rather “competitive” with weakening competitiveness (also evidenced by the leather- reach” of comparator countries. Its medium-level VA-to-output ratio and gaps with comparator footwear and wearing apparel subsectors’ wage growth higher than their LP growth, and especially countries have narrowed. Notably within the sub-sector, the motorcycle industry experienced a wearing apparel’s damaging competitiveness wage growth in addition to its long struggle to move significant improvement in “local” value chains and very high levels of “domestic” content, though vertically up value chains). Future competitiveness depends on several factors: (i) foreign firms’ ability possibly mainly produced by Viet Nam-based FDI firms. The sub-sector, mainly motorcycles (and to maintain competitiveness of products, (ii) Viet Nam’s sub-sector LP and VA-to-output ratio’s ability production of bicycles/parts), was assessed as “competitive” and its competitiveness could be greatly to continue growing and (iii) how the high risk of losing repetitive jobs to automation will unfold. Given enhanced through stronger linkages between FDI and Vietnamese firms to allow the latter to climb

8 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 9 local value chain. Strong linkages between Vinfast, the first leading domestic firm in electric bicycles, and domestic suppliers as well as effective domestic market competition to build capacity for future exports will be key to the success of Vinfast and Viet Nam’s electric bicycle industry.

Wood (excluding furniture), printing, tobacco and non-metallic mineral products: Small-sized and low-tech sub-sectors, aside from the later which is large-sized and medium tech, are dominated by domestic private firms (SOEs are prominent in tobacco). Gaps between Viet Nam and comparator countries in VA-to-output ratios of wood (excluding furniture) and printing were closing, while distant in tobacco and non-metallic mineral products. LP gaps with comparator countries in these four sub- sectors remain wide and slowly narrowing. Overall, these sub-sectors were assessed as having “low competitiveness”. Given their large domestic markets and presence of domestic firms, there is significant scope for policy interventions to elevate productivity and competitiveness. Tailored support to build stronger linkages between wood (excluding furniture) and local firms in the oil refinery and chemicals sub-sectors as well as suppliers of bamboo, rattan and other related materials available in Viet Nam and for sub-sector firms to upgrade technologies, branding and marketing capabilities must be explored. High and medium technology sub-sectors with negative net exports

FDI firms have high VA shares in almost all sub-sectors in this group, while domestic firms have relatively higher VA shares in chemicals and fabricated metals and significant ones in rubber-plastics and medical precision-optical equipment. SOEs only have medium employment and high VA in coke-petroleum- nuclear fuel and repair and installation of machinery and equipment. FDI in resource-based industries tended to have higher backward-forward linkages with domestic firms (backward linkages in basic metals and chemicals were 96 and 62 percent, respectively and backward-forward linkages in rubber- plastics were 25 and 24 percent, respectively). Linkages in other sub-sectors were small during 2011- 2015.

The basic metals, paper, chemicals, coke-petroleum-nuclear fuel and motor vehicles sub-sectors had high LP, with gaps to comparator countries narrowing. However, the performance measured by VA-to- output revenue showed only motor vehicles had a high ratio and reduced gaps to comparator countries, while paper, coke-petroleum-nuclear fuel, rubber-plastics, chemicals and basic metals had low/medium VA-to-output ratios and sizable gaps to other nations.

The motor vehicle sub-sector was assessed as “competitive” as long as FDI car-assembly plants in Viet Nam remained competitive. The main risks were trade agreements (including ASEAN), which could provide tax incentives for products with higher (20 percent and more) local content and fluid movement of goods encouraging FDI firms to relocate car-assembly plants if Vietnamese firms failed to become FDI suppliers. With the Vinfast car project operational in late 2018, its active presence in the domestic market and development of local value chains will elevate this sub-sector’s productivity and competitiveness.

Basic metals, paper, rubber-plastics, chemicals, coke-petroleum-nuclear fuel sub-sectors were assessed as having “medium competitiveness” if the supply of local resource-based (as well as electricity supplies for energy-intensive basic metal sub-sector) inputs continued. Key challenges included: (i) accelerating LP and VA-to-revenue ratios, (ii) improving logistics and linkages with other sub-sectors in local value chains and (iii) applying higher environmental standards and changing competition rules towards application of more environment-friendly technologies. Moving up value chains in plastics and the rubber industry moving away from exporting rubber raw materials to producing high-quality plastics and rubber products for other sub-sectors (cars, bikes, motorcycles and electronics or high-quality rubber- plastics products for healthcare) is necessary for the industry to become more competitive.

Fabricated metals, electrical equipment, other manufacturing and machinery-equipment n.e.c., medical- precision equipment and installation of machinery have medium or low LP and VA-to-output ratios, with large LP and VA-to-output ratio gaps with comparator countries. These sub-sectors were assessed as having “low competitiveness”.

10 Productivity and Competitiveness of Viet Nam’s Enterprises local value chain. Strong linkages between Vinfast, the first leading domestic firm in electric bicycles, Looking Forward, this Report Recommends and domestic suppliers as well as effective domestic market competition to build capacity for future exports will be key to the success of Vinfast and Viet Nam’s electric bicycle industry. Enhancing Vietnamese firms’ productivity and competitiveness, particularly: (i) linkages and upward movements in local and global value chains, (ii) increased productivity, value addition and (iii) local market Wood (excluding furniture), printing, tobacco and non-metallic mineral products: Small-sized shares and especially export volumes and values must be achieved through implementing a wide range and low-tech sub-sectors, aside from the later which is large-sized and medium tech, are dominated of integrated and concerted policy actions. This should be a common prioritized goal within a wide range by domestic private firms (SOEs are prominent in tobacco). Gaps between Viet Nam and comparator of national policies, from industrialization, SOE reforms and private sector/enterprise development to countries in VA-to-output ratios of wood (excluding furniture) and printing were closing, while distant trade, FDI attraction, R&D, skills training and public investment. As lower middle-income Viet Nam, in its in tobacco and non-metallic mineral products. LP gaps with comparator countries in these four sub- next stage of development, pursues an inclusive growth pathway to generate more productive jobs for sectors remain wide and slowly narrowing. Overall, these sub-sectors were assessed as having “low its workers, enhance productivity and competitiveness, Vietnamese manufacturing firms must be at the competitiveness”. Given their large domestic markets and presence of domestic firms, there is significant centre of Viet Nam’s growth strategy. scope for policy interventions to elevate productivity and competitiveness. Tailored support to build stronger linkages between wood (excluding furniture) and local firms in the oil refinery and chemicals Urgent action is required to address well-known weaknesses of limited linkages between trade sub-sectors as well as suppliers of bamboo, rattan and other related materials available in Viet Nam and negotiations, industrial policies and programmes supporting enterprise development. Consultation for sub-sector firms to upgrade technologies, branding and marketing capabilities must be explored. and engagement of domestic private firms in formulation and implementation of such policies and programmes are paramount. High and medium technology sub-sectors with negative net exports SOE reform efforts must focus on enhancing SOEs’ (especially those at higher stages or leading local FDI firms have high VA shares in almost all sub-sectors in this group, while domestic firms have relatively value chains with relatively high productivity and competitiveness) effectiveness and efficiencies, plus higher VA shares in chemicals and fabricated metals and significant ones in rubber-plastics and medical backward-forward linkages with domestic firms. Equitization of SOEs in industries/sub-sectors where precision-optical equipment. SOEs only have medium employment and high VA in coke-petroleum- domestic private firms are ready to take over leading roles from SOEs (with equitization going hand-in- nuclear fuel and repair and installation of machinery and equipment. FDI in resource-based industries hand with building capacity of domestic private firms) should be accelerated. tended to have higher backward-forward linkages with domestic firms (backward linkages in basic metals and chemicals were 96 and 62 percent, respectively and backward-forward linkages in rubber- Public investment should aim to crowd-in domestic private sector investment. Public investment could plastics were 25 and 24 percent, respectively). Linkages in other sub-sectors were small during 2011- help create demand for domestic private firms’ products and services, and thus incentives to invest in 2015. business and technology upgrades. Public investment could also provide domestic private firms with opportunities to build capacity, including through learning-by-doing and receiving technology transfers The basic metals, paper, chemicals, coke-petroleum-nuclear fuel and motor vehicles sub-sectors had through public investment from ODA loan projects. high LP, with gaps to comparator countries narrowing. However, the performance measured by VA-to- output revenue showed only motor vehicles had a high ratio and reduced gaps to comparator countries, Public investment in development of IT/telecommunications infrastructure (cloud computing, network while paper, coke-petroleum-nuclear fuel, rubber-plastics, chemicals and basic metals had low/medium and data security as well as e-commerce platforms, including for intermediate goods), e-payments and VA-to-output ratios and sizable gaps to other nations. e-banking (similar to e-tax, e-customs and e-government payments), will not only help private firms (especially SMEs) improve IR4.0 readiness and efficiency, but also value chain connections. While public The motor vehicle sub-sector was assessed as “competitive” as long as FDI car-assembly plants in services (R&D and skills training) will benefit SMEs in general, special services such as testing and Viet Nam remained competitive. The main risks were trade agreements (including ASEAN), which could certification (and perhaps radiation and cold storage) have strong potential to enhance access and provide tax incentives for products with higher (20 percent and more) local content and fluid movement competitiveness of food and agriculture processing SMEs in global markets. of goods encouraging FDI firms to relocate car-assembly plants if Vietnamese firms failed to become FDI suppliers. With the Vinfast car project operational in late 2018, its active presence in the domestic market Within the overarching goals of sustainable development and creation of productive jobs for Vietnamese and development of local value chains will elevate this sub-sector’s productivity and competitiveness. workers, Viet Nam must shift its focus from quantity to attraction of higher-quality FDI and bridge gaps between FDI and domestic private firms. The most important element of higher-quality FDI firms is long- Basic metals, paper, rubber-plastics, chemicals, coke-petroleum-nuclear fuel sub-sectors were assessed term partnerships with local firms (as key players in production chains) as the core of their international as having “medium competitiveness” if the supply of local resource-based (as well as electricity supplies competition strategy. Government should cooperate with such FDI firms, in a “win-win” approach, to for energy-intensive basic metal sub-sector) inputs continued. Key challenges included: (i) accelerating support capacity development of domestic firms to benefit from technology transfers and connect LP and VA-to-revenue ratios, (ii) improving logistics and linkages with other sub-sectors in local value them as first-tier suppliers to leading FDI firms and GVCs. chains and (iii) applying higher environmental standards and changing competition rules towards application of more environment-friendly technologies. Moving up value chains in plastics and the Domestic sector development will need to be prioritized in the upcoming Socio-Economic Development rubber industry moving away from exporting rubber raw materials to producing high-quality plastics and Strategy (2021-2030) and Plan (2021-2025). Key policy objectives should support domestic private firms rubber products for other sub-sectors (cars, bikes, motorcycles and electronics or high-quality rubber- to grow in size, accelerate transition to formalization and enhance productivity and competitiveness plastics products for healthcare) is necessary for the industry to become more competitive. through development of local value chains, improved linkages within and upward movement in domestic and global value chains, with special attention to facilitate the emergence of big domestic firms to Fabricated metals, electrical equipment, other manufacturing and machinery-equipment n.e.c., medical- lead domestic value chains and become significant GVC players. In addition to continued efforts to precision equipment and installation of machinery have medium or low LP and VA-to-output ratios, with nuance the business environment and support domestic private enterprises to access land and credit, large LP and VA-to-output ratio gaps with comparator countries. These sub-sectors were assessed as more tailored support is needed for SMEs to elevate their: (a) capacity for business management and having “low competitiveness”. marketing, (b) linkages in domestic and international value chains and (c) technical capacity to adopt new

10 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 11 technologies and readiness to grasp opportunities unlocked by IR4.0 and a new wave of higher quality FDI. Establishment of independent (para-governmental) institutions specialized in providing training and R&D support to Vietnamese firms is necessary. Besides access to credit, guidance on technology upgrades is needed for domestic private firms to improve integration and generate upward movement in local and GVCs.

The assessment of productivity and competitiveness performance of the manufacturing sector and its sub-sectors presented in this report identified challenges and opportunities for Viet Nam to greatly enhance domestic firms’ competitive edge through capturing more value added from a bigger local and global value chain footprint. With the enhancement of domestic firms’ productivity and competitiveness as a central tenet in Viet Nam’s growth strategy for its next development stage, it is the time for interlinked policy actions to be formulated and implemented within an integrated policy framework with concerted efforts by different stakeholders from government and business sectors.

12 Productivity and Competitiveness of Viet Nam’s Enterprises technologies and readiness to grasp opportunities unlocked by IR4.0 and a new wave of higher quality FDI. Establishment of independent (para-governmental) institutions specialized in providing training Contents and R&D support to Vietnamese firms is necessary. Besides access to credit, guidance on technology upgrades is needed for domestic private firms to improve integration and generate upward movement in local and GVCs.

The assessment of productivity and competitiveness performance of the manufacturing sector and its Acknowledgements ...... 4 sub-sectors presented in this report identified challenges and opportunities for Viet Nam to greatly enhance domestic firms’ competitive edge through capturing more value added from a bigger local and List of Abbreviations ...... 5 global value chain footprint. With the enhancement of domestic firms’ productivity and competitiveness as a central tenet in Viet Nam’s growth strategy for its next development stage, it is the time for interlinked Executive summary ...... 7 policy actions to be formulated and implemented within an integrated policy framework with concerted Introduction ...... 18 efforts by different stakeholders from government and business sectors. 1. Labour Productivity at the Economy-Wide Level ...... 20 1.1. Viet Nam’s Macroeconomic context: Key highlights ...... 20 1.2. Labour Productivity Performance ...... 23 1.3. Sources of Labour Productivity Growth: a Shift Share Analysis ...... 25 1.4. Determinants of Labour Productivity at Firm Level ...... 27

2. Productivity and Competitiveness at Sector and Sub-sector Levels ...... 31 2.1. Measurements to assess productivity and competitiveness of manufacturing sector and its sub-sectors ...... 31 2.2. Productivity and Competitiveness of Manufacturing ...... 32 5. Conclusions ...... 89 References ...... 90

Appendix 1. Technical Notes on Productivity Analysis ...... 92 Appendix 2. NACE Industrial Classification ...... 99 Appendix 3. Results of Econometric Analysis of Enterprise Census 2012 and 2017 ...... 102 Appendix 4. List of Manufacturing Sub-sectors and VSIC codes ...... 109

12 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 13 List of Figures, tables, boxes

List of Figures

Figure 1.1: Economic growth rate, 1991-2016 (%) ...... 20 Figure 1.2: Investment rate, 1991-2016: international comparison (%) ...... 20 Figure 1.3: Viet Nam’s trade performance, 1995-2017 (USD billion) ...... 21 Figure 1.4: Trade balance, 2011-2017 (USD billion) ...... 21 Figure 1.5: Labour productivity in selected countries, 1991-2016: Average annual growth rate (%)...... 24 Figure 1.6: Structural transformation in Viet Nam during 1991-2016 ...... 26 Figure 1.7: Sectoral labour productivity in Viet Nam during 1991-2016 ...... 26 Figure 1.8: Sources of labour productivity growth (%) ...... 27 Figure 1.9: Size Premium ...... 28 Figure 1.10: Labour productivity of firms in sectors relative to ones in low-tech manufacturing ...... 29 Figure 1.11; Ownership and productivity (reference: private firms) ...... 30 Figure 1.12: Labour productivity of firms in different regions (reference: northern central and central coast)...... 30 Figure 2.1: MVA as a percentage of GDP ...... 33 Figure 2.2a: Global MVA share of selected countries (%) ...... 33 Figure 2.2b: Manufacturing VA per capita (USD) ...... 34 Figure 2.3: CIP rankings ...... 34 Figure 2.4: Viet Nam’s manufacturing LP as a share of other countries’ manufacturing LP ...... 35 Figure 2.5: Manufacturing VA-to-output ratio ...... 35 Figure 2.6: Manufacturing RCA of Viet Nam and comparator countries ...... 37 Figure 2.7: Proportions of SOEs, domestic private and FDI firms at different IR4.0 readiness levels ....39 Figure 2.8: Readiness scores by size of firms ...... 39 Figure 2.9: Level of enterprises applying IR4.0 technologies ...... 39 Figure 2.10: Employment size of manufacturing sub-sectors ...... 41 Figure 2.11: Sub-sector shares in manufacturing employment (%)...... 41 Figure 2.12: Sub-sector employment structure by ownership (2016) ...... 42 Figure 2.13: Large and small firm shares in sub-sector employment (%), 2016 (Source: Authors’ calculation based on 2017 Enterprise Census data) ...... 43 Figure 2.14: Revenue of manufacturing sub-sectors (VND billion, 2016 prices)t ...... 44 Figure 2.15:: Sub-sector revenue shares in manufacturing revenue (%) ...... 44 Figure 2.16: (upper panel): Domestic private firms, SOEs and FDI shares in sub-sector revenue (%), 2016 ...... 45 Figure 2.16: (lower panel): Large and SME shares in sub-sector revenue (%), 2016 ...... 45 Figure 2.17: (upper panel): Value added of manufacturing sub-sector, VND billion (2016 prices) ...... 46 Figure 2.17: (lower panel): Sub-sector share in MVA ...... 46 Figure 2.18: FDI, SOE and domestic firms’ shares in sub-sector VA (%) 2016 ...... 47

14 Productivity and Competitiveness of Viet Nam’s Enterprises Figure 2.19: Large firms and SME shares of sub-sector VA (%), 2016 ...... 47 List of Figures, tables, boxes |Figure 2.20: Sub-sectors’ export volume (USD million, current price) ...... 48 Figure 2.21: Manufacturing sub-sector net exports (USD million, current prices) ...... 48 Figure 2.22: Proportion (%) of enterprises’ IR4.0 readiness levels by sub-sector ...... 49 Figure 2.23: Firms’ usage of cloud computing by sub-sectors (%) ...... 49 List of Figures Figure 2.24: Backward-linkages between FDI and domestic firms in Viet Nam’s manufacturing industry (%, max=100%) ...... 50 Figure 1.1: Economic growth rate, 1991-2016 (%) ...... 20 Figure 2.25: Forward-linkages between FDI and domestic firms Figure 1.2: Investment rate, 1991-2016: international comparison (%) ...... 20 in Viet Nam’s manufacturing industry (%, max = 100%) ...... 51 Figure 1.3: Viet Nam’s trade performance, 1995-2017 (USD billion) ...... 21 Figure 2.26: Labour productivity of manufacturing sub-sectors (VND million, 2016 prices) ...... 55 Figure 1.4: Trade balance, 2011-2017 (USD billion) ...... 21 Figure 2.27a: Relative labour productivity (manufacturing = 1) ...... 55 Figure 1.5: Labour productivity in selected countries, 1991-2016: Average annual growth rate (%)...... 24 Figure 2.27b: Relative labour productivity (manufacturing = 1), 2015 (source: Figure 1.6: Structural transformation in Viet Nam during 1991-2016 ...... 26 Authors’ calculation using UNIDO database) ...... 56 Figure 1.7: Sectoral labour productivity in Viet Nam during 1991-2016 ...... 26 Figure 2.28: Productivity level (2011), growth rate and average employment Figure 1.8: Sources of labour productivity growth (%) ...... 27 size (2011-2016) of manufacturing sub-sectors ...... 57 Figure 1.9: Size Premium ...... 28 Figure 2.29: SOE, FDI and domestic private firms’ average labour productivity by sub-sector Figure 1.10: Labour productivity of firms in sectors relative to ones in low-tech manufacturing ...... 29 (VND million, 2016 prices) ...... 57 Figure 1.11; Ownership and productivity (reference: private firms) ...... 30 Figure 2.30: Labour productivity level in 2011 and annual growth rate in 2011-2016 (%) ...... 58 Figure 1.12: Labour productivity of firms in different regions (reference: northern central Figure 2.31: Decomposition of labour productivity growth in manufacturing (2011-2016) ...... 59 and central coast)...... 30 Figure 2.32: Size premium ...... 60 Figure 2.1: MVA as a percentage of GDP ...... 33 Figure 2.33: Internet usage and labour productivity ...... 61 Figure 2.2a: Global MVA share of selected countries (%) ...... 33 Figure 2.34: Labour productivity of firms in high- and medium-tech sectors relative Figure 2.2b: Manufacturing VA per capita (USD) ...... 34 to firms in low-tech manufacturing ...... 61 Figure 2.3: CIP rankings ...... 34 Figure 2.35: Ownership and productivity of companies with different ownership forms (reference: Figure 2.4: Viet Nam’s manufacturing LP as a share of other countries’ manufacturing LP ...... 35 private companies) ...... 62 Figure 2.5: Manufacturing VA-to-output ratio ...... 35 Figure 2.36: Labour productivity of firms in different regions (reference: Ho Chi Minh City)...... 62 Figure 2.6: Manufacturing RCA of Viet Nam and comparator countries ...... 37 Figure 2.38: Sub-sectors’ RCA of Viet Nam and comparator countries ...... 65 Figure 2.7: Proportions of SOEs, domestic private and FDI firms at different IR4.0 readiness levels ....39 Figure 2.40: Value added-to-revenue ratio (%), 2011 and 2015 ...... 68 Figure 2.8: Readiness scores by size of firms ...... 39 Figure 2.41.: 2016 Value added-to-revenue ratio (%), using Enterprise Census data ...... 69 Figure 2.9: Level of enterprises applying IR4.0 technologies ...... 39 Figure 2.42a: Value added-to-revenue ratio by ownership (2016) ...... 69 Figure 2.10: Employment size of manufacturing sub-sectors ...... 41 Figure 2.42b: Value added-to-revenue ratio by firm size (2016) ...... 70 Figure 2.11: Sub-sector shares in manufacturing employment (%)...... 41 Figure 2.43: Comparator countries’ VA-to-output ratios as percentage of Viet Nam’s ...... 70 Figure 2.12: Sub-sector employment structure by ownership (2016) ...... 42 Figure 2.44: Regional minimum wages, CPI and GDP per capita in Viet Nam, 2009-2016 (2008=100) ...71 Figure 2.13: Large and small firm shares in sub-sector employment (%), 2016 (Source: Figure 2.45: Viet Nam’s productivity (LPG) and wage growth in manufacturing, 2011-2016 (%) ...... 72 Authors’ calculation based on 2017 Enterprise Census data) ...... 43 Figure 2.46: Labour productivity of food processing industries ...... 82 Figure 2.14: Revenue of manufacturing sub-sectors (VND billion, 2016 prices)t ...... 44 Figure 2.15:: Sub-sector revenue shares in manufacturing revenue (%) ...... 44 Figure 2.16: (upper panel): Domestic private firms, SOEs and FDI shares in sub-sector revenue (%), 2016 ...... 45 Figure 2.16: (lower panel): Large and SME shares in sub-sector revenue (%), 2016 ...... 45 Figure 2.17: (upper panel): Value added of manufacturing sub-sector, VND billion (2016 prices) ...... 46 Figure 2.17: (lower panel): Sub-sector share in MVA ...... 46 Figure 2.18: FDI, SOE and domestic firms’ shares in sub-sector VA (%) 2016 ...... 47

14 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 15 List of Tables

Table 1.1: Number of years required for Viet Nam to catch regional countries in labour productivity ..25 Table 2.1: Decomposition of TFP ...... 36 Table 2.2: Firms’ usage of cloud computing by size and ownership (%) ...... 40 Table 2.3: Summary of key sub-sector characteristics ...... 52 Table 2.4: Labour productivity of Viet Nam relative to comparator countries (%) ...... 54 Table 2.5: Measures of variation in labour productivity (2011 and 2016) ...... 58 Table 2.7: Productivity and wage growth and competitiveness in selected manufacturing sub-sectors..... 73 Table 2.8: Summary of manufacturing sub-sector characteristics and performance ...... 74 List of Tables List of Boxes

Table 1.1: Number of years required for Viet Nam to catch regional countries in labour productivity ..25 Box 1. 1: How long does Viet Nam need to fully close absolute labour productivity gaps Table 2.1: Decomposition of TFP ...... 36 with ASEAN countries? ...... 24 Table 2.2: Firms’ usage of cloud computing by size and ownership (%) ...... 40 Box 2.1: Decomposing Total Factor Productivity: A stochastic frontier approach in measuring Table 2.3: Summary of key sub-sector characteristics ...... 52 performance of Viet Nam’s manufacturing sector ...... 36 Table 2.4: Labour productivity of Viet Nam relative to comparator countries (%) ...... 54 Box 2.2: RCA as a measurement of international competitiveness ...... 38 Table 2.5: Measures of variation in labour productivity (2011 and 2016) ...... 58 Box 2.3: Firms’ size and ownership ...... 43 Table 2.7: Productivity and wage growth and competitiveness in selected manufacturing sub-sectors..... 73 Box 2.4: Relative labour productivity using UNIDO 2015 data ...... 56 Table 2.8: Summary of manufacturing sub-sector characteristics and performance ...... 74 Box 2.5: LP growth and employment size of sub-sectors ...... 56 BOX 2.6: Value added-to-revenue ratio calculated using Enterprise Census data ...... 69 Box 2.7: Signs of strengthening linkages between Samsung and domestic firms ...... 77 Box 2.9: Deeper look at food processing industries’ labour productivity ...... 82

Volume 1: Manufacturing 17 Introduction

Viet Nam achieved remarkable economic growth during 1991-2018, exceeding 6.8 percent. Per capita income increased 22-fold, from less than USD100 in the late 1980s to USD2,356 in 2017. During this time, Viet Nam passed the World Bank’s USD1,000 income per capita threshold in 2008 to join the rank of lower middle-income countries. Despite these positive developments, it is widely acknowledged that Viet Nam’s current growth model relies too heavily on cheap labour and the exploitation of natural resources.

The challenge for lower middle-income Viet Nam to ensure more inclusive and sustainable growth is to transition to a new growth model based on rapid productivity growth, innovation, high-value addition and enhanced internationally competitiveness to provide more productive employment for the majority of Vietnamese people. Recognizing this challenge, Viet Nam’s Socio-Economic Development Strategy (2010-2020) and five-year Socio-Economic Development Plans (SEDPs), notably 2010-2015 and 2016-2020, have highlighted the importance of industrialization as well as increased productivity and competitiveness. Among the 2016-2020 SEDP’s nine economic targets, two are to increase the contribution of total factor productivity (TFP) to growth and achieve 5 percent average annual increases in labour productivity.

A wide range of policies and programmes (SOE reforms, an enhanced environment for start-ups and to do business, support of SMEs to access credit and improve the quality of human resources) has been and are being formulated and implemented. Despite these policies, improvements in productivity and competitiveness remain below policy-makers’ expectations with the weak implementation a key reason often cited. While this may be true, it is also widely acknowledged that weak policy designs (not accounting for the needs of enterprises in different sectors/sub-sectors and sizes, for example), political economy, incentives and capacity for implementation and institutional/cross-sector coordination aspects may have also contributed to the limited results.

One of the negative byproducts of weak policy design and to a lesser extent implementation, is the paucity of in-depth knowledge of key bottlenecks that impede productivity and competitiveness in sectors, especially the sub-sector/cross-sector and enterprise levels. While some research is available on TFP/productivity and competitiveness at economy-wide and some sector levels, it was based on the application of different methodologies and data sources that encountered numerous data and measurement consistency issues.

These limitations hamper the formulation of concrete and tailored policies and actions, as opposed to a one-size-fits-all approach, to address bottlenecks in sub-sectors and enterprises and support them to enhance productivity and competitiveness. This study sets out to close some of these knowledge gaps through utilizing the latest macro and micro-level data available from enterprise census, labour force and household surveys. It analyzes data within a coherent framework and uses a consistent set of productivity and competitiveness measurements cognizant of the data’s limited scope.

The first part of this report briefly analyzes, within Viet Nam’s macroeconomic context, the country’s labour productivity performance, its sources and factors contributing to growth at economy-wide level. The second part assesses productivity and competitiveness in more detail at sectoral (manufacturing, service and agriculture1) and (2-digit VSIC) sub-sector levels. In this first volume, the report focuses on manufacturing and its sub-sectors, while an assessment of the service and agriculture sectors and

1 This report did not assess the construction and mining sub-sectors, important to Viet Nam’s economy, due to data shortcomings as well as the non-tradable nature of these sub-sectors. Thus, they are less relevant to any assessment of international competitiveness.

18 Productivity and Competitiveness of Viet Nam’s Enterprises respective sub-sectors is delivered in volume two of this report. By examining the productivity and Introduction competitiveness performance of enterprises at sector and sub-sector levels, this report will provide technical notes on measurements and data used for such assessments. Importantly, a summary of assessments and policy recommendations will be provided in the last section of this report.

18 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 19 1. Labour Productivity at the Economy-Wide Level 1.1. Viet Nam’s Macroeconomic context: Key highlights Economic growth has been bottoming down

Figure 1.1: Economic growth rate, 1991-2016 (%)

9.00 WorldThế giới 8.00 8.21

7.00 6.96 6.90 6.32 6.26.2 6.00 5.925.92 VietnamViệt Nam 5.00 4.00 LowCác incomenước 3.00 countriesthu nhập thấp 2.00

1.00 LowerCác n ưmiddleớc incomethu nhậ countriesp 0.00 trung bình thấp 1991-1995 1996-2000 2001-2005 2006-20102006-2010 2011-20152011-2015 20162016

Source: Authors’ calculation from World Bank’s World Development Indicators

Figure 1.1 highlights economic growth in Viet Nam and low-income, lower middle-income countries and globally over a 25-year period (1991-2016). It shows that Viet Nam outperformed these comparator countries from 1991 to 2006. However, Viet Nam’s growth rate declined and as a consequence, it was no longer superior to the low-income and lower middle-income groups during the ensuing period (2007- 2015). Since 2016, Viet Nam’s GDP growth has shown signs of iut, but it is too early to determine whether the country has resumed its robust growth performance over comparator countries. The relatively high and increasing investment-to-GDP ratio reversed after Viet Nam reached lower middle-income status

Viet Nam’s growth performance is arguably attributed to high investment rates sustained over an extended period of time. Figure 1.2 shows that during 1996-2010, the ratio of investment to GDP in Viet Nam was consistently and significantly higher than low- and middle-income countries. However, this rate subsequently fell to a level similar to comparator countries, reflecting significant changes that brought this important index to a more sustainable level. Specifically, the savings-investment gap in Viet Nam changed from -6 percent of GDP during 2006-2010 to +2.32 percent of GDP in 2011-2015 and +2.40 percent in 2016. This was in part attributed to improvements in efficiency of capital use, as evidenced by a reduction in the incremental capital output ratio (ICOR) from more than 6 in the late 2000s to approximately 5 in recent years.

Figure 1.2: Investment rate, 1991-2016: international comparison (%)

40.040.0 Các nước 36.690536.6905 Low income 35.035.0 thucountries nhập thấp 31.748231.7482 30.030.0 Các nước 28.1040 28.1040 27.640527.6405 26.600 thu nhập 26.600 trLowerung bì middlenh thấp 25.025.0 income countries 21.917221.9172 20.020.0 Việt Nam 15.015.0 Vietnam Thế giới 10.010.0

5.05.0 World .0.0 1991-19951991-1995 1996-20001996-2000 2001-20052001-2005 2006-20102006-2010 2011-20152011-2015 20162016

Source: Authors’ calculation from World Bank’s World Development Indicators

20 Productivity and Competitiveness of Viet Nam’s Enterprises 1. Labour Productivity at the Economy-Wide Level Trade performance improved with exports and imports rapidly rising, while the trade balance turned positive in recent years 1.1. Viet Nam’s Macroeconomic context: Key highlights The value of Viet Nam’s exports and imports increased each year during 1995-2017, except for 2009. These Economic growth has been bottoming down figures highlight Viet Nam’s strong trade performance, with exports and imports having accelerated over the last decade (Figure 1.3). Figure 1.1: Economic growth rate, 1991-2016 (%) 9.00 Figure 1.3: Viet Nam’s trade performance, 1995-2017 (USD billion) 8.21 Thế giới 8.00 250 2114.01010 7.00 6.96 6.90 2111.1.110 6.32 6.2 200 6.00 5.92 Việt Nam 200 1175.90 1173.30 166.1164 150.1861162.053 5.00 148.049 150 132.033 4.00 132.103332.033 114.529 Các nước 106.750113.780 3.00 thu nhập thấp 96.9.906 100 84.839 100 80.714 84.839 69.9.949 72.2.237 62.765 62.685 2.00 57.0.096 48.561 44.891 48.561 36.761 39.826 50 31.96936.761 1.00 Các nước 31.969 32.447 25.256 26.485 thu nhập 15.63716.218 19.746 20.149 11.144 11.592 11.63011.71412.514511.643.478136.218 8.155 11.17.44256 9.185 9.360 14.483 15.02916.706 0.00 trung bình thấp 5.449 1991-1995 1996-2000 2001-2005 2006-2010 2011-2015 2016 - 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 Source: Authors’ calculation from World Bank’s World Development Indicators NhậpImport khẩu XuấtExport khẩu

Figure 1.1 highlights economic growth in Viet Nam and low-income, lower middle-income countries Source: GSO and globally over a 25-year period (1991-2016). It shows that Viet Nam outperformed these comparator A positive trend also emerged with respect to the trade balance in recent years. If imports increased countries from 1991 to 2006. However, Viet Nam’s growth rate declined and as a consequence, it was no faster than exports during 2000-2008 and resulted in a larger trade deficit, from 2009 until now the longer superior to the low-income and lower middle-income groups during the ensuing period (2007- trade deficit has gradually decreased to move into a surplus in recent years, except in 2015. Such a 2015). Since 2016, Viet Nam’s GDP growth has shown signs of iut, but it is too early to determine whether positive change was largely driven by the FDI’s strong trade performance, while the domestic sector the country has resumed its robust growth performance over comparator countries. persistently ran a large trade deficit (Figure 1.4).

The relatively high and increasing investment-to-GDP ratio reversed after Viet Nam Figure 1.4: Trade balance, 2011-2017 (USD billion) reached lower middle-income status 030030 004 2.600 2.9100 Viet Nam’s growth performance is arguably attributed to high investment rates sustained over an 2.1378 002 020020 .700.700 extended period of time. Figure 1.2 shows that during 1996-2010, the ratio of investment to GDP in Viet .00.00 - 25.8100 Nam was consistently and significantly higher than low- and middle-income countries. However, this 010010 21.6500 13.3400 (002) 9.7400 13.3400 rate subsequently fell to a level similar to comparator countries, reflecting significant changes that 4.100 6.500 -- --.900.900 -4.1113 (004) brought this important index to a more sustainable level. Specifically, the savings-investment gap in Viet --3.400 --6.500 --8.900 -7.6022 (006) Nam changed from -6 percent of GDP during 2006-2010 to +2.32 percent of GDP in 2011-2015 and +2.40 -17.4513 -19.0500 -010-010 -22.900 percent in 2016. This was in part attributed to improvements in efficiency of capital use, as evidenced (008) -020-020 by a reduction in the incremental capital output ratio (ICOR) from more than 6 in the late 2000s to --9.800 (010) approximately 5 in recent years. -030-030 (012) 20112011 20122012 20132013 2014 2015 2016 2017 Figure 1.2: Investment rate, 1991-2016: international comparison (%) FDIFDI (USD billion) DomesticDoanh nghiệp sector trong (USD nước billion) NetXuấtkhẩu export ròng 40.0 Các nước 36.6905 Source: GSO 35.0 thu nhập thấp 31.7482 30.0 Các nước FDI is an increasingly important source of investment 28.1040 27.6405 26.600 thu nhập 25.0 trung bình thấp 21.9172 The volume of FDI inflows into Viet Nam accelerated in a stable manner, especially after World Trade 20.0 Việt Nam Organization (WTO) accession, to stand at a relatively high level in ASEAN - only below Indonesia and 15.0 Singapore in 2015. The FDI share of total investment in Viet Nam reached a record level of 30.9 percent in 10.0 Thế giới 2008, and firmed up again to around 23.4 percent in 2015 and 2016. The FDI sector has made increasing 5.0 contributions to account for around 20 percent of the country’s GDP (from 15.2 percent in 2005), 72 .0 percent of Viet Nam’s exports (from 57 percent in 2005), 18 percent of government revenue and creating 1991-1995 1996-2000 2001-2005 2006-2010 2011-2015 2016 3.7 million jobs for Vietnamese workers in 2017.

Source: Authors’ calculation from World Bank’s World Development Indicators

20 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 21 FDI in manufacturing accounts for almost 70 percent of total FDI into Viet Nam, the largest proportion of FDI in manufacturing in ASEAN followed by Indonesia (40 percent) and the Philippines (38 percent). GSO data shows that 64.6 percent of newly-registered FDI capital was in manufacturing, 10.1 percent in real estate and 2.4 percent in wholesale, retail and car-motorbike repairs in 2016. In 2017, electricity production and distribution, gas and air-conditioning attracted the most newly-registered FDI with USD8.4 billion (42.3 percent of total) and manufacturing only USD6.3 billion (31.7 percent).

The sources of FDI inflows into Viet Nam were diversified with Japan, RoK and Singapore the top investors out of 68 countries and territories. Job vulnerability to automation and AI

Technological progress has accelerated in the recent decade, which is termed by numerous commentators as the Fourth Industrial Revolution (IR4.0). It is changing the global economic landscape, with far reaching implications for all actors – government, businesses and society at large. Among the key concerns of IR4.0 are (simple skilled/repetitive) job loses to automation and Artificial Intelligence (AI).

The report “The future of jobs at risks of automation”2 (ILO 2016) estimated that in the next few decades among ASEAN-5 (Cambodia, Indonesia, the Philippines, Thailand and Viet Nam), the share of jobs with a high probability of automation was lowest in Thailand (44 percent) and highest in Viet Nam (70 percent). In the Philippines, Indonesia and Cambodia, the shares were 49, 56 and 57 percent, respectively (although some jobs, such as in agriculture, will be affected by structural challenges beyond mechanization). Within the average of 70.4 percent of all jobs in Viet Nam’s economy assessed at risk of automation, some sectors/sub-sectors with very high proportions of jobs with a high probability of automation risks include agriculture, forestry and fisheries (83.3 percent), manufacturing (74.4 percent), food and beverages (68 percent), garments (85 percent) and electronics (75 percent), wholesale, retail and repair of motor vehicles (84.1 percent), service sector (32 percent), retail (70 percent), hotel and banking (around 40 percent). Occupations with the highest risk of disappearing in the future include shop sales assistants (2.1 million), garden labourers (1 million) and sewing-machine operators in garment manufacturing (770,000). Oxford Economics and Cisco’s September 2018 report “Technology and the future of ASEAN jobs - The impact of AI on workers in ASEAN’s six largest economies” (Oxford Economics and Cisco, 2018) forecasted that by 2028, “displaced workers” would number 9.5 million in Indonesia, 7.5 million in Viet Nam, 4.9 million in Thailand and 4.5 million in the Philippines. The report estimated varying employment impacts across ASEAN-6 countries (Indonesia, Malaysia, the Philippines, Singapore, Thailand and Viet Nam) would be partly driven by differences in respective economic structures. It predicted that technological displacement would occur most intensively in the agriculture sector, affecting 13 percent of the workforce, which equates to around 10 million full time equivalent (FTE) jobs and would affect agriculture-dependent Indonesia and Viet Nam more heavily (sector accounts for 13 and 17 percent of their respective GDPs). In the manufacturing sector, also a major employer across ASEAN-6, technology was projected to displace up to 10 percent of the workforce over the next decade, equating to more than four million FTE jobs.

At the same time, the Oxford Economics and Cisco report found that as technological innovations would be more widely adopted in the coming years, labour productivity would be enhanced across all ASEAN-6 economies, as the technology that would displace some workers would also boost economic growth and create more jobs. New technologies cut production costs, which lowers the prices of goods, services and raises population spending power (known as the ‘income effect’). The report outlined a modelling scenario in which Viet Nam would benefit from: (i) a vibrant mobile economy, driven by a large, young, digital-savvy workforce, (ii) high levels of investment in advanced infrastructure meant 5G connectivity would be established in cities and most rural areas covered by internet services, (iii) state-of-the-art

2 “The future of jobs at risk of automation”, International Labour Organization, July 2016. Available at: http://www.ilo.org/actemp/ publications/WCMS_579554/lang--en/index.htm

22 Productivity and Competitiveness of Viet Nam’s Enterprises FDI in manufacturing accounts for almost 70 percent of total FDI into Viet Nam, the largest proportion IoT technologies supported export manufacturing and logistics, (iv) data localization policies proved of FDI in manufacturing in ASEAN followed by Indonesia (40 percent) and the Philippines (38 percent). an impediment to advanced use of the cloud, distributed computing and AI development and (v) the GSO data shows that 64.6 percent of newly-registered FDI capital was in manufacturing, 10.1 percent prevalence of cheap and abundant labour meant the domestic economy would be characterized by in real estate and 2.4 percent in wholesale, retail and car-motorbike repairs in 2016. In 2017, electricity outdated practices. Under this scenario, the report forecasted that by 2028: production and distribution, gas and air-conditioning attracted the most newly-registered FDI with USD8.4 billion (42.3 percent of total) and manufacturing only USD6.3 billion (31.7 percent). • Sectors with the most displaced jobs were agriculture: 3.4 million displaced (17.1 percent of workforce), manufacturing: 1.3 million (13.2 percent) and wholesale and retail: 840,000 (10.9 percent) The sources of FDI inflows into Viet Nam were diversified with Japan, RoK and Singapore the top investors out of 68 countries and territories. • Sectors with most jobs created (through the above-mentioned ‘income effect’) were manufacturing: 1.7 million jobs created (8.5 percent of workforce), wholesale and retail: 1.6 million (16.4 percent) and Job vulnerability to automation and AI hotels and restaurants: 1.3 million (16.8 percent)

Technological progress has accelerated in the recent decade, which is termed by numerous • 1.8 million existing roles would disappear from the labour market, with more than 90 percent of commentators as the Fourth Industrial Revolution (IR4.0). It is changing the global economic landscape, redundancies in agriculture, pushing workers into other industries and occupations. with far reaching implications for all actors – government, businesses and society at large. Among the key concerns of IR4.0 are (simple skilled/repetitive) job loses to automation and Artificial Intelligence 1.2. Labour Productivity Performance (AI). Data from the World Bank’s World Development Indicators shows labour productivity in Viet Nam grew by 4.7 percent per year on average during 1991-2016. The growth rate was the highest in ASEAN, greatly The report “The future of jobs at risks of automation”2 (ILO 2016) estimated that in the next few exceeding the -0.7 and 2.5 percent of low- and lower middle-income countries respectively, but much decades among ASEAN-5 (Cambodia, Indonesia, the Philippines, Thailand and Viet Nam), the share lower than China (9 percent) during this period (Figure 1.5). of jobs with a high probability of automation was lowest in Thailand (44 percent) and highest in Viet Nam (70 percent). In the Philippines, Indonesia and Cambodia, the shares were 49, 56 and 57 percent, This was largely attributed to Viet Nam’s impressive performance during the 1991-2006 sub-period, respectively (although some jobs, such as in agriculture, will be affected by structural challenges beyond when labour productivity grew by 5.1 percent per annum on average, the highest in ASEAN. It must mechanization). Within the average of 70.4 percent of all jobs in Viet Nam’s economy assessed at risk be noted, however, that Viet Nam’s labour productivity growth decreased since 2007, with 4.1 percent of automation, some sectors/sub-sectors with very high proportions of jobs with a high probability of average annual growth during the 2007-2016 sub-period3 and the rate was only the second highest automation risks include agriculture, forestry and fisheries (83.3 percent), manufacturing (74.4 percent), (behind Lao PDR) among ASEAN countries. Across the same period, average labour productivity in low- food and beverages (68 percent), garments (85 percent) and electronics (75 percent), wholesale, retail and lower middle-income groups grew at 2 and 3.7 percent, respectively. As such, labour productivity and repair of motor vehicles (84.1 percent), service sector (32 percent), retail (70 percent), hotel and growth in Viet Nam accelerated during 1991-2006 (when Viet Nam was in the low-income group) and banking (around 40 percent). Occupations with the highest risk of disappearing in the future include slowed during 2007-2016 (Viet Nam joined the lower middle-income group) in contrast to the low- and shop sales assistants (2.1 million), garden labourers (1 million) and sewing-machine operators in garment lower middle-income groups’ trend to slow in the first sub-period and accelerate in the second. manufacturing (770,000). Oxford Economics and Cisco’s September 2018 report “Technology and the future of ASEAN jobs - The impact of AI on workers in ASEAN’s six largest economies” (Oxford Economics and Cisco, 2018) forecasted that by 2028, “displaced workers” would number 9.5 million in Indonesia, 7.5 million in Viet Nam, 4.9 million in Thailand and 4.5 million in the Philippines. The report estimated varying employment impacts across ASEAN-6 countries (Indonesia, Malaysia, the Philippines, Singapore, Thailand and Viet Nam) would be partly driven by differences in respective economic structures. It predicted that technological displacement would occur most intensively in the agriculture sector, affecting 13 percent of the workforce, which equates to around 10 million full time equivalent (FTE) jobs and would affect agriculture-dependent Indonesia and Viet Nam more heavily (sector accounts for 13 and 17 percent of their respective GDPs). In the manufacturing sector, also a major employer across ASEAN-6, technology was projected to displace up to 10 percent of the workforce over the next decade, equating to more than four million FTE jobs.

2 “The future of jobs at risk of automation”, International Labour Organization, July 2016. Available at: http://www.ilo.org/actemp/ 3 The year 2007 is an important milestone in Viet Nam’s modern economic history, as the country graduated from one of the least publications/WCMS_579554/lang--en/index.htm developed nations to a lower middle-income one. On the integration front, Viet Nam also officially became a member of the WTO.

22 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 23 Figure 1.5: Labour productivity in selected countries, 1991-2016: Average annual growth rate (%)

10.00%

8.00%

6.00% 5.06% 4.73% 4.22% 4.00%

2.00%

0.00% AverageTrung bình 1991-2016 1991-2016 AverageTrung bình 1991-2006 1991-2006 TrungAverage bình 2007-2016 2007-2016

-2.00%

Vietnam Laos Laos CambodiaCambodia Thailand Singapore Indonesia MalaysiaMalaysia PhilippinesPhilippines Brunei China India Lower middleTrung bìnhincome thấp LowTrung & middle bình và income thấp LowThu nhậpincome thấp TrungUpper bìnhmiddle cao income

Source: Authors’ calculations from data of World Bank’s World Development Indicators

Outside ASEAN, Viet Nam’s labour productivity growth rate was consistently below those of China in both sub-periods and India’s 6 percent during 2007-2016. Viet Nam’s average annual labour productivity growth during 1991-2016 was south of these two countries by large margins. This demonstrates that labour productivity growth can keep pace with rapid development in giant economies.

Box 1. 1: How long does Viet Nam need to fully close absolute labour productivity gaps with ASEAN countries?

The issue of “catching-up” in labour productivity has recently featured prominently in policy debates in Viet Nam, as the country has set an objective of rapid and sustainable growth for the five-year 2016-2020 period and beyond. Thanks to its highest average labour productivity growth rate during 1991-2016, Viet Nam somewhat closed labour productivity gaps with other ASEAN countries relatively fast - especially compared to Brunei and Cambodia. However, gaps between Viet Nam and other ASEAN countries remain significant. If and how many years it would take Viet Nam to close these gaps can be assessed through the following “simple algebra”.

Assessment model: Suppose that in year 0 (the initial year), relative labour productivity of country i over Viet Nam is , where Z denotes labour productivity, i – country i, v – Viet Nam

Suppose that in year t, labour productivity of country i and Viet Nam will be , and respectively; where are respectively the average rates of labour productivity of country I and Viet Nam in the period from 0 to t

Then relative labour productivity of country i against Viet Nam in year t will be

(1.1)

The catch up, implies that ⇔

⇔ (1.2), where t is the number of years that are needed for Viet Nam to fully close the labour productivity gaps with comparator countries.

24 Productivity and Competitiveness of Viet Nam’s Enterprises Figure 1.5: Labour productivity in selected countries, 1991-2016: Average annual growth rate (%) Equation (1.2) says that if the initial relative labour productivity is large (then the numerator is 10.00% large) and/or the difference in growth rates of labour productivity is small (then the denominator

8.00% is small), the number of years required for the catch up would be bigger. Assessment results: If the year 2016 is used as the initial one, under different scenarios of growth 6.00% 5.06% rates comparable to those achieved in 1991-2016 and 2007-2016 periods as presented in Figure 4.73% 4.22% 1.5, the number of years required for Viet Nam to fully close the absolute gaps with the comparator 4.00% countries is presented in Table 1.1.

2.00% Table 1.1: Number of years required for Viet Nam to catch regional countries in labour productivity 0.00% Trung bình 1991-2016 Trung bình 1991-2006 Trung bình 2007-2016 Countries Absolute LP in Relative LP in 2016 The number of years required, if countries keep -2.00% 2016 (in USD PPP (Việt Nam = 1) growing as in the period of Vietnam Laos Cambodia Thailand Singapore 2011) 1991 - 2016 2007 - 2016 Indonesia Malaysia Philippines Brunei China India Trung bình thấp Trung bình và thấp Thu nhập thấp Trung bình cao Brunei 156,100 15.8 48 46 Singapore 144,424 14.6 149 101 Source: Authors’ calculations from data of World Bank’s World Development Indicators malaysia 55,350 5.6 81 59 Thái Lan 27,165 2.7 60 66 Outside ASEAN, Viet Nam’s labour productivity growth rate was consistently below those of China in Indonesia 23,352 2.4 45 135 both sub-periods and India’s 6 percent during 2007-2016. Viet Nam’s average annual labour productivity Philippines 17,373 1.8 20 39 growth during 1991-2016 was south of these two countries by large margins. This demonstrates that labour productivity growth can keep pace with rapid development in giant economies. Laos 11,192 1.1 72 Never Vietnam 9,891 1 Cambodia (*) 6,254 0.6 Never Never Box 1. 1: How long does Viet Nam need to fully close absolute labour productivity gaps China 25,530 2.6 Never Never with ASEAN countries? India 16,282 1.6 Never Never The issue of “catching-up” in labour productivity has recently featured prominently in policy Source: Authors’ calculations. Note: If past performance is a good predictor of the future, Viet Nam would never debates in Viet Nam, as the country has set an objective of rapid and sustainable growth for the catch China and India as the current gaps are large, but more importantly labour productivity in Viet Nam has five-year 2016-2020 period and beyond. Thanks to its highest average labour productivity growth grown consistently and considerably slower than in these two countries. Viet Nam would never catch Lao PDR rate during 1991-2016, Viet Nam somewhat closed labour productivity gaps with other ASEAN if future labour productivity growth rates between the pair remained the same as the 2007-2016 sub-period. countries relatively fast - especially compared to Brunei and Cambodia. However, gaps between Cambodia would not catch Viet Nam if its labour productivity growth remained the same. Viet Nam and other ASEAN countries remain significant. If and how many years it would take Viet Nam to close these gaps can be assessed through the following “simple algebra”. This implies Viet Nam must do considerably more to achieve better labour productivity growth than in the recent past if it wants to catch up. Assessment model: Suppose that in year 0 (the initial year), relative labour productivity of country i over Viet Nam is , where Z denotes labour productivity, i – country i, v – Viet Nam 1.3. Sources of Labour Productivity Growth: a Shift Share Analysis

Suppose that in year t, labour productivity of country i and Viet Nam will be To understand sources of labour productivity growth, a shift share analysis is a useful tool. It allows for , and respectively; where are respectively the average rates of labour the decomposition of labour productivity growth into three components: (i) the ‘within sector effect’ that productivity of country I and Viet Nam in the period from 0 to t measures the contribution of each sector’s productivity growth to overall productivity growth, (ii) the ‘structural change effect’ that measures the contribution of structural change as measured by changing Then relative labour productivity of country i against Viet Nam in year t will be sectoral employment shares to overall productivity growth and (iii) the ‘interactive effect’ that measures (1.1) the contribution of interactions within a sector and shift effects to overall productivity growth. Details on this methodology are given in Appendix A.1.2. The catch up, implies that ⇔ Viet Nam’s economy has been structurally transformed over the last few decades, as evidenced by a ⇔ (1.2), where t is the number of years that are needed for Viet Nam to fully substantial reduction in agriculture’s employment share from 73 percent in 1991 to 43 percent in 2016. close the labour productivity gaps with comparator countries. During the same period, employment shares of industry and services rose from 9 and 18 percent in 1991 to 23 and 35 percent in 2016, respectively (Figure 1.6). Such structural change is a key driver of labour productivity growth, as productivity gaps between agriculture and industry and services sectors have been large, yet narrowed over time. Specifically, labour productivity of industry and services were 4.6

24 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 25 and 3.6 times that of agriculture in 1991, and 3.8 and 3.1-fold in 2016 (Figure 1.7). Therefore, such a shift in workers out of agriculture towards industry and services resulted in productivity gains through the ‘reallocation effect’.

Figure 1.6: Structural transformation in Viet Nam during 1991-2016

TỷEmployment phần lao động share của of cácsectors ngành 100% 100% 27% 18% 18% 26% 35% 80% 9% 21% 60% 19% 23%

40% 73% 53% 56% 20% 43%

0% 2011 1991 2011 1991 1997 2012 2013 2015 1997 2014 1992 2012 1993 2013 2016 1995 2010 2001 2015 1994 2014 1992 1998 1993 1999 1996 2016 1995 2001 2010 1994 1998 1999 1996 2007 2007 2002 2003 2005 2004 2002 2008 2003 2009 2006 2005 2000 2004 2008 2009 2006 2000

NôngShare nghiệp ofAgr. CôngShare nghiệp ofInd. DịchShare vụ ofServ.

Source: Authors’ calculations based on data of the World Bank’s World Development Indicators

Figure 1.7: Sectoral labour productivity4 in Viet Nam during 1991-2016

NăngLabour suất lao productivity động ngành of công industry nghiệp and và services dịch vụ relativeso với năng to agriculture suất lao động ngành nông nghiệp 1000%1000% 461%461% 364%364% 800%800% 455% 600%600% 527% 311%311% 378%378% 400%400% 200%200% 0%0%

Relative_Ind_PCông nghiệp DịchRelative vụ Serv_P Source: Authors’ calculations based on data of the World Bank’s World Development Indicators

Shift share analysis results are presented in Figure 1.8. The upper panel captures 1991-2006 and the lower panel 2007-2016. A comparison of the two periods reveals the contribution of the ‘structural change effect’ on productivity growth in Viet Nam declined over time, from 33 percent during 1991-2006 to 28 percent during 2007-2016, while the influence of the ‘within sector effect’ rose significantly (46 to 71 percent). The ‘interaction effect’ was positive across both periods, as the employment shares of sectors with lower-than-average labour productivity shrunk, while those of sectors with higher-than- average labour productivity rose. As such, the ‘within sector’ and ‘structural change’ effects reinforced one another to raise labour productivity over time in Viet Nam. However, the ‘interaction effect’ declined significantly from 21 percent in the first period to only 2 percent in the second.

4 Labour productivity is calculated in USD, 2011 PPP.

26 Productivity and Competitiveness of Viet Nam’s Enterprises and 3.6 times that of agriculture in 1991, and 3.8 and 3.1-fold in 2016 (Figure 1.7). Therefore, such a shift Figure 1.8: Sources of labour productivity growth (%) in workers out of agriculture towards industry and services resulted in productivity gains through the GiaiPeriod đoạn 1991-2006 1991-2006 ‘reallocation effect’. 120% 2% 100% 7% 2% 6% 10% 19% 11% 12% 10% 12%12% 21%21% 12%12% Figure 1.6: Structural transformation in Viet Nam during 1991-2016 80% 20% 6% 40% 22% 22%22% 22%22% 55% 60% 113%113% 33%33% Tỷ phần lao động của các ngành 40% 87% 76% 66% 71% 66% 65%65% 100% 53% 27% 20% 39% 46%46% 18% 26% 35% 80% 9% 0% --11%11% 21% --3%3% 19% -20% 60% China Indonesia Malaysia Philippines Thailand TBLower thấp ThấpLow &và middle TB ThấpLow ViệtVietnam Nam Upper TB cao middle 23% middleincome income income income 40% 53% 73% Hiệu ứng nội ngành Hiệu ứng chuyển dịch cơ cấu Hiệu ứng tương tác 56% Within Effect Structural change effect Interaction effect 20% 43% 150%150% GiaiPeriod đoạn 2007-2016 2007-2016 0% 100%100% 12% 1%1% 5%5% 4%4% 4%4% 5%5% 2%2% 4%4% 17%17% 10% 37% 24%24% 36%36% 24%24% 20%20% 28%28% 2011 1991 1997 2012 2013 2015 2014 1992 1993 2016 1995 2010 2001 1994 1998 1999 1996 2007 2002 2003 2005 2004 2008 2009 2006 2000 75%75% 50%50% 102%102% 107%107% 78% 79%79% 78% 67% 72%72% 60%60% 72%72% 71%71% Nông nghiệp Công nghiệp Dịch vụ 0%0% -6%-6% -4% -3% Source: Authors’ calculations based on data of the World Bank’s World Development Indicators -4% -3% -2%-2% -50%-50% China IndonesiaIndonesia MalaysiaMalaysia PhilippinesPhilippines ThailandThailand TBLower th ấp TLowhấp &v àmiddle TB ThấLowp ViVietnamệt Nam UpperTB cao middle middleincome income income income Figure 1.7: Sectoral labour productivity4 in Viet Nam during 1991-2016 Hiệu ứng nội ngành Hiệu ứng chuyển dịch cơ cấu Hiệu ứng tương tác Within Effect Structural change effect Interaction effect Năng suất lao động ngành công nghiệp và dịch vụ so với năng suất lao động ngành nông nghiệp Source: Authors’ calculation based on data of the World Bank’s World Development Indicators 1000% 461% 364% 800% Figure 1.8 shows that sources of labour productivity growth in Viet Nam appeared to follow low middle- 455% 600% 527% 311% income group patterns in both sub-periods. During 1991-2006 (when Viet Nam a a low middle-income 378% 400% nation, Viet Nam’s shares of ‘structural change’ and ‘within sector’ effects were 33 and 46 percent respectively, resembling average shares in lower middle-income countries (20 and 71 percent) and in 200% 2007-2016 when Viet Nam was a lower middle-income nation the shares were 28 and 71 percent versus 0% the average of 20 and 75 percent of lower middle-income countries.

Công nghiệp Dịch vụ It is interesting to note the case of Thailand, which has a much higher share ‘reallocation effect’ contribution to productivity growth than Viet Nam and other countries in comparison. Thailand clearly Source: Authors’ calculations based on data of the World Bank’s World Development Indicators is at a more advanced level of development than Viet Nam and some other comparators, and common Shift share analysis results are presented in Figure 1.8. The upper panel captures 1991-2006 and the wisdom suggests the former’s economy would have a more stable structure and thus less scope lower panel 2007-2016. A comparison of the two periods reveals the contribution of the ‘structural for structural changes to contribute to productivity gains. But, Thailand’s data shows otherwise and change effect’ on productivity growth in Viet Nam declined over time, from 33 percent during 1991-2006 suggests room may exist for Viet Nam within its next higher development stage. to 28 percent during 2007-2016, while the influence of the ‘within sector effect’ rose significantly (46 1.4. Determinants of Labour Productivity at Firm Level to 71 percent). The ‘interaction effect’ was positive across both periods, as the employment shares of sectors with lower-than-average labour productivity shrunk, while those of sectors with higher-than- The previous section showed how the ‘within sector effect’ became a dominant source of labour average labour productivity rose. As such, the ‘within sector’ and ‘structural change’ effects reinforced productivity growth in Viet Nam in recent periods. Therefore, it is important to analyze the determinants one another to raise labour productivity over time in Viet Nam. However, the ‘interaction effect’ declined of labour productivity at firm level, presumably a major component of the ‘within sector effect’ in labour significantly from 21 percent in the first period to only 2 percent in the second. productivity increases.

As labour productivity in firms is measured by the ratio of value added over the number of workers, this section examines three categories of “standard” factors - workers, firms and their operational environments - that affect value added and thus influence labour productivity of firms. Econometric analysis of the 2017 Enterprise Census (which covered more than 330,000 firms, including 265,000 firms in the 2012-2017 two-wave panel dataset)5 suggests the following determinants of (or more rigorously, factors associated with) firms’ labour productivity:

5 Brief description of Enterprise Census is given in Section A.3.1. of Appendix 3. Details on regression results are provided in Section 4 Labour productivity is calculated in USD, 2011 PPP. A.3.2. of Appendix 3.

26 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 27 Worker-related factors

The presence of foreign workers in a firm is beneficial to productivity. Raising their share by 1 percent boosts productivity by 119 percent (e779 - 1), indicating the presence of strong spill-over effects from foreign to Vietnamese workers. With regard to human capital of workers as proxied by the level of educational attainment, most indicators (e.g. ratios of workers with primary vocational certificates, secondary (vocational) or (vocation) college degrees, ratio of workers with degrees or higher) had unexpectedly negative characteristics, except the ratio of workers with short-term training certificates. This may indicate production is predominantly based on simple skilled labour and short-term training yields positive returns to firms’ labour productivity, while the yield in other indicators is more ambiguous. Age of workers also matters. Firms with a bigger share of workers under 30 years of age were the most productive and this may reinforce the above explanation. Firm-related factors Capital deepening is important:

Raising the ratio of capital over labour, which measures how well workers are equipped, by 1 percent will increase labour productivity by 0.37 percent (0.38 percent, if firms not in the 2012-2017 panel are excluded).

Firms’ size matters:

The size premium, as measured by the productivity gap between firms with different sizes and micro- firms with less than five workers as a reference group, becomes larger at a diminishing speed as a firm’s size increases (Figure 1.9). For example, firms with 50-99 workers have higher productivity, by 119 percent (i.e. e0.785 - 1) more than firms with less than five workers, with other things being equal. If firms in the 2012-2017 panel were taken into account, the size premium exhibits an inverted U-shape pattern with firms’ sizes: (i) more than five workers are more productive than the latter and (ii) between 50-99 workers are the most productive (with labour productivity higher by 106.3 percent (i.e. e0.724 - 1) than firms with less than five workers), with other things being equal (Figure 1.9). This finding indicates that a larger firm size depicts higher labour productivity, as it presumably facilitates the learning process among peers as well as other advantages arising from economies of scale. The inverted U-shape indicates a turning point in the size premium where the additional costs, associated with limits on managerial capabilities for example, outweigh the additional benefits of bigger size.

Figure 1.9: Size Premium

140%140% 128% 125% 128% 119% 121% 120%120% 108% 106% 106% 104% 103% 103% 96% 100%100% 91% 82% 80%80% 59% 60%60% 53%

40%40%

20%20%

0% 11 2 3 4 5 6 7 8

AllTất firms, cả DN, 2016 2016 CảcFirms DN in trong 2015-2016 điều tra panel 2015-2016

Source: Authors’ calculation using 2017 Enterprises Census data

28 Productivity and Competitiveness of Viet Nam’s Enterprises Worker-related factors Firms’ managerial capabilities matter

The presence of foreign workers in a firm is beneficial to productivity. Raising their share by 1 percent If a firm’s top manager holds a masters degree or higher, labour productivity increases by 2.1 percent boosts productivity by 119 percent (e779 - 1), indicating the presence of strong spill-over effects from from the base case, where the manager only has a junior college degree or lower. A manager’s experience foreign to Vietnamese workers. With regard to human capital of workers as proxied by the level of as proxied by its age also matters, with experience premium following an inverted U-shaped pattern. educational attainment, most indicators (e.g. ratios of workers with primary vocational certificates, secondary (vocational) or (vocation) college degrees, ratio of workers with degrees or higher) had Digitization makes a difference unexpectedly negative characteristics, except the ratio of workers with short-term training certificates. Firms that use computers more intensively, have a website and use the internet in various activities This may indicate production is predominantly based on simple skilled labour and short-term training are more productive. Specifically, productivity of firms with computers is 9.1 percent higher than those yields positive returns to firms’ labour productivity, while the yield in other indicators is more ambiguous. without. Firms with a website, an important sign of being online, are 5.7 percent more productive than Age of workers also matters. Firms with a bigger share of workers under 30 years of age were the most off-line firms. Companies that use the internet for operational management are 2.8 percent more productive and this may reinforce the above explanation. productive than others.

Firm-related factors Participating in the global market helps

Capital deepening is important: Firms engaged in export and/or import activities are 29.6 percent more productive than those not.

Raising the ratio of capital over labour, which measures how well workers are equipped, by 1 percent Level of technological sophistication in manufacturing and knowledge intensity in services will increase labour productivity by 0.37 percent (0.38 percent, if firms not in the 2012-2017 panel are matters excluded). With all other factors being equal, firms in medium and high-tech manufacturing as well as services and Firms’ size matters: construction are more productive than ones in low-tech manufacturing, while companies in agriculture, forestry and fisheries, electricity and mining are less productive (Figure 1.10). The size premium, as measured by the productivity gap between firms with different sizes and micro- firms with less than five workers as a reference group, becomes larger at a diminishing speed as a Figure 1.10: Labour productivity of firms in sectors relative to ones in low-tech manufacturing firm’s size increases (Figure 1.9). For example, firms with 50-99 workers have higher productivity, by 119 20%20% percent (i.e. e0.785 - 1) more than firms with less than five workers, with other things being equal. If firms 14% 14%16%16% 14% 13% 12% 14% 15%15% 11% 12%12% 12% in the 2012-2017 panel were taken into account, the size premium exhibits an inverted U-shape pattern 7% 11% 7% 10%10% 5% with firms’ sizes: (i) more than five workers are more productive than the latter and (ii) between 50-99 5% 7% 7% 5%5% 0.724 workers are the most productive (with labour productivity higher by 106.3 percent (i.e. e - 1) than firms 0%0% Agriculture,nông-lâm Forestry,nghiệp Mining,khai khoáng-điện Electricity ConstructionXây dựng High-techchế tạo Manuf. Medium-techchế tạo Manuf Knowledge-intensiveserv.dịch vụ Lessdịch knowledge- vụ ít with less than five workers), with other things being equal (Figure 1.9). This finding indicates that a larger -5% Fishery 2% intensive serv. -5% --2% công nghệ cao công nghệ trung bình thâm dụng tri thức thâm dụng tri thức firm size depicts higher labour productivity, as it presumably facilitates the learning process among -10%-10% -15% peers as well as other advantages arising from economies of scale. The inverted U-shape indicates -15% --4% -20% a turning point in the size premium where the additional costs, associated with limits on managerial -20% -25%-25% --22%22% capabilities for example, outweigh the additional benefits of bigger size. --24%24% -30%-30% Figure 1.9: Size Premium AllTất firms, cả DN, 2016 2016 AllCảc matched DN trong firms điều 2015-2016 tra 2015-2016 140% 125% 128% Source: Authors’ calculation using Enterprises Census data 119% 121% 120% 108% 106% 104% 103% 103% 96% 100% 91% 82% 80% 59% 60% 53%

40%

20%

0% 1 2 3 4 5 6 7 8

Tất cả DN, 2016 Cảc DN trong điều tra 2015-2016

Source: Authors’ calculation using 2017 Enterprises Census data

28 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 29 Foreign firms are the most productive Figure 1.11; Ownership and productivity (reference: private firms)

60.0% 56.0%

50.0% 42.2% 40.0% 32.3% 31.4% 30.0% 18.5% 20.0% 16.1%

10.0% CollectiveHợp tác xã 0.0% StateDNNN Comp. Hình thứcMixed sở Comp. hữu hỗn hợp MixedFDI Comp. -10.0%

-20.0% -18.0%-17.1%

AllTất firms, cả DN, 2016 2016 AllCảc matched DN trong firms điều 2015-2016 tra 2015-2016

Source: Authors’ calculation using Enterprises Census data

With other factors being equal, foreign firms are the most productive with labour productivity exceeding private firms’ by 56 percent (if all firms in Enterprise Census are taken into account) or 42 percent (if only firms in the 2012-2017 panel are taken into account), outperforming SOEs by 32 and 31 percent, respectively and mixed-ownership companies by 19 and 16 percent, respectively. Cooperatives are the least productive, with labour productivity below private firms’ by 18 and 17 percent (Figure 1.11).

Location matters Figure 1.12: Labour productivity of firms in different regions (reference: northern central and central coast) 45.0% 42.0% 40.0%

35.0% 31.6% 31.4% 28.9% 30.0%

25.0% 19.6% 20.0% 17.6%

15.0%

10.0% 5.3% 5.5% 5.4% 5.7% 5.0% 3.4% 0.9% 0.0% TâyCentral Nguyên ĐồngMekong Bằng MêRiver Công MiềnNorthern núi phía bắc HaHà NoiNội ĐồngRed bằng River sông Delta HồngHo ChiTP HCM Minh City Highlands Delta Mountain (except(trừ Hà Ha Nội) Noi)

AllTất firms, cả DN, 2016 2016 AllCảc matched DN trong firms điều 2015-2016 tra 2015-2016

Source: Authors’ calculation using Enterprises Census data

Firms in Ho Chi Minh City are the most productive with other factors being equal, followed by companies in the South East, Mekong River Delta and Ha Noi. Firms in other regions are considerably less productive than ones in Ho Chi Minh City (Figure 1.12).

Agglomeration effect is evident A 1 percent increase in the spatial concentration of firms from the same sector raises their labour productivity by 0.54 percent.

30 Productivity and Competitiveness of Viet Nam’s Enterprises Foreign firms are the most productive 2. Productivity and Competitiveness at Sector and Sub-sector Levels Figure 1.11; Ownership and productivity (reference: private firms) 2.1. Measurements to assess productivity and competitiveness of manufacturing 60.0% 56.0% sector and its sub-sectors

50.0% 42.2% The measurements used in this report to assess the manufacturing sector’s productivity and 40.0% competitiveness include: (i) Manufacturing Value Added (MVA)-to-GDP ratio, in which MVA is the total 32.3% 31.4% value added of manufacturing sector and the increase (decrease) in MVA-to-GDP ratio is commonly 30.0% seen as a sign of industrialization (de-industrialization), (ii) Viet Nam’s (and comparator countries’) MVA 18.5% 20.0% 16.1% share of the global MVA, (iii) per capita MVA, (iv) Competitive Industrial Performance (CIP) index ranking6,

10.0% (v) labour productivity (LP), (vi) value added-to-output ratio and (vii) Revealed Comparative Advantage 7 Hợp tác xã (RCA) . 0.0% DNNN Hình thức sở hữu hỗn hợp FDI Measurements used to (i) describe key characteristics of the manufacturing VSIC 2-digit sub-sectors -10.0% include revenue, employment, value added (VA), net exports, FDI backward and forward linkages and (ii) -20.0% -18.0%-17.1% to assess productivity and competitiveness of sub-sectors LP, RCA, domestic contents of exports, VA- to-output ratios and wage growth are applied. Tất cả DN, 2016 Cảc DN trong điều tra 2015-2016

Source: Authors’ calculation using Enterprises Census data Data for international comparisons of MVA-to-GDP ratios, MVA share of global MVA, per capita MVA and LP are drawn from WDI and UNIDO database (2017) and for RCA and net exports data from UN Comtrade With other factors being equal, foreign firms are the most productive with labour productivity exceeding database is utilized. Viet Nam’s data from the Enterprise Census for 2012 and 2017 are used to analyze private firms’ by 56 percent (if all firms in Enterprise Census are taken into account) or 42 percent (if productivity and competitiveness performance at sub-sector level, especially for revenue, employment, only firms in the 2012-2017 panel are taken into account), outperforming SOEs by 32 and 31 percent, VA, FDI backward and forward linkages, VA-to-output ratios, LP and wage growth. respectively and mixed-ownership companies by 19 and 16 percent, respectively. Cooperatives are the least productive, with labour productivity below private firms’ by 18 and 17 percent (Figure 1.11). Important points on Enterprise Census data and methods used to estimate value added are:

Location matters • The Enterprise Census only covers formal enterprises in Viet Nam. The household business sector, comprising unregistered household enterprises, is large in terms of number and estimated Figure 1.12: Labour productivity of firms in different regions (reference: northern central and at nine million. It contributes an estimated 23 percent of GDP (Doumer et al. 2017). However, the central coast) collection of data on household businesses was not conducted by GSO systematically, but instead 45.0% 42.0% in an ad hoc manner by research institutions (Cling et. al. 2009, Doumer et. al 2017). Therefore, the 40.0% household business sector was not included in this study and analyses results in this report are only 35.0% representative for the formal enterprise sector. 31.6% 31.4% 28.9% 30.0% • VA can be calculated from either production method or income method. This report used VA data 25.0% derived from both methods, namely: UNIDO’s VA data estimated by using the production method for 19.6% 20.0% 17.6% international comparisons and VA estimated by using Enterprise Census data and income method for comparison between subsectors within Viet Nam. 15.0%

10.0% • UNIDO’s recommended VA calculation method and data: the VA of the manufacturing industry refers 5.3% 5.5% 5.4% 5.7% 5.0% 3.4% 0.9% 6 As a performance indicator, CIP reflects a country’s productivity, structural change and competitiveness. These concepts are 0.0% taken as a departure point for the selection of indicators under the three major dimensions of the CIP (Dimension 1 “Capacity to Tây Nguyên Đồng Bằng Mê Công Miền núi phía bắc Hà Nội Đồng bằng sông Hồng TP HCM produce and export” measured by indicators: 1. manufacturing value added per capita, 2. manufacturing export per capita; Dimension (trừ Hà Nội) 2 “Technological upgrading and deepening” measured indicators: 3. share of manufacturing high-technology (MHT) activities in total Tất cả DN, 2016 Cảc DN trong điều tra 2015-2016 MVA, 4. share of MVA in GDP, 5. share of MHT manufactures exports, and 6. share of manufactures export in total exports; Dimension 3 “Impact on world production and trade” measured indicators, 7.sShare of the country in world MVA and 8. share of the country in Source: Authors’ calculation using Enterprises Census data world manufactures exports). The first dimension includes MVA per capita, which is the ratio of output to the country’s population. This indicator represents the level of overall productivity and quantifies the country’s capacity to produce. Another indicator of the Firms in Ho Chi Minh City are the most productive with other factors being equal, followed by companies same dimension shows the extent of the realization of domestic manufacturing products in external markets. The second dimension in the South East, Mekong River Delta and Ha Noi. Firms in other regions are considerably less productive of the CIP consists of indicators relating to the intensity of industrialization and quality of manufacturers’ exports. As industrialization progresses, two forms of major structural change may occur. First, the manufacturing sector’s position in the overall economy may than ones in Ho Chi Minh City (Figure 1.12). strengthen (increased share of MVA in GDP) and second, a gradual shift from low-technology and resource-based to high-technology products may occur. Increasing levels of industrialization trigger the export of high-technology and high quality products. The third Agglomeration effect is evident dimension comprises indicators on the country’s share in the world market and thus introduces exogenous factors into the analytical framework of the CIP. Source: https://unstats.un.org/unsd/ccsa/isi/2013/Paper-UNIDO.pdf A 1 percent increase in the spatial concentration of firms from the same sector raises their labour 7 RCA is calculated as the proportion of a country’s share of export of a certain class of goods or services (from, for example, productivity by 0.54 percent. a certain sub-sector or sector) in the country’s export divided by the proportion of world exports that are of that class. When a country’s RCA value higher than unity on a certain class of goods or services, the comparative advantage of the country in this class is “revealed” and when the RCA is less than unity, comparative disadvantage is “revealed”.

30 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 31 to the given industry’s net output derived from the difference in gross output and intermediate consumption calculated without deducing consumption of fixed assets. Viet Nam’s VA data in UNIDO’s database is at producer prices, meaning VA at producers’ prices = Gross Output at producers’ price minus Intermediate Consumption at purchasers’ price, where: the producer price is the amount receivable by the producer, inclusive of taxes on products except deductible value added tax and exclusive of subsidies on products and the purchaser price is the amount payable by the purchaser (purchaser price = producer price + trade and transport margins + non-deductible Value Added Tax). As survey data on industry VA may disregard the contribution of small and household- based manufacturing units, and weak quality of the primary data for estimating gross output and intermediate consumption, producer’s and purchaser’s prices, industry VA is best used to measure the growth and structure, but not the level8.

• VA estimation using the Enterprise Census data: data for calculating intermediate consumption of enterprises is only collected in the Enterprise Census (EC) every five years and 2012 was the latest year when EC collected such data. For years when data is not collected, the intermediate consumption of enterprises is estimated based on data collected in the latest year. Given the fast changes in enterprises’ production and prices of intermediate goods, such estimates may contain a degree of inaccuracy. As such, this study employs the income method to estimate VA of manufacturing sub- sectors based on the Enterprise Census data. Namely, also based on the same principle of VA is what producers get from output less the cost of intermediate goods used for producing the output, the income method estimates VA as a sum of: (i) after-tax profit + taxes + payments of wages and employer’s compulsory social and health insurance contributions + depreciations (of fixed capital). As is the case when VA is calculated by the standard method recommended by UNIDO, the quality of primary data relating to taxes, after tax profit, wages and compulsory insurance contributions paid by employers, and especially depreciation of fixed assets also resulted in inaccuracy in VA estimation, and for these reasons, VA estimated by this method is also used to measure the growth and structure or in ratios rather than absolute level. In this study, when observed any contradiction or missing information, the VA calculated by income method was replaced by the estimation by production method, in which the intermediate expenditure was estimated from the latest (2012) I-O table provided by GSO.

• The report (in its analysis of growth and structure of VA and other indicators using VA) used VA calculated by both methods, while VA data from the UNIDO database (calculated by the standard method) are utilized more for international comparisons and VA estimates by income method using EC data – more for comparisons between subsectors in Viet Nam. Consistent results of VA calculated by both methods are presented in this report with differences highlighted. 2.2. Productivity and Competitiveness of Manufacturing 2.2.1. Performance at Sector Level Performance measured by manufacturing value added (MVA)

Viet Nam’s MVA-to-GDP ratio rose by 0.9 percent during 2005-2017. Compared to selected countries (Figure 2.1), such a rise in absolute terms was lower than those of Cambodia, China, Japan, India and RoK, while exceeding other comparator countries (Indonesia, Malaysia and Thailand) which saw these ratios shrink. The decline in manufacturing’s contribution to GDP (indicating de-industrialization) in these countries was mainly due to domestic production having shifted to the service sector with higher VA and as manufacturing costs in industrialized countries rise, companies move manufacturing plants to developing nations to sustain international competitiveness.

8 Further information on VA can be found in “What is manufacturing value added?” at http://stat.unido.org/content/focus/what-is- manufacturing-value-added%3F;jsessionid=82D5D3FAAC4ECE658DBFE31D5B1A4686

32 Productivity and Competitiveness of Viet Nam’s Enterprises to the given industry’s net output derived from the difference in gross output and intermediate Figure 2.1: MVA as a percentage of GDP consumption calculated without deducing consumption of fixed assets. Viet Nam’s VA data 35.0 31.6 in UNIDO’s database is at producer prices, meaning VA at producers’ prices = Gross Output at 29.9 29.0 28.8 producers’ price minus Intermediate Consumption at purchasers’ price, where: the producer price is 30.0 28.3 25.7 the amount receivable by the producer, inclusive of taxes on products except deductible value added 24.8 24.0 25.0 22.9 21.7 tax and exclusive of subsidies on products and the purchaser price is the amount payable by the 19.1 16.6 20.0 17.7 17.9 purchaser (purchaser price = producer price + trade and transport margins + non-deductible Value 16.4 15.7 15.2 14.1 Added Tax). As survey data on industry VA may disregard the contribution of small and household- 15.0 based manufacturing units, and weak quality of the primary data for estimating gross output and intermediate consumption, producer’s and purchaser’s prices, industry VA is best used to measure 10.0 8 the growth and structure, but not the level . 5.0

• VA estimation using the Enterprise Census data: data for calculating intermediate consumption of 0.0 enterprises is only collected in the Enterprise Census (EC) every five years and 2012 was the latest China Thailand R. Korea Malaysia Indonesia Viet Nam Japan Cambodia India year when EC collected such data. For years when data is not collected, the intermediate consumption 2005 2017 of enterprises is estimated based on data collected in the latest year. Given the fast changes in enterprises’ production and prices of intermediate goods, such estimates may contain a degree of Source: Calculated from UNIDO MVA data (2018) inaccuracy. As such, this study employs the income method to estimate VA of manufacturing sub- Despite climbing, Viet Nam’s share in global MVA is still small compared to others in the region. sectors based on the Enterprise Census data. Namely, also based on the same principle of VA is Specifically, in 2005, Viet Nam’s MVA accounted for only 0.15 percent of total global MVA and increased what producers get from output less the cost of intermediate goods used for producing the output, to 0.22 percent in 2017. During the same period, China’s more than doubled from 6.12 to 12.77 percent and the income method estimates VA as a sum of: (i) after-tax profit + taxes + payments of wages and as a result, it became a major global manufacturing powerhouse. Japan’s share of global MVA decreased employer’s compulsory social and health insurance contributions + depreciations (of fixed capital). from 9.78 to 7.77 percent, but the latter figure is still significant. Malaysia and Thailand’s shares only As is the case when VA is calculated by the standard method recommended by UNIDO, the quality slightly changed in this period (Figure 2.2a). of primary data relating to taxes, after tax profit, wages and compulsory insurance contributions paid by employers, and especially depreciation of fixed assets also resulted in inaccuracy in VA Figure 2.2a: Global MVA share of selected countries (%) estimation, and for these reasons, VA estimated by this method is also used to measure the growth 14.000 and structure or in ratios rather than absolute level. In this study, when observed any contradiction 12.772 or missing information, the VA calculated by income method was replaced by the estimation by 12.000 production method, in which the intermediate expenditure was estimated from the latest (2012) I-O 9.777 table provided by GSO. 10.000 7.767 • The report (in its analysis of growth and structure of VA and other indicators using VA) used VA 8.000 6.121 calculated by both methods, while VA data from the UNIDO database (calculated by the standard 6.000 method) are utilized more for international comparisons and VA estimates by income method using EC data – more for comparisons between subsectors in Viet Nam. Consistent results of VA calculated 4.000 3.307 1.697 by both methods are presented in this report with differences highlighted. 1.379 1.908 2.000 1.542 0.221 0.532 0.459 0.985 0.023 0.489 2.2. Productivity and Competitiveness of Manufacturing 0.014 0.147 0.353 0.000 2.2.1. Performance at Sector Level Cambodia Viet Nam Thailand Malaysia Indonesia India R. Korea Japan China 2005 2017 Performance measured by manufacturing value added (MVA) Source: Calculated from UNIDO data (2017) Viet Nam’s MVA-to-GDP ratio rose by 0.9 percent during 2005-2017. Compared to selected countries (Figure 2.1), such a rise in absolute terms was lower than those of Cambodia, China, Japan, India and RoK, Similarly, Figure 2.2b shows that Viet Nam’s VA per capita, despite increasing, is lower than all comparator while exceeding other comparator countries (Indonesia, Malaysia and Thailand) which saw these ratios countries except India. shrink. The decline in manufacturing’s contribution to GDP (indicating de-industrialization) in these countries was mainly due to domestic production having shifted to the service sector with higher VA and as manufacturing costs in industrialized countries rise, companies move manufacturing plants to developing nations to sustain international competitiveness.

32 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 33 Figure 2.2b: Manufacturing VA per capita (USD)

15000 10000 5000 0

China Indonesia India Japan Korea, Rep.

Malaysia Philippines Singapore Thailand Vietnam

Source:https://tcdata360.worldbank.org/indicators/mva.per.cap?country=BRA&indicator=3798&viz=line_ chart&years=1990,2014

Performance measured by Competitive Industrial Performance (CIP) index

In the early 1990s, Viet Nam had the worst performing in CIP ranking by some distance to comparator countries. Thanks to significant improvements in export value, MVA, VA of high-tech and medium- tech manufacturing over the last three decades, Viet Nam significantly improved its CIP ranking and catch the middle-income group (India, Indonesia) and gradually narrowed the gap with industrialized countries (Figure 2.3). However, this CIP improvement is mainly attributed to the FDI sector, with limited contributions from the domestic sector.

Figure 2.3: CIP rankings

100 Vietnam 90 Thailand 80 70 Malaysia 60 Indonesia 50 India 40 China 30 20 Korea

10 Japan 0 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 Source: UNIDO (2017)

Kết quả đo bằng Năng suất lao động (LP) Performance measured by Labour Productivity (LP)

Labour productivity of Viet Nam’s manufacturing industry is low compared to others in Asia, standing at 63.5 percent of India’s LP, 29.26 percent of Indonesia’s, 27.3 percent of Malaysia’s, 36.4 percent of the Philippines’, 7.2 percent of RoK’s and 7.8 percent of Japan’s in 2015. It is noted, however, that Viet Nam’s manufacturing industry closed such LP gaps with India (by 28.6 percentage point), Malaysia (10.1 percent point), the Philippines (9.8 percent point), RoK (4.3 percent point) and Japan (5.2. percent point) between 2005-2015 and with Thailand (21.1 percent point) between 2005-2011. Between 2005-2015, Viet Nam’s manufacturing LP gap between China and Indonesia widened by 5.4 and 3 percent, point respectively (Figure 2.4). Despite improvements, Viet Nam still faces an uphill task to close these wide LP gaps.

34 Productivity and Competitiveness of Viet Nam’s Enterprises Figure 2.2b: Manufacturing VA per capita (USD) Figure 2.4: Viet Nam’s manufacturing LP as a share of other countries’ manufacturing LP

15000 250.0 10000 202.6 5000 200.0 0 150.0

China Indonesia India Japan Korea, Rep. 103.6 Malaysia Philippines Singapore Thailand Vietnam 100.0

63.5 Source:https://tcdata360.worldbank.org/indicators/mva.per.cap?country=BRA&indicator=3798&viz=line_ 34.9 29.5 50.6 chart&years=1990,2014 50.0 33.5 36.4 27.5 30.5 27.3 26.6 22.1 17.2 Performance measured by Competitive Industrial Performance (CIP) index 2.9 7.2 2.6 7.8 0.0 In the early 1990s, Viet Nam had the worst performing in CIP ranking by some distance to comparator Bangladesh China India Indonesia Malaysia Philippines Thailand RoK Japan countries. Thanks to significant improvements in export value, MVA, VA of high-tech and medium- 2005 2011 2015 tech manufacturing over the last three decades, Viet Nam significantly improved its CIP ranking and Source: Authors’ calculation from UNIDO data catch the middle-income group (India, Indonesia) and gradually narrowed the gap with industrialized countries (Figure 2.3). However, this CIP improvement is mainly attributed to the FDI sector, with limited Performance measured by value added-to-output ratio contributions from the domestic sector. Another indicator to assess the competitiveness of firms (in sub-sectors/sector) is the VA-to-routput Figure 2.3: CIP rankings ratio, which measures what firms get out of their output after deducting the cost of intermediate consumption (costs of intermediate goods used for production). Viet Nam’s manufacturing sector 100 Vietnam performance measured by this indicator resembles its manufacturing performance measured by 90 Thailand other (presented above) indicators. Figure 2.5, using the UNIDO database, shows that Viet Nam’s 80 manufacturing VA-to-output ratio improved and is slightly higher than that of China, India and Malaysia, 70 Malaysia while lower than Japan, the Philippines and RoK. Enterprise Census data from 2017 reveals Viet Nam’s 60 Indonesia 9 50 average manufacturing VA-to-revenue increased from 20.53 percent in 2011 to 26.22 percent in 2016. India 40 China 30 Figure 2.5: Manufacturing VA-to-output ratio Korea 20 50.0 10 Japan 45.0 0 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 40.0 Source: UNIDO (2017) 35.0 30.0 26.1 Kết quả đo bằng Năng suất lao động (LP) 25.0 23.4 19.4 20.0 Performance measured by Labour Productivity (LP) 15.0 10.0 Labour productivity of Viet Nam’s manufacturing industry is low compared to others in Asia, standing at 5.0 63.5 percent of India’s LP, 29.26 percent of Indonesia’s, 27.3 percent of Malaysia’s, 36.4 percent of the 0.0 Philippines’, 7.2 percent of RoK’s and 7.8 percent of Japan’s in 2015. It is noted, however, that Viet Nam’s 2005 2011 2015 manufacturing industry closed such LP gaps with India (by 28.6 percentage point), Malaysia (10.1 percent Bangladesh China India Indonesia Malaysia Philippines Thailand ROK Japan Viet Nam point), the Philippines (9.8 percent point), RoK (4.3 percent point) and Japan (5.2. percent point) between Source: UNIDO database, authors’ calculation 2005-2015 and with Thailand (21.1 percent point) between 2005-2011. Between 2005-2015, Viet Nam’s manufacturing LP gap between China and Indonesia widened by 5.4 and 3 percent, point respectively (Figure 2.4). Despite improvements, Viet Nam still faces an uphill task to close these wide LP gaps.

9 As the UNIDO database provides data on output, not revenue, the “VA-to-output” ratios are calculated, not “VA-to-revenue” as analyzed using the Enterprise Census which provides data on firms’ revenue. Strictly speaking, output is different from revenue as the former only takes monetary values of outputs into account, while revenue also includes other incomes (such as royalties and donations) of firms.

34 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 35 2.1: Decomposing Total Factor Productivity: A stochastic frontier approach in measuring performance of Viet Nam’s manufacturing sector

Following recent developments in the measurement of productivity growth, a stochastic frontier production function is applied to decompose TFP growth in Vietnamese manufacturing into technical progress and changes in technical efficiency. Along with technical progress, changes in technical efficiency (gap between frontier technology and a firm’s actual production) can also contribute to productivity growth. Stochastic frontier models assume that firms do not fully utilize existing technology because of various non-price and organization factors that lead to inevitable technical inefficiencies in production. Under these circumstances, TFP growth may arise from improvements in technical efficiency (TE), without technical progress (TP).

From a policy perspective, researchers acknowledge that the decomposition of TFP into efficiency and technical changes provide useful information in productivity analysis. Policy recommendations that are based on the better understanding on the sources of variation in productivity growth can lead to more effective policies in enhancing the productivity of firms or sectors. For example, if low productivity results from slow TP, then a policy to induce technological innovation should be recommended to shift up the production frontier. If high rates of TP coexist with deteriorating TE, resulting in slow productivity growth, then a policy to increase the efficiency (with a known and applied technology) is required and might include aiming at improvements in learning-by-doing processes and managerial practices.

The growth rate of TFP can be decomposed into four components, i.e. the change of productivity due to TP, technical efficiency (TEC), scale effect (SEC) and factor reallocation effect (FAEC). That is,

TFP = TPit + TECit + SECit + FAECit SEC captures the productivity improvement resulting from the evolution of scale economy of industrial sectors. FAEC is referred to as factor allocative efficiency change. Under the two inputs of capital and labour, FAEC consists of two kinds of efficiency due to labour reallocation and capital reallocation. FAEC depend on the relative growth magnitude of two inputs. If the sum of labour allocation efficiency and capital reallocation efficiency is substantial, factor shifts have an impact on productivity. In turn, we can expect that the FAEC term reflects the industrial restricting efforts at the reallocation of factors in order to increase industrial productivity and growth.

Table 2.1: Decomposition of TFP

Growth in Growth in Growth in Growth in total Growth high-tech medium-tech low-tech Time period manufacturing components manufacturing manufacturing manufacturing sample (%) sectors (%) sectors (%) sectors (%) TP 2001-2015 6.82 6.86 6.39 6.05 TEC 2001-2015 -0.28 -0.35 -0.44 -1.15 SEC 2001-2015 5.17 6.78 3.92 7.38 FAEC 2001-2015 18.04 -27.10 8.02 53.77 TFP 2001-2015 29.74 -13.82 17.89 66.05

Table 2.1 presents the average of the rates of technical progress (TP, technical efficiency change (TEC), scale efficiency change (SEC), factor allocative efficiency change (FAEC) and the total factor productivity growth for the manufacturing industry, high-technology manufacturing, medium- technology manufacturing and low-technology manufacturing sectors samples during 2001-2015.

36 Productivity and Competitiveness of Viet Nam’s Enterprises 2.1: Decomposing Total Factor Productivity: A stochastic frontier approach in measuring At the total manufacturing industry sample level, the contribution of FAEC (allocative effect) to performance of Viet Nam’s manufacturing sector TFP growth is significant and dominates for the whole sample period. TC ranked second and was slightly higher than SEC at the same time. TE exerted a negative effect on TFP growth. Following recent developments in the measurement of productivity growth, a stochastic frontier production function is applied to decompose TFP growth in Vietnamese manufacturing into In terms of the high-tech manufacturing sector sample, estimated FAEC has a negative value technical progress and changes in technical efficiency. Along with technical progress, changes and given its magnitude negative FAEC resulted in reduced TFP. This implies the existence of in technical efficiency (gap between frontier technology and a firm’s actual production) can also an inefficient allocation of inputs (when factor prices are not equal to their marginal product) in contribute to productivity growth. Stochastic frontier models assume that firms do not fully utilize production of firms in high-tech manufacturing sector. In contrast, FAEC was estimated to be existing technology because of various non-price and organization factors that lead to inevitable positive and much larger for the medium- and low-tech manufacturing sectors. This discrepancy technical inefficiencies in production. Under these circumstances, TFP growth may arise from in FAEC amongst sub-sectors indicates that the degree of market distortion varied across these improvements in technical efficiency (TE), without technical progress (TP). sectors. The resulting inefficiency costs were generally greater in high-tech than low-tech manufacturing sectors. From a policy perspective, researchers acknowledge that the decomposition of TFP into efficiency and technical changes provide useful information in productivity analysis. Policy recommendations In all three groups, scale efficiency is a factor contributing slightly more to TFP growth than that are based on the better understanding on the sources of variation in productivity growth can technical progress. On the other hand, technical efficiency in all three groups deteriorates. lead to more effective policies in enhancing the productivity of firms or sectors. For example, if Obviously, TFP growth highly depends strongly on the reallocation in factors. low productivity results from slow TP, then a policy to induce technological innovation should be recommended to shift up the production frontier. If high rates of TP coexist with deteriorating TE, resulting in slow productivity growth, then a policy to increase the efficiency (with a known and Performance measured by Revealed Comparative Advantage (RCA) applied technology) is required and might include aiming at improvements in learning-by-doing Figure 2.6: Manufacturing RCA of Viet Nam and comparator countries processes and managerial practices. 1.40 The growth rate of TFP can be decomposed into four components, i.e. the change of productivity 1.11 1.20 1.10 1.10 1.07 1.02 1.05 1.05 due to TP, technical efficiency (TEC), scale effect (SEC) and factor reallocation effect (FAEC). That is, 0.98 1.00 0.82 TFP = TPit + TECit + SECit + FAECit 0.80 SEC captures the productivity improvement resulting from the evolution of scale economy of 0.60 industrial sectors. FAEC is referred to as factor allocative efficiency change. Under the two inputs 0.40 of capital and labour, FAEC consists of two kinds of efficiency due to labour reallocation and 0.20 capital reallocation. FAEC depend on the relative growth magnitude of two inputs. If the sum of 0.00 labour allocation efficiency and capital reallocation efficiency is substantial, factor shifts have an Viet Nam Bangladesh China RCA Indonesia India RCA Cambodia Thailand Malaysia Philippines impact on productivity. In turn, we can expect that the FAEC term reflects the industrial restricting RCA RCA RCA RCA RCA RCA RCA efforts at the reallocation of factors in order to increase industrial productivity and growth. 2005 2011 2016

Table 2.1: Decomposition of TFP Source: Authors’ calculation based on UN Comtrade database Figure 2.6 shows that Viet Nam’s manufacturing RCA steadily increased from 0.75 in 2005 to 0.963 in 2011, Growth in Growth in Growth in Growth in total and for the first time Viet Nam’s manufacturing RCA had a value (1.024) higher than unity in 2016. That is, Growth high-tech medium-tech low-tech Time period manufacturing in 2016, Viet Nam’s manufacturing comparative advantage was “revealed” in relation to countries, such components manufacturing manufacturing manufacturing sample (%) as Indonesia and Malaysia, with lower-than-unity RCAs (that “revealed” their comparative disadvantages sectors (%) sectors (%) sectors (%) in manufacturing). A more detailed analysis of manufacturing RCA improvements at sub-sector level is TP 2001-2015 6.82 6.86 6.39 6.05 provided later in this report. TEC 2001-2015 -0.28 -0.35 -0.44 -1.15 SEC 2001-2015 5.17 6.78 3.92 7.38 FAEC 2001-2015 18.04 -27.10 8.02 53.77 TFP 2001-2015 29.74 -13.82 17.89 66.05

36 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 37 Box 2.2: RCA as a measurement of international competitiveness

It is noted that RCA as a measurement has some limitations (for example, RCA index does not take into account the size/volume of countries’ exports) and comparisons between RCA values higher than unity of countries may mislead. For example, China’s RCA value almost at the same level of Bangladesh and Cambodia’s and just a little higher than Viet Nam’s may seem to counter the commonly accepted fact that China is the world manufacturing hub. Indeed, China’s share in total world manufacturing export is 14.16 percent, much higher than Bangladesh’s 0.24 percent, Cambodia’s 0.07 percent and Viet Nam’s 1.1 percent in 2016 (sources: Authors’ calculation using UN Comtrade database) and these numbers indicate that Viet Nam’s manufacturing export share in world manufacturing exports is higher than those of Bangladesh and Cambodia (though Viet Nam’s manufacturing RCA is lower). The RCA index calculation method, described above, is defined by its own manufacturing export share in its total exports (while the share of world manufacturing exports in total world exports is the same for all countries), and thus if a country has a high manufacturing export share in its total exports, the RCA is high. China’s manufacturing export share in its total exports is very high (98.65 percent), but Viet Nam’s, Cambodia’s and Bangladesh’s manufacturing export shares in their total exports are also very high (91.5, 98.25 and 99.02 percent respectively), (sources: Authors’ calculation using UN Comtrade database). The manufacturing RCA values of these countries are rather close. A country’s lower (but higher than unity) manufacturing RCA value may indicate the country has more significant shares of exports in agriculture and services sectors in its total exports than other countries (as is the case when comparing Viet Nam to Bangladesh and Cambodia).

IR4.0 readiness of manufacturing10

As noted in the first section of this report, several studies have suggested disruptive technologies and innovation will affect manufacturing in major ways. In this context, manufacturing firms must be ready to capture the opportunities and rise to the challenges associated with IR4.0.

A study on the IR4.0 readiness of manufacturing firms jointly conducted by MOIT, VASS and UNDP in late 2017 and early 2018 found the overwhelming majority of Viet Nam’s industrial enterprises, more than 85 percent, were not engaged in IR4.0 (termed by VDMA as “outsiders” to IR4.0), 13 percent of enterprises in the survey were at “beginner” level, only 2 percent were at “intermediate” level and negligible numbers were at “experienced” and “expert” levels and zero at “” “top performer” level.

With regards factors associated with IR4.0 readiness, a firm’s size was the strongest predictor of its participation in IR4.0, other factors being equal. In terms of ownership, SOEs have the highest level of participation in IR4.0, followed by foreign-invested enterprises. Domestic private firms have the lowest participation rates. However, differences between these types of ownership may be derived from other firms’ characteristics, crucial for IR4.0 readiness, i.e. level of capital intensity, employment size, technology level, industry concentration and level of technology used (Figures 2.7 and 2.8).

10 This section draws on key findings of a study on the IR4.0 readiness of manufacturing firms jointly conducted by MOIT, VASS and UNDP in late 2017 and early 2018. Under this study: (i) a survey of 2,659 industrial firms, of which the overwhelming majority were from manufacturing (sampled firms in water and electricity, oil and gas sub-sectors were also included in the survey), (ii) methodology developed by the German Mechanical Engineering Industry Association (Verband Deutscher Maschinen- und Anlagenbau – VDMA) was used for assessing firms’ IR4.0 readiness. Firms/sub-sectors are classified at “outsider” level if the score (based on scores in six dimensions: Strategy and Organization, Smart Factory, Smart Operations, Smart Product, Data-Driven Services and Employee Skills) is 0, from 0-1, beginner; 1-2, intermediate; 2-3, experienced; 3-4, expert; and top performer: 4-5.

38 Productivity and Competitiveness of Viet Nam’s Enterprises Figure 2.7: Proportions of SOEs, domestic private and FDI firms at different IR4.0 readiness Box 2.2: RCA as a measurement of international competitiveness levels

100% 89% It is noted that RCA as a measurement has some limitations (for example, RCA index does not 81% take into account the size/volume of countries’ exports) and comparisons between RCA values 80% higher than unity of countries may mislead. For example, China’s RCA value almost at the same 60% 47% 37% level of Bangladesh and Cambodia’s and just a little higher than Viet Nam’s may seem to counter 40% 16% 16% the commonly accepted fact that China is the world manufacturing hub. Indeed, China’s share in 20% 10% 1% 3% 1% 0% 0% 0% 0% 0% 0% 0%0% 0%0% total world manufacturing export is 14.16 percent, much higher than Bangladesh’s 0.24 percent, 0% Cambodia’s 0.07 percent and Viet Nam’s 1.1 percent in 2016 (sources: Authors’ calculation using NgoàiOutsider cuộc MớiBeginner bắt đầu TrungIntermediate bìnhExperienced Kinh nghiệmExpert Chuyên giaTop ĐiPerformer đầu UN Comtrade database) and these numbers indicate that Viet Nam’s manufacturing export State-ownedDNNN DN tưDomestic nhân trong Private nước FDI share in world manufacturing exports is higher than those of Bangladesh and Cambodia (though Source: MOIT-VASS-UNDP (2018) Viet Nam’s manufacturing RCA is lower). The RCA index calculation method, described above, is defined by its own manufacturing export share in its total exports (while the share of world Figure 2.8 shows the larger firms (with other characteristics equal) correspond with higher scores manufacturing exports in total world exports is the same for all countries), and thus if a country (higher IR4.0 readiness level). has a high manufacturing export share in its total exports, the RCA is high. China’s manufacturing Figure 2.8: Readiness scores by size of firms export share in its total exports is very high (98.65 percent), but Viet Nam’s, Cambodia’s and Bangladesh’s manufacturing export shares in their total exports are also very high (91.5, 98.25 1.21.2 1.051.05 and 99.02 percent respectively), (sources: Authors’ calculation using UN Comtrade database). The 11 94.9% 95.4% manufacturing RCA values of these countries are rather close. A country’s lower (but higher than 0.8 unity) manufacturing RCA value may indicate the country has more significant shares of exports 0.80.8 76.5% in agriculture and services sectors in its total exports than other countries (as is the case when 0.67 57.4% 0.60.6 57.4% comparing Viet Nam to Bangladesh and Cambodia). ReadinessĐiểm sẵn sàng score 0.39

Readiness score % of Outsider Điểm sẵn sàng 0.40.4 % "Đứng ngoài cuộc" IR4.0 readiness of manufacturing10 0.20.2

As noted in the first section of this report, several studies have suggested disruptive technologies and 0 0 More than From 200-300 from 20-200 Less than 10 workers innovation will affect manufacturing in major ways. In this context, manufacturing firms must be ready Trên300workers 300 lao động Từ 200workers-300 lao động Từ 20workers-200 lao động Dưới 10 lao động to capture the opportunities and rise to the challenges associated with IR4.0. QuiEnterprise mô doanh size nghiệp (number (số of lao workers) động)

A study on the IR4.0 readiness of manufacturing firms jointly conducted by MOIT, VASS and UNDP in late Source: MOIT-VASS-UNDP (2018) 2017 and early 2018 found the overwhelming majority of Viet Nam’s industrial enterprises, more than 85 Furthermore, the study found the overwhelming majority of manufacturing firms were unfamiliar with percent, were not engaged in IR4.0 (termed by VDMA as “outsiders” to IR4.0), 13 percent of enterprises in IR4.0’s disruptive technologies. Figure 2.9 shows a small percentage of enterprises had applied typical the survey were at “beginner” level, only 2 percent were at “intermediate” level and negligible numbers IR4.0 technology. Only cloud computing and device-to-device/products were used by more than 15 and were at “experienced” and “expert” levels and zero at “” “top performer” level. 12 percent, respectively of surveyed firms, with the usage rate of other technologies below 10 percent. With regards factors associated with IR4.0 readiness, a firm’s size was the strongest predictor of its Less than 1 percent of manufacturing firms used 3D printing and big data in their businesses. participation in IR4.0, other factors being equal. In terms of ownership, SOEs have the highest level Figure 2.9: Level of enterprises applying IR4.0 technologies of participation in IR4.0, followed by foreign-invested enterprises. Domestic private firms have the 100%100% lowest participation rates. However, differences between these types of ownership may be derived from 14.814.8 12.612.6 90%90% 21 21.8 22.7 22.9 other firms’ characteristics, crucial for IR4.0 readiness, i.e. level of capital intensity, employment size, 38.1 80%80% 45 technology level, industry concentration and level of technology used (Figures 2.7 and 2.8). 70%70% 60%60% 81.5 50%50% 65.665.6 68.968.9 64.6 40%40% 70.1 72.2 72.8 58.7 30%30% 51.4 20%20% 10%10% 14.1 0%0% 10 This section draws on key findings of a study on the IR4.0 readiness of manufacturing firms jointly conducted by MOIT, VASS and Điện toán Kết nối thiết Công nghệ Công nghệ Công nghệ Công nghệ Trí tuệ nhân Công nghệ Phân tích và UNDP in late 2017 and early 2018. Under this study: (i) a survey of 2,659 industrial firms, of which the overwhelming majority were đám mây bị với thiết cảm biến thiết bị đầu định vị thời nhận dạng tạo chế tạo đắp quản trị dữ from manufacturing (sampled firms in water and electricity, oil and gas sub-sectors were also included in the survey), (ii) methodology bị/sản phẩm cuối di động gian thực bằng sóng dần (in 3D) liệu (Big vô tuyến (RFID) Data) developed by the German Mechanical Engineering Industry Association (Verband Deutscher Maschinen- und Anlagenbau – VDMA) was used for assessing firms’ IR4.0 readiness. Firms/sub-sectors are classified at “outsider” level if the score (based on scores in six Đang áp dụng Sẽ áp dụng Không có kế hoạch Không liên quan dimensions: Strategy and Organization, Smart Factory, Smart Operations, Smart Product, Data-Driven Services and Employee Skills) is 0, from 0-1, beginner; 1-2, intermediate; 2-3, experienced; 3-4, expert; and top performer: 4-5. Applied Will apply No plan Not relevant Source: MOIT-VASS-UNDP (2018) 38 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 39 While many IR4.0 disruptive technologies will remain out of reach for many firms globally in the foreseeable future, cloud computing has become increasingly popular in the business sector, particularly among SMEs. The is due to this on-demand service’s reliance on third parties’ digital resources to substantially cut firms’ operating costs. This is crucial for firms to stay competitive in an increasingly digitized world. However, with mean and median values of 14 and 13 percent, the adoption rate of cloud computing by Viet Nam’s manufacturing firms appears modest.

Size and ownership also make a difference. Table 2.2 shows the utilization rate and scale increases in tandem (except for groups of 10-200 and 200-300 employees). In terms of ownership, this proportion differed little between foreign-invested enterprises and private domestic firms.

Table 2.2: Firms’ usage of cloud computing by size and ownership (%)

In use To be used No plan Not relevant All By size (%) Less than 10 workers 10.8 2.4 73.8 13.0 100 10-200 workers 17.0 4.3 64.6 14.0 100 200-300 workers 16.1 4.4 58.3 21.2 100 > 300 workers 22.3 17.5 32.9 27.2 100 By ownership (%) SOEs 28.7 22.4 23.3 25.6 100 Private firms 14.5 4.2 68.3 12.9 100

Source: MOIT-VASS-UNDP (2018)

2.2.2 Performance at Sub-sector Level Important characteristics of manufacturing sub-sectors11

Before assessing productivity and competitiveness performance at sub-sector level, the following section highlights important characteristics of manufacturing sub-sectors, such as: (i) size of employment and share in manufacturing employment, (ii) revenue, VA volumes, share of manufacturing VA and revenue, (iii) FDI, SOE, domestic private and large firms’ and SMEs’ shares of sub-sector revenues, VA and employment and (iv) imports and exports as well as sub-sectors’ IR4.0 readiness. These characteristics help explain the analysis of productivity and competitiveness performance at sub-sector level and deliver recommendations synergized with sub-sector characteristics.

Employment

Figure 2.10 paints a picture of employment size of manufacturing sub-sectors. Apparel is the largest employer with almost 1.4 million workers in 2016, up from 0.9 million in 2011. Leather is second, with almost 1.2 million workers in 2016, up from 0.8 million in 2011. These million-worker sub-sectors are followed by electronics, computers and optical products as well as food processing, each employing more than 500,000 workers. In contrast, some sub-sectors only employ more than 5,000 workers (coke- refined petroleum products-nuclear fuel) or approximately 11,000 workers (tobacco).

11 While Enterprise Census data allows for an accurate analysis of trends and relative ratios, the absolute values - such as ones used in this section - must be treated with caution.

40 Productivity and Competitiveness of Viet Nam’s Enterprises While many IR4.0 disruptive technologies will remain out of reach for many firms globally in the foreseeable Figure 2.10: Employment size of manufacturing sub-sectors future, cloud computing has become increasingly popular in the business sector, particularly among Number ofSố workers lao động of trong manufacturing các tiểu ngành sub-sectors chế tạo SMEs. The is due to this on-demand service’s reliance on third parties’ digital resources to substantially cut firms’ operating costs. This is crucial for firms to stay competitive in an increasingly digitized world. WearingWearing apparel, apparel, fur fur Leather,Leather, leather leather products products and and footwear footwear However, with mean and median values of 14 and 13 percent, the adoption rate of cloud computing by ManufactureManufacture of of computer, computer, electronic electronic and and optical optical Food Food Viet Nam’s manufacturing firms appears modest. FurnitureFurniture FabricatedFabricated metal metal products products Size and ownership also make a difference. Table 2.2 shows the utilization rate and scale increases in RubberRubber and and plastics plastics products products Non-metallicNon-metallic mineral mineral products products tandem (except for groups of 10-200 and 200-300 employees). In terms of ownership, this proportion TextilesTextiles ElectricalElectrical equipment equipment differed little between foreign-invested enterprises and private domestic firms. OtherOther manufacturing manufacturing Wood,Wood, bamboo bamboo products products (excl. (excl. furniture) furniture) Table 2.2: Firms’ usage of cloud computing by size and ownership (%) MotorMotor vehicles, vehicles, trailers, trailers, semi-trailers semi-trailers ChemicalsChemicals and and chemical chemical products products OtherOther vehicles vehicles In use To be used No plan Not relevant All PaperPaper and and paper paper products products BasicBasic metals metals By size (%) PrintingPrinting and and publishing publishing MachineryMachinery and and equipment equipment n.e.c. n.e.c. Less than 10 workers 10.8 2.4 73.8 13.0 100 beveragesbeverages Medical,Medical, precision precision and and optical optical instruments instruments 10-200 workers 17.0 4.3 64.6 14.0 100 RepairRepair and and installation installation of of machinery machinery and and equipment equipment TobaccoTobacco products products 200-300 workers 16.1 4.4 58.3 21.2 100 Coke,Coke, refined refined petroleum petroleum products, products, nuclear nuclear fuel fuel > 300 workers 22.3 17.5 32.9 27.2 100 00 200000200000 400000400000 600000600000 800000800000 10000001000000 12000001200000 14000001400000 16000001600000 20162016 20112011 By ownership (%) SOEs 28.7 22.4 23.3 25.6 100 Source: Authors’ calculation based on 2017 Enterprise Census data Private firms 14.5 4.2 68.3 12.9 100 While almost all sub-sectors (except non-metallic mineral products and tobacco) experienced increases Source: MOIT-VASS-UNDP (2018) in workers between 2011-2016, the rises varied. They were highest in the three sub-sectors with the most workers (apparel, leather and especially electronics, computer and optical products) resulting in 2.2.2 Performance at Sub-sector Level higher shares in total manufacturing employment in 2016 compared to 2011. Slightly higher shares were also observed in motor vehicles-trailers-semitrailers, textiles, medical-optical-precision equipment, Important characteristics of manufacturing sub-sectors11 while the remaining sub-sectors experienced declining shares (Figure 2.11). Before assessing productivity and competitiveness performance at sub-sector level, the following section Figure 2.11: Sub-sector shares in manufacturing employment (%) highlights important characteristics of manufacturing sub-sectors, such as: (i) size of employment and share in manufacturing employment, (ii) revenue, VA volumes, share of manufacturing VA and revenue, Wearing apparel, Mayfur Leather, leather products and footDawe giầyar (iii) FDI, SOE, domestic private and large firms’ and SMEs’ shares of sub-sector revenues, VA and Manufacture of computer, electronic and oĐiệnptic atửl Chế biến thực Foophẩmd employment and (iv) imports and exports as well as sub-sectors’ IR4.0 readiness. These characteristics FurniĐồtu rgỗe help explain the analysis of productivity and competitiveness performance at sub-sector level and FabricatedChế me tbiếnal pro kimduc loạits Rubber and plaCaostics su pro vàduc nhựats deliver recommendations synergized with sub-sector characteristics. SảnNo phẩmn-me khoángtallic mi sảnnera phil pro kimduc loạits TextilDệtes Employment ElectricalThiết equip bịm điệnent SảnOth phẩmer ma nufchếa tạoctu rkhácing WSảnood ,phẩm bamboo gỗ trepro (khôngducts ( excbaol. gồm furn itđồur gỗ)e) Figure 2.10 paints a picture of employment size of manufacturing sub-sectors. Apparel is the largest Motor vehicles, trailers, Xesem cói- tđộngraile rcơs CheHóamica chấtls and và csảnhem phẩmical p hóaroduc chấtts employer with almost 1.4 million workers in 2016, up from 0.9 million in 2011. Leather is second, with Phương tiện giaoOth ethôngr vehi clkháces almost 1.2 million workers in 2016, up from 0.8 million in 2011. These million-worker sub-sectors are Paper and paper producGiấyts KimBa sloạiic m cơet bảnals followed by electronics, computers and optical products as well as food processing, each employing PrintingXuất and bảnpubl vàish inin ấng MáyMac móchine rphụy and tùng equip chưame phânnt n.e.c loại. more than 500,000 workers. In contrast, some sub-sectors only employ more than 5,000 workers (coke- Medical, precision and optical instruĐồm euốngnts refined petroleum products-nuclear fuel) or approximately 11,000 workers (tobacco). Công cụ y tế, chính xác vàb quangeverage họcs Repair and installation of Sửamac hichữanery và a ndlắp e đặtqui pthiếtmen bịt Tobacco prThuốcoduct slá Than Coke,Coke, các refi sảnned phẩm petro hóaleum dầu, pro ducnhiênts, liệunucl hạtear nhânfuel

0 200000 400000 600000 800000 1000000 1200000 14000001400000 16000001600000

2016 2011

Source: Authors’ calculation based on 2017 Enterprise Census data

Figure 2.12 shows FDI employed the highest number of workers in manufacturing (55.52 percent), followed by domestic private firms (47.88 percent). SOEs employed the least number of workers (3.6 11 While Enterprise Census data allows for an accurate analysis of trends and relative ratios, the absolute values - such as ones used in this section - must be treated with caution. percent).

40 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 41 In a few sub-sectors, where SOEs play somewhat significant roles, their share in sub-sector employment was modest (except the dominate share of 94.02 percent in tobacco), 36.17 percent in coke-petroleum products-nuclear fuel, 28.40 percent in repair and installation of machinery and equipment, 21.1 percent in chemicals, 15.34 percent in basic metals, 13.3 percent in beverages, 12.95 percent in other vehicles and 10.56 percent in non-metallic mineral products. Notably all these sub-sectors employ less than 150,000 workers, while only non-metallic mineral products accounts for 300,000 workers in 2016 (Figure -2.10).

In contrast, FDI dominates in terms of number of workers in around half of large-employing manufacturing sub-sectors. Its shares in sub-sector employment are: 96.13 percent in computer-electronics and optical equipment (third-largest employing sub-sector), 81.66 percent in leather (second-largest employing sub-sector), 56.3 percent in apparel (largest employing sub-sector), 79.02 percent in motors-trailers- semitrailers, 66.53 percent in other vehicles, 77.88 percent in other manufacturing, 77.62 percent in electric equipment, 48.6 percent in textiles, 52.24 percent in furniture, 49.94 percent in machinery and equipment n.e.c., 38.31 percent in rubber-plastics – all with “medium” numbers of workers.

Domestic private firms generally dominate in terms of sub-sectors with small and medium numbers of workers: 82.97 percent in wood and bamboo products (excluding furniture), 75.99 percent in non- metallic mineral products, 75.84 percent medical-precision and optical equipment, 73.76 percent in food (fourth-largest employing sub-sector), 70.85 percent in printing, 66.65 percent in paper, 64.72 percent in repair and installation of machines and equipment, 61.76 percent in fabricated metal products (sixth- largest employing sub-sector), 56.36 percent in basic metals, 53.15 percent in beverages and 48.64 percent in rubber-plastics.

Figure 2.12: Sub-sector employment structure by ownership (2016)

SảnWoo phẩmd, bamboo gỗ tre p (khôngroducts bao (exc gồml. fu rđồnit gỗ)ure) Sản phẩmNon-m khoángetallic misảnne phiral pkimroduc loạits MedicCôngal, p rcụec yis tế,ion chính and o xácptic avàl i quangnstrum họcents Chế biến thực phẩmFood PrintiXuấtng and bản pu vàbl iinsh ấning Paper and paper producGiấyts Repair and installation ofSửa m acchữahine vàry lắpand đặt eq uthiếtipm enbị t FabricatChếed m biếnetal pkimroduc loạits KimB loạiasic cơ m ebảntals bĐồeve uốngrages Rubber and plCaoasti sucs vàpro nhựaducts ChHóaemi chấtcals andvà sản ch ephẩmmical hóa pro ducchấtts MáyM amócchin phụery andtùng e chưaquipme phânnt n.e.c loại . Than Coke,Cok cáce, r esảnfined phẩm petro hóaleu dầu,m p ronhiênduct liệus, nucl hạte anhânr fuel FurĐồnit gỗure TexDệttiles Wearing appareMayl, fur Ngành chếM biếnanuf chếactu tạoring ElectricaThiếtl equ bịip điệnment SảnO phẩmther ma chếnuf tạoac tkhácuring Phương tiện giaoOth thônger ve hikháccles Leather, leather products and fDaoot giầywear Motor vehicles, trailersXe, s cóem độngi-trai lecơrs Manufacture of computer, electronic andĐiện opt itửcal TobaccoThuốc produ ctlás

0%0% 10%10% 20%20% 30%30% 40%40% 50%50% 60% 70% 80% 90% 100%

FDIFDI SOESOE DomesticDomestic privateprivate

Source: Authors’ calculation based on 2017 Enterprise Census data

Figure 2.13 shows big firms dominate in terms of workers in large and medium-employing sub-sectors where FDI is prominent, while small firms are visible in small and medium-employing sub-sectors where domestic private firms are influential. The few exceptions include tobacco, coke-refined petroleum products-nuclear fuel, basic metals and repair and installation of machinery and equipment, in which large (SOE) firms have bigger shares in employment.

42 Productivity and Competitiveness of Viet Nam’s Enterprises In a few sub-sectors, where SOEs play somewhat significant roles, their share in sub-sector employment Figure 2.13: Large and small firm shares in sub-sector employment (%), 2016 (Source: Authors’ was modest (except the dominate share of 94.02 percent in tobacco), 36.17 percent in coke-petroleum calculation based on 2017 Enterprise Census data) products-nuclear fuel, 28.40 percent in repair and installation of machinery and equipment, 21.1 percent PrintinXgu ấtand bả pun vàbli shin inấng in chemicals, 15.34 percent in basic metals, 13.3 percent in beverages, 12.95 percent in other vehicles SWảoon phẩmd, bamboo gỗ tr ep ro(kduchôngts b(excao gồl. fmur đồnit urgỗe)) FabricatedCh mế biếnetal p kiromduc loạtis and 10.56 percent in non-metallic mineral products. Notably all these sub-sectors employ less than Paper and paper producGiấtys Repair and installation ofSử maac chhiữnea vàry alắpnd đ eặtqu thipiếmten bịt 150,000 workers, while only non-metallic mineral products accounts for 300,000 workers in 2016 (Figure MáyMa mchóinc pheryụ and tùng e qchuipưame phânt nn.e.c loại. ChHeómia chấtcals andvà s cảhne phẩmmical hpóroaduc chấtts -2.10). Sản Nophẩmn-m kehtaolálingc mi sảnen rphal ip kiromduc loạtis Rubber and plCaastio cssu p vàro nduchựtas beĐồve uốnragegs Than CokeCok, cáce, r esfiảnedn phẩm petro hóleua dmầ u,pro nduchiênt s,liệu nucl hạtea nr hâfuenl In contrast, FDI dominates in terms of number of workers in around half of large-employing manufacturing KimB laosạici cơme btảanls TextiDệlets sub-sectors. Its shares in sub-sector employment are: 96.13 percent in computer-electronics and optical MedicCaôngl, pr cecụ iys iotế,n chíandn ho ptxácica vàl ins qutarumng enthọcs FurĐồnit ugrỗe equipment (third-largest employing sub-sector), 81.66 percent in leather (second-largest employing Chế biến thực phẩFoomd Ngành chMế abiếnnuf acchtếu ritạnog sub-sector), 56.3 percent in apparel (largest employing sub-sector), 79.02 percent in motors-trailers- SảnO thphẩmer ma chnufế tạacot kuhárincg Phương tiện giaOoth theôngr vehi kháclecs semitrailers, 66.53 percent in other vehicles, 77.88 percent in other manufacturing, 77.62 percent in ElectricaThl eiếqut bịip mđiệennt Motor vehicles, trailersX, es ecmó iđộng-traile crơs electric equipment, 48.6 percent in textiles, 52.24 percent in furniture, 49.94 percent in machinery and Wearing appareMal, fuyr Tobacco prThouốducct lás equipment n.e.c., 38.31 percent in rubber-plastics – all with “medium” numbers of workers. Manufacture of computer, electronic andĐiện opti ctửal Leather, leather products and footDa wegiầayr Domestic private firms generally dominate in terms of sub-sectors with small and medium numbers 00%% 1010%% 20%20% 30%30% 40%40% 50%50% 60%60% 70%70% 80%80% 90%90% 100%100% of workers: 82.97 percent in wood and bamboo products (excluding furniture), 75.99 percent in non- Large FirmFirms SMEsSME metallic mineral products, 75.84 percent medical-precision and optical equipment, 73.76 percent in food Source: Authors' calculation based on 2017 Census data Enterprise (fourth-largest employing sub-sector), 70.85 percent in printing, 66.65 percent in paper, 64.72 percent in repair and installation of machines and equipment, 61.76 percent in fabricated metal products (sixth- largest employing sub-sector), 56.36 percent in basic metals, 53.15 percent in beverages and 48.64 Box 2.3: Firms’ size and ownership percent in rubber-plastics. Enterprise Census data points to the large size of FDI firms, as evidenced by the negative correlation Figure 2.12: Sub-sector employment structure by ownership (2016) between FDI and SME labour shares across manufacturing sub-sectors: the coefficient of

Sản phẩm gỗ tre (không bao gồm đồ gỗ) correlation estimated at minus 0.55. In sub-sectors with higher participations of domestic private Sản phẩm khoáng sản phi kim loại Công cụ y tế, chính xác và quang học firms (lower participation of FDI), domestic firms are often small- and medium-sized, while SOEs Chế biến thực phẩm Xuất bản và in ấn tend to be large. However, “small- and medium-sized enterprises” in this report are based on the Giấy Sửa chữa và lắp đặt thiết bị definition within Viet Nam’s Enterprise Law, which specifies that SMEs are registered firms “with Chế biến kim loại Kim loại cơ bản Đồ uống Cao su và nhựa “The category of micro, small- and medium-sized enterprises is made up of enterprises which Hóa chất và sản phẩm hóa chất Máy móc phụ tùng chưa phân loại employ fewer than 250 persons and which have an annual turnover not exceeding 50 million Than Coke, các sản phẩm hóa dầu, nhiên liệu hạt nhân Đồ gỗ euros, and/or an annual balance sheet total not exceeding 43 million euros.” Dệt May Ngành chế biến chế tạo Thiết bị điện Sản phẩm chế tạo khác Phương tiện giao thông khác Revenue Da giầy Xe có động cơ Điện tử Thuốc lá In terms of revenue, all sub-sectors (except coke-refined petroleum products-nuclear fuel) enjoyed

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% increases during 2011-2016, with the robust electronics sub-sector having pushed food to second in

FDI SOE Domestic private 2016 (Figure 2.14). Source: Authors’ calculation based on 2017 Enterprise Census data The biggest revenue jump in electronics and relatively large rises in leather-footwear, wearing apparel, motor vehicles-trailers-semitrailers resulted in increased shares in total manufacturing revenue, while Figure 2.13 shows big firms dominate in terms of workers in large and medium-employing sub-sectors other sub-sectors declined except optical and precision equipment (Figure 2.15). where FDI is prominent, while small firms are visible in small and medium-employing sub-sectors where domestic private firms are influential. The few exceptions include tobacco, coke-refined petroleum products-nuclear fuel, basic metals and repair and installation of machinery and equipment, in which large (SOE) firms have bigger shares in employment.

42 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 43 Figure 2.14: Revenue of manufacturing sub-sectors (VND billion, 2016 prices)t

Manufacture of computer, electronic andĐiện opti ctửal Chế biến thực phẩmFood FabricatedChế m biếnetal pkimroduc loạits KimB loạiasic cơ m ebảntals Wearing appareMayl, fur Sản Nophẩmn-m khoángetallic mi sảnne phiral pkimroduc loạits Leather, leather products and footDa giầywear Rubber and plCaoasti cssu pvàro nhựaducts Motor vehicles, trailersXe, s ecóm độngi-traile cơrs ChHóaemi chấtcals andvà sản che phẩmmical hóapro ducchấtts TextiDệtles ElectricaThiếtl equ bịip mđiệnent Phương tiện giaoOth thônger ve hikháccles FurĐồnit ugỗre bĐồeve uốngrages Paper and paper producGiấyts SảnWoo phẩmd, bamboo gỗ tre p ro(khôngducts bao (exc gồml. fur đồnit urgỗ)e) Than Coke,Cok cáce, r esảnfined phẩm petro hóaleu dầu,m pro nhiênduct s,liệu nucl hạtea nhânr fuel MáyMa mócchin phụery and tùng e qchưauipme phânnt n.e.c loại. SảnO phẩmther ma chếnuf tạoact khácuring MedicCôngal, pr cụec iys iotế,n chínhand o ptxácic avàl i nsquangtrum họcents PrintinXuấtg and bản pu vàbli insh ấning Tobacco Thuốcproduct lás Repair and installation ofSửa mac chữahine vàry lắpand đặt eq uthiếtipmen bịt

00 40000004000000 8000000 12000000 16000000

20162016 20112011

Figure 2.15:: Sub-sector revenue shares in manufacturing revenue (%)

Manufacture of computer, electronic and Điệnoptical tử Chế biến thực phẩmFood FabricatedChế metal biến products kim loại KimBasic loại metalscơ bản Wearing apparel,May fur SảnNon-metallic phẩm khoáng mineral sản phi products kim loại Leather, leather products and footwearDa giầy Rubber and plasticsCao su products và nhựa Motor vehicles, trailers,Xe semi-trailers có động cơ ChemicalsHóa chất and và sảnchemical phẩm products hóa chất TextilesDệt ElectricalThiết equipment bị điện Phương tiện giaoOther thông vehicles khác FurnitureĐồ gỗ beveragesĐồ uống Paper and paper productsGiấy Wood,Sản phẩm bamboo gỗ tre products (không (excl. bao gồm furniture) đồ gỗ) Than Coke,Coke, các refined sản phẩm petroleum hóa dầu, products, nhiên liệu nuclear hạt nhân fuel MáyMachinery móc phụ and tùng equipment chưa phân n.e.c. loại SảnOther phẩm manufacturing chế tạo khác Medical,Công precision cụ y tế, and chính optical xác và instruments quang học PrintingXuất and bản publishing và in ấn Tobacco productsThuốc lá Repair and installation of Sửamachinery chữa và and lắp equipmentđặt thiết bị

0.00% 5.00% 10.00% 15.00% 20.00% 25.00%

2016 2011

Source: Authors’ calculation based on 2017 Enterprise Census data

Similar to the employment structure, FDI had the largest share (54.66 percent) in total manufacturing revenue, followed by domestic private firms (34.79 percent) and SOEs (10.55 percent) in 2016. Naturally, shares of FDI revenue were robust in sub-sectors with high employment shares, but FDI had: (i) relatively high revenue shares in some sub-sectors with medium employment shares (chemicals, medical and precision equipment and fabricated metal) and (ii) medium revenue shares where its employment share was low (non-metallic mineral products and tobacco). SOEs, similar to their employment share, only have higher revenue shares in tobacco and coke-refined petroleum-nuclear fuel and medium revenue shares in beverages and repair-installation of equipment. Domestic firms’ revenue shares were: (i) high in wood and bamboo (excluding furniture), paper, medical and precision equipment, rubber-plastics, non-metallic mineral products, basic metal, fabricated metal products, furniture and repair-installation of machinery and (ii) at a medium level in beverages, textiles, wearing apparel, chemicals, electrical equipment, machinery and equipment n.e.c and motor vehicles (Figure 2.16 - upper panel).

Figure 2.16 (lower panel) shows that large enterprises dominated almost all sub-sectors in terms of revenue. Large firms’ share of total manufacturing revenue was 86.11 percent and only three sub-sectors (printing, wood (excluding furniture) and repair-installation of machinery) had SME revenue shares of around 60 percent.

44 Productivity and Competitiveness of Viet Nam’s Enterprises Figure 2.14: Revenue of manufacturing sub-sectors (VND billion, 2016 prices)t Figure 2.16: (upper panel): Domestic private firms, SOEs and FDI shares in sub-sector revenue

Điện tử (%), 2016 Chế biến thực phẩm Chế biến kim loại Ngành chếManufacturing biến chế tạo Kim loại cơ bản Wood,Sản phẩmbamboo gỗ productstre (không (excl. bao gồmfurniture) đồ gỗ) May Medical,Công precision cụ y tế, and chính optical xác instrumentsvà quang học Sản phẩm khoáng sản phi kim loại Chế biến thựcFood phẩm Da giầy FabricatedChế metal biến products kim loại Cao su và nhựa PrintingXuất and bảnpublishing và in ấn Xe có động cơ SảnNon-metallic phẩm khoáng mineral sản phi products kim loại Hóa chất và sản phẩm hóa chất Paper and paper productsGiấy Dệt Repair and installation of machinerySửa chữa và and lắp equipment đặt thiết bị Thiết bị điện KimBasic loại metals cơ bản Phương tiện giao thông khác Rubber and plasticsCao suproducts và nhựa Đồ gỗ FurnitureĐồ gỗ Đồ uống Wearing apparel, furMay Giấy Electrical Thiếtequipment bị điện Sản phẩm gỗ tre (không bao gồm đồ gỗ) ChemicalsHóa chất and và chemical sản phẩm products hóa chất Than Coke, các sản phẩm hóa dầu, nhiên liệu hạt nhân MáyMachinery móc phụ and tùng equipment chưa phân n.e.c. loại Máy móc phụ tùng chưa phân loại TextilesDệt Sản phẩm chế tạo khác Motor vehicles, trailers, semi-trailersXe có động cơ Công cụ y tế, chính xác và quang học beveragesĐồ uống Xuất bản và in ấn SảnOther phẩm manufacturing chế tạo khác Thuốc lá Leather, leather products and footwearDa giầy Sửa chữa và lắp đặt thiết bị Phương tiện giaoOther thông vehicles khác Coke, refined petroleum products, nuclear fuel 0 4000000 8000000 12000000 16000000 Than Coke, các sản phẩm hóa dầu, nhiên liệu hạt nhân Manufacture of computer, electronic and opticalĐiện tử Tobacco productsThuốc lá 2016 2011 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

Figure 2.15:: Sub-sector revenue shares in manufacturing revenue (%) FDI SOEsDNNN DomesticDN tư nhân private trong firmsnước

Điện tử Chế biến thực phẩm Figure 2.16: (lower panel): Large and SME shares in sub-sector revenue (%), 2016 Chế biến kim loại Kim loại cơ bản Ngành chếManufacturing biến chế tạo May PrintingXuất and bản publishing và in ấn Sản phẩm khoáng sản phi kim loại Wood,Sản phẩm bamboo gỗ tre products (không bao(excl. gồm furniture) đồ gỗ) Da giầy Repair and installation ofSửa machinery chữa và andlắp đặtequipment thiết bị Cao su và nhựa Paper and paper productsGiấy Xe có động cơ FabricatedChế metal biến products kim loại Hóa chất và sản phẩm hóa chất Rubber and plasticsCao su products và nhựa Dệt MáyMachinery móc phụ and tùng equipment chưa phân n.e.c. loại Thiết bị điện SảnNon-metallic phẩm khoáng mineral sản phi products kim loại Phương tiện giao thông khác FurnitureĐồ gỗ Đồ gỗ ChemicalsHóa chất and và sản chemical phẩm hóaproducts chất Đồ uống SảnOther phẩm manufacturing chế tạo khác Giấy TextilesDệt Sản phẩm gỗ tre (không bao gồm đồ gỗ) Medical,Công precision cụ y tế, andchính optical xác và instruments quang học Than Coke, các sản phẩm hóa dầu, nhiên liệu hạt nhân Chế biến thực phẩmFood Máy móc phụ tùng chưa phân loại ElectricalThiết equipment bị điện Sản phẩm chế tạo khác KimBasic loại cơmetals bản Công cụ y tế, chính xác và quang học Wearing apparel,May fur Xuất bản và in ấn Phương tiện giaoOther thông vehicles khác Thuốc lá beveragesĐồ uống Sửa chữa và lắp đặt thiết bị Motor vehicles, trailers,Xe semi-trailers có động cơ Leather, leather products and footwear 0.00% 5.00% 10.00% 15.00% 20.00% 25.00% Da giầy Than Coke,Coke, các refined sản phẩm petroleum hóa dầu, products, nhiên liệu nuclear hạt nhân fuel Tobacco productsThuốc lá 2016 2011 Manufacture of computer, electronic andĐiện optical tử Source: Authors’ calculation based on 2017 Enterprise Census data 0% 10% 20% 30% 40% 50% 60%60% 70%70% 80%80% 90%90% 100%100% DoanhLarge Firm nghiệp lớn SMEDN VVN Similar to the employment structure, FDI had the largest share (54.66 percent) in total manufacturing Source: Authors’ calculation based on 2017 Enterprise Census data revenue, followed by domestic private firms (34.79 percent) and SOEs (10.55 percent) in 2016. Naturally, shares of FDI revenue were robust in sub-sectors with high employment shares, but FDI had: (i) relatively Value added high revenue shares in some sub-sectors with medium employment shares (chemicals, medical and precision equipment and fabricated metal) and (ii) medium revenue shares where its employment share In terms of value added, food processing was the biggest sub-sector in 2011, but overtaken by electronics was low (non-metallic mineral products and tobacco). SOEs, similar to their employment share, only in 2016 and the wearing apparel sub-sector was only the second and third largest sub-sector in 2011 and have higher revenue shares in tobacco and coke-refined petroleum-nuclear fuel and medium revenue 2016, respectively (Figure 2.17 - upper panel)12. shares in beverages and repair-installation of equipment. Domestic firms’ revenue shares were: (i) high In terms of shares in total MVA, electronics’ performance was exceptional as its share more than doubled, in wood and bamboo (excluding furniture), paper, medical and precision equipment, rubber-plastics, from 7.4 percent in 2011 to 15.65 percent and thus it was the largest manufacturing sub-sector in 2016 non-metallic mineral products, basic metal, fabricated metal products, furniture and repair-installation (in absolute volume of VA and share in total MVA). The basic metal products sub-sector is another rising of machinery and (ii) at a medium level in beverages, textiles, wearing apparel, chemicals, electrical star, with its VA share rising from 3.9 percent in 2011 to 10.84 percent to claim the second biggest share equipment, machinery and equipment n.e.c and motor vehicles (Figure 2.16 - upper panel). in MVA in 2016, leaving food processing’s VA share (reduced from 13.36 percent in 2011 to 9.53 percent Figure 2.16 (lower panel) shows that large enterprises dominated almost all sub-sectors in terms of in 2016) in third place. Notably, the VA shares of wearing apparel, non-metallic mineral products and revenue. Large firms’ share of total manufacturing revenue was 86.11 percent and only three sub-sectors leather-footwear having ranked second, third and fifth, respectively in 2011 slipped in 2016 to fifth, (printing, wood (excluding furniture) and repair-installation of machinery) had SME revenue shares of fourth and sixth, respectively (Figure 2.17 - lower panel). Figure 2.17 – upper panel also shows that while around 60 percent. 12 As noted in Section 2.1, due to estimation methods and data problems, MVA is used to measure the growth and structure, but not the level. The VA values provided in this paragraph and upper panel of Figure 2.17 are only for an illustration setting basis for the “structure” of VA distribution across manufacturing sub-sectors in the lower panel of Figure 2.17.

44 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 45 all sub-sectors (except coke-refined petroleum-nuclear fuel) experienced VA rises between 2011-2016, the higher 2016 VA shares (meaning relatively faster VA growth compared to other manufacturing sub- sectors) compared to 2011 shares were only observed in some sub-sectors (electronics, basic metal, rubber-plastics, textiles, electrical equipment, motor vehicles and other manufacturing).

Figure 2.17: (upper panel): Value added of manufacturing sub-sector, VND billion (2016 prices)

Điện tử Manufacture of computer, electronic and optical Chế biến thực Foodphẩm KimBasic loại metals cơ bản Wearing apparel, Mayfur SảnNon-metallic phẩm khoáng mineral sản phi products kim loại Leather, leather products and footwearDa giầy Rubber and plasticsCao su products và nhựa FabricatedFabricated metal metal products products TextilesDệt ChemicalsHóa chất and và chemical sản phẩm products hóa chất ElectricalThiết equipment bị điện FurnitureĐồ gỗ Phương tiện giaoOther thông vehicles khác Motor vehicles, trailers, semi-trailersXe có động cơ beveragesĐồ uống Paper and paper productsGiấy SảnOther phẩm manufacturing chế tạo khác Wood,Sản phẩm bamboo gỗ productstre (không (excl. bao gồmfurniture) đồ gỗ) MáyMachinery móc phụ and tùng equipment chưa phân n.e.c. loại Medical,Công precision cụ y tế, and chính optical xác instrumentsvà quang học PrintingXuất and bảnpublishing và in ấn Than Coke,Coke, các refined sản phẩm petroleum hóa dầu, products, nhiên liệunuclear hạt nhânfuel Repair and installation of machinerySửa chữa và and lắp equipment đặt thiết bị Tobacco productsThuốc lá

0 50000000 100000000100000000 150000000150000000 200000000200000000250000000250000000300000000300000000

20162016 20112011

Source: Authors’ calculation based on 2017 Enterprise Census data

Figure 2.17: (lower panel): Sub-sector share in MVA

Điện tử KimBas loạiic m ecơta lbảns Chế biến thực phẩm Sản phẩmNon-m khoángetallic m isảnneral phi produc kimt loạis May Leather, leather products and footDawe agiầyr Cao su và nhựa FabricatedChế me tbiếnal pro kimduct loạis Dệt Electrical equipment Thiết bị điện MotHóaor ve hchấticles và, tr asảnilers phẩm, semi- thóaraile chấtrs Xe có động cơ Phương tiện giao thôngFurnitu khácre Đồ gỗ Paper and paper proĐồduc uốngts Giấy MachineSảnry and phẩm equip chếmen tạot n.e khác.c. Máy móc phụ tùng chưa phân loại SảnMed iphẩmcal, prec gỗis itreon a(khôngnd optic baoal in sgồmtrum eđồnt sgỗ) Công cụ y tế, chính xác và quang học Repair and installation of machineryXuất and bảnequi pvàm einnt ấn Sửa chữa và lắp đặt thiết bị Than Coke, các sản phẩm hóa dầu, nhiênTobac liệuco p rhạtoduc nhânts Thuốc lá 0.00% 5.00% 10.00% 15.00% 20.00% 0.00% 2.00% 4.00% 6.00% 8.00% 10.00% 12.00% 14.00% 16.00% 18.00% 2016 2011 2016 2011 Source: Authors’ calculation based on 2017 Enterprise Census data

FDI’s share in total manufacturing VA in 2016 (similar to its employment and revenue shares) was the largest: 64.44 percent and higher than its employment share (47.88 percent) and revenue share (54.66 percent) in manufacturing employment and revenue, respectively. SOEs’ VA share (4.77 percent) was smallest (higher than their employment share (3.6 percent) and lower than their revenue share (6.07 percent), and domestic private firms’ VA share was 30.79 percent (lower than their employment share (40.88 percent) and revenue share (35.47 percent) in 2016 (Figure 2.18).

In addition to sub-sectors where FDI VA shares were high and corresponded to large FDI shares in manufacturing and employment, it is interesting to note that FDI had: (i) relatively high VA shares, while its shares in manufacturing employment and revenue were only at medium level in beverages and paper sub-sectors and (ii) of its medium VA shares, its employment share was low in tobacco, non-metallic metal products and basic metals, indicating higher productivity than SOEs, domestic private firms and SMEs in these sub-sectors. SOEs only had higher VA shares in tobacco, repair-installation of equipment and coke-refined petroleum-nuclear fuel and a medium VA share in beverages. Domestic firms had relatively high VA shares in food, wood and bamboo (excluding furniture), printing, medical and precision equipment, rubber-plastics and non-metallic mineral products, in contrast to low VA shares in tobacco,

46 Productivity and Competitiveness of Viet Nam’s Enterprises all sub-sectors (except coke-refined petroleum-nuclear fuel) experienced VA rises between 2011-2016, electronics, vehicles and other manufacturing and medium VA shares in the remainder (Figure 2.18). the higher 2016 VA shares (meaning relatively faster VA growth compared to other manufacturing sub- sectors) compared to 2011 shares were only observed in some sub-sectors (electronics, basic metal, Figure 2.19 shows large enterprises dominated almost all sub-sectors in terms of VA. Large firms’ share rubber-plastics, textiles, electrical equipment, motor vehicles and other manufacturing). of total manufacturing revenue was 81.69 percent and only three sub-sectors (printing, wood excluding furniture and repair-installation of machinery) had SME VA shares of more than 50 percent and SME VA Figure 2.17: (upper panel): Value added of manufacturing sub-sector, VND billion (2016 prices) share is close to 50 percent in rubber-plastics, and fabricated metal products.

Điện tử Chế biến thực phẩm Figure 2.18: FDI, SOE and domestic firms’ shares in sub-sector VA (%) 2016 Kim loại cơ bản May Sản phẩm khoáng sản phi kim loại NgànhTotal manufacturingchế biến chế tạo Da giầy Wood,Sản bamboo phẩm gỗ products tre (không (excl. bao furniture)gồm đồ gỗ) Cao su và nhựa Medical,Công precision cụ y tế, and chính optical xác instruments và quang học Fabricated metal products Chế biến thựcFood phẩm Dệt Printing Xuấtand publishingbản và in ấn Hóa chất và sản phẩm hóa chất Repair and installation of machinerySửa chữa vàand lắp equipment đặt thiết bị Thiết bị điện Rubber and plasticsCao suproducts và nhựa Đồ gỗ Fabricated Chếmetal biến products kim loại Phương tiện giao thông khác SảnNon-metallic phẩm khoáng mineral sản phi products kim loại Xe có động cơ FurnitureĐồ gỗ Đồ uống Paper and paper productsGiấy Giấy Wearing apparel, furMay Sản phẩm chế tạo khác Electrical equipmentThiết bị điện Sản phẩm gỗ tre (không bao gồm đồ gỗ) ChemicalsHóa chất and và chemical sản phẩm products hóa chất Máy móc phụ tùng chưa phân loại MachineryMáy móc phụ and tùng equipment chưa phân n.e.c. loại Công cụ y tế, chính xác và quang học TextilesDệt Xuất bản và in ấn beveragesĐồ uống Than Coke, các sản phẩm hóa dầu, nhiên liệu hạt nhân Motor vehicles, trailers, semi-trailersXe có động cơ Sửa chữa và lắp đặt thiết bị Leather, leather products and footwearDa giầy Thuốc lá SảnOther phẩm manufacturing chế tạo khác Than Coke, refinedcác sản petroleumphẩm hóa dầu,products, nhiên nuclear liệu hạt fuelnhân Basic metals 0 50000000 100000000 150000000 200000000 250000000 300000000 Kim loại cơ bản Phương tiện Othergiao thông vehicles khác Manufacture of computer, electronic and opticalĐiện tử 2016 2011 Tobacco productsThuốc lá 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Source: Authors’ calculation based on 2017 Enterprise Census data 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% FDIFDI SOEsDNNN DomesticDN tư nhân private trong firmsnước

Figure 2.17: (lower panel): Sub-sector share in MVA Source: Authors’ calculation based on 2017 Enterprise Census data

Điện tử Kim loại cơ bản Chế biến thực phẩm Figure 2.19: Large firms and SME shares of sub-sector VA (%), 2016 Sản phẩm khoáng sản phi kim loại May Da giầy NgànhTotal chế manufacturing biến chế tạo Cao su và nhựa Repair and installation ofSửa machinery chữa và and lắp đặtequipment thiết bị Chế biến kim loại PrintingXuất and bản publishing và in ấn Dệt Wood,Sản phẩm bamboo gỗ tre products (không (excl.bao gồm furniture) đồ gỗ) Thiết bị điện Cao su và nhựa Hóa chất và sản phẩm hóa chất Rubber and plastics products Xe có động cơ FabricatedChế metal biến products kim loại Phương tiện giao thông khác MáyMachinery móc phụ and tùng equipment chưa phân n.e.c. loại Đồ gỗ Paper and paper productsGiấy Đồ uống ElectricalThiết equipment bị điện Giấy ChemicalsHóa chất and và sảnchemical phẩm productshóa chất Sản phẩm chế tạo khác Đồ gỗ Máy móc phụ tùng chưa phân loại Furniture Sản phẩm gỗ tre (không bao gồm đồ gỗ) TextilesDệt Công cụ y tế, chính xác và quang học SảnOther phẩm manufacturing chế tạo khác Xuất bản và in ấn SảnNon-metallic phẩm khoáng mineral sản phi products kim loại Sửa chữa và lắp đặt thiết bị Chế biến thực phẩmFood Than Coke, các sản phẩm hóa dầu, nhiên liệu hạt nhân Công cụ y tế, chính xác và quang học Thuốc lá Medical, precision and optical instruments Wearing apparel,May fur 0.00% 2.00% 4.00% 6.00% 8.00% 10.00% 12.00% 14.00% 16.00% 18.00% beveragesĐồ uống Motor vehicles, trailers,Xe semi-trailers có động cơ 2016 2011 Phương tiện giaoOther thông vehicles khác Than Coke,Coke, các refined sản phẩm petroleum hóa dầu, products, nhiên liệu nuclear hạt nhân fuel Source: Authors’ calculation based on 2017 Enterprise Census data Leather, leather products and footwearDa giầy KimBasic loại metalscơ bản Manufacture of computer, electronic andĐiện optical tử FDI’s share in total manufacturing VA in 2016 (similar to its employment and revenue shares) was the Tobacco productsThuốc lá largest: 64.44 percent and higher than its employment share (47.88 percent) and revenue share (54.66 0% 10% 20% 30% 40% 50% 60% 70%70% 80%80% 90%90% 100%100% percent) in manufacturing employment and revenue, respectively. SOEs’ VA share (4.77 percent) was DoanhLarge Firms nghiệp lớn SMEsDoanh nghiệp VVN smallest (higher than their employment share (3.6 percent) and lower than their revenue share (6.07 Source: Authors’ calculation based on 2017 Enterprise Census data percent), and domestic private firms’ VA share was 30.79 percent (lower than their employment share (40.88 percent) and revenue share (35.47 percent) in 2016 (Figure 2.18). Exports and net exports

In addition to sub-sectors where FDI VA shares were high and corresponded to large FDI shares in All sub-sectors experienced increases in export volumes between 2005-2016. The largest rises manufacturing and employment, it is interesting to note that FDI had: (i) relatively high VA shares, while occurred in electronics, leather-footwear to make these sub-sectors the largest and third largest export its shares in manufacturing employment and revenue were only at medium level in beverages and paper volume sub-sectors in 2016. Wearing apparel’s export volume was second largest, followed by food and sub-sectors and (ii) of its medium VA shares, its employment share was low in tobacco, non-metallic beverages (fourth largest), textiles (fifth) and furniture (sixth) in 2016 (Figure 2.20). metal products and basic metals, indicating higher productivity than SOEs, domestic private firms and SMEs in these sub-sectors. SOEs only had higher VA shares in tobacco, repair-installation of equipment and coke-refined petroleum-nuclear fuel and a medium VA share in beverages. Domestic firms had relatively high VA shares in food, wood and bamboo (excluding furniture), printing, medical and precision equipment, rubber-plastics and non-metallic mineral products, in contrast to low VA shares in tobacco,

46 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 47 |Figure 2.20: Sub-sectors’ export volume (USD million, current price)

Office, accounting and computing machineryĐiện tử Wearing apparel,May fur Leather, leather products and footwearDa giầy Chế biến thực phầmFood and và beverages đồ uống TextilesDệt Furniture; manufacturingĐồ n.e.c. gỗ Electrical machinery andThiết apparatus bị điện Medical, precision andCông optical cụ y instruments, tế, chính xác watches và quang and clocks học HóaChemicals chất và andsản chemical phẩm hóa products chất Kim Basicloại cơ metals bản Rubber and Caoplastics su vàproducts nhựa MachinerySản phẩm and chế equipment tạo khác n. Fabricated metal products except machineryChế andbiến equipment kim loại Sản phẩm gỗ Woodtre (không products bao (excl. gồm furniture) đồ gỗ) Sản phẩmNon-metallic khoáng sản mineral phi productskim loại PhươngOther tiện transport giao thông equipment khác Motor vehicles, trailers,Xe semi-trailerscó động cơ Paper and paper productsGiấy TobaccoThuốc products lá Publishing, printing and reproductionXuất of recorded bản và media in ấn Than Coke, các sảnCoke,refined phẩm hóa petroleum dầu, nhiên products,nuclear liệu hạt nhân fuel

-- 5,000.00 10,000.0010,000.00 15,000.0015,000.00 20,000.00 20,000.00 25,000.00 25,000.00 30,000.00 30,000.00 35,000.00 35,000.00 40,000.00 40,000.00

2016 2011 20052005

Source: Authors’ calculation, UN Comtrade

Figure 2.21 shows net export (exports minus imports) values of sub-sectors and indicates that only few sub-sectors have positive net export values. This together with the level of FDI, SOE and domestic private firms’ participation (as indicated by the analysis on employment, revenue and VA shares) suggest that: (i) electronics (mainly smart phones), wearing apparel, leather-footwear were the largest export- oriented and FDI-led sub-sectors (noting that wearing apparel had a medium participation level from domestic private firms), (ii) furniture, food and beverages and wood (excluding furniture) were medium export-oriented and domestic private firms led sub-sectors (noting that furniture had high and food and beverages had medium participation levels of FDI). The other vehicles (mainly motorbikes) sub-sector was also a positive net exporter and led by FDI. The only sub-sector to be a positive net exporter led by SOEs was tobacco. The remaining sub-sectors (chemicals, machinery and equipment n.e.c., basic metals, textiles13, motor vehicles, fabricated metal, rubber-plastics, paper, electrical equipment) had high and medium levels of FDI participation with negative net export-import substitution sub-sectors.

Figure 2.21: Manufacturing sub-sector net exports (USD million, current prices)

Manufacture of computer, electronic andĐiện optical tử Wearing apparel,May fur Leather, leather products and footwearDa giầy Furniture; manufacturing Đồn.e.c. gỗ Chế biến thựcFood phẩm and và beverages đồ uống Sản phẩmWood gỗ tre products (không (excl.bao gồm furniture) đồ gỗ) Phương tiện giaoOther thông vehicles khác SảnNon-metallic phẩm khoáng mineral sản phi products kim loại Tobacco productsThuốc lá Publishing, printing and reproduction ofXuất recorded bản và media in ấn Medical, precision and opticalCông instruments, cụ y tế, chính watches xác và quangand clocks học Than Coke,Coke,refined các sản phẩm petroleum hóa dầu, products,nuclear nhiên liệu hạt nhân fuel ElectricalThiết equipment bị điện Paper and paper productsGiấy Rubber and plasticsCao su products và nhựa Fabricated metal products except machineryChế and biến equipment kim loại Motor vehicles, trailers,Xe semi-trailers có động cơ TextilesDệt KimBasic loại cơmetals bản MachinerySản and phẩm equipment chế tạo n.e.c. khác ChemicalsHóa chất and và sản chemical phẩm productshóa chất

(20,000.00) (20,000.00) (15,000.00)(15,000.00) (10,000.00)(10,000.00) (5,000.00)(5,000.00) -- 5,000.00 10,000.00 15,000.0015,000.00 20,000.0020,000.00 25,000.0025,000.00 30,000.00 30,000.00

2016 2011 2005 Source: Authors’ calculation, UN Comtrade

13 As trade agreements require more origin countries’ inputs in exported products, demand for made-in-Viet Nam textile products (inputs to wearing apparel – Viet Nam manufacturing’s second largest export sub-sector) increases. This explains: (i) the volume of textile exports was the third highest in 2016, but the sub-sector net export was negative, (ii) increased FDI inflows to Viet Nam in this sub-sector and suggests (trade negotiations and industrial policies may not have worked in tandem and resulted in the fact that) domestic firms may not have effectively captured the increased demand created by trade agreements and negatively affected the local content of value added in this sub-sector.

48 Productivity and Competitiveness of Viet Nam’s Enterprises |Figure 2.20: Sub-sectors’ export volume (USD million, current price) IR4.0 readiness of manufacturing enterprises at sub-sector level

Điện tử May The MOIT-VASS-UNDP study on IR4.0 readiness of Viet Nam’s industry firms found that manufacturing Da giầy Chế biến thực phầm và đồ uống sub-sectors were assessed at “outsider” level with readiness scores varying from 0.42 to 0.80. The Dệt proportion of sub-sectors’ firms at “outsider” level ranged from 73-92 percent, at “beginner” level from Đồ gỗ Thiết bị điện 4-25 percent and at “intermediate” level from 1-6 percent (Figure 2.22). Công cụ y tế, chính xác và quang học Hóa chất và sản phẩm hóa chất Kim loại cơ bản Figure 2.22: Proportion (%) of enterprises’ IR4.0 readiness levels by sub-sector Cao su và nhựa Sản phẩm chế tạo khác Chế biến kim loại NgoàiOutsider cuộc MớiBeginner bắt đầu IntermediateTrình độ cơ bản ExpriencedKinh nghiệm ExpertChuyên gia LeadingĐi đầu Sản phẩm gỗ tre (không bao gồm đồ gỗ) Sản phẩm khoáng sản phi kim loại Khai thácElectronics dầu khí Phương tiện giao thông khác Sản phẩmChemicals điện tử Điện, khí đốt, nước Xe có động cơ Others Hóa chất và sản phẩm hóa chất Giấy Các ngànhMotor SXCN vehicle khác Thuốc lá SX xe Machinerycó động cơ Xuất bản và in ấn Máy móc,Basic thiết metal bị Than Coke, các sản phẩm hóa dầu, nhiên liệu hạt nhân Sản xuấtBeverage kim loại TOÀN NGÀNHFood CÔNG processing THƯƠNG - 5,000.00 10,000.00 15,000.00 20,000.00 25,000.00 30,000.00 35,000.00 40,000.00 Sản xuất đồ uống ElectricalChế biến equipmentthực phẩm Other transportation vehicles 2016 2011 2005 Thiết bị điện Tàu,Paper thuyền, product xe lửa Giấy vàWearing sản phẩm apparel giấy Source: Authors’ calculation, UN Comtrade Fabricated netalMay Sản xuất sản phẩm cơ khí Textile Dệt Leather - footwearDa - Giầy Figure 2.21 shows net export (exports minus imports) values of sub-sectors and indicates that only RubberCao & su,plastics nhựa few sub-sectors have positive net export values. This together with the level of FDI, SOE and domestic Source: Survey on IR4.0 Readiness of Viet Nam’s Industry Firms (MOIT-VASS-UNDP 2018) private firms’ participation (as indicated by the analysis on employment, revenue and VA shares) suggest that: (i) electronics (mainly smart phones), wearing apparel, leather-footwear were the largest export- Sub-sectors with high export volumes (wearing apparel, leather-footwear) had the lowest IR4.0 readiness oriented and FDI-led sub-sectors (noting that wearing apparel had a medium participation level from scores and the highest proportions of firms at “outsider” level. Figure 2.22 shows that rubber-plastics, domestic private firms), (ii) furniture, food and beverages and wood (excluding furniture) were medium fabricated metals, textiles, wearing apparel, and leather-footwear sub-sectors had very high (above 90 export-oriented and domestic private firms led sub-sectors (noting that furniture had high and food and percent) proportions of firms at “outsider” level, compared to 75 percent in chemicals and electronics beverages had medium participation levels of FDI). The other vehicles (mainly motorbikes) sub-sector sub-sectors. was also a positive net exporter and led by FDI. The only sub-sector to be a positive net exporter led by SOEs was tobacco. The remaining sub-sectors (chemicals, machinery and equipment n.e.c., basic This IR4.0 study also found differences across manufacturing sub-sectors (Figure 2.23) in the rate of metals, textiles13, motor vehicles, fabricated metal, rubber-plastics, paper, electrical equipment) had cloud computing: (i) sub-sectors with higher cloud computing adoption rates included fabricated metal high and medium levels of FDI participation with negative net export-import substitution sub-sectors. (denoted in Figure 2.23 as “mechanics”) (26 percent), electrical equipment (23 percent), electronics (23 percent) and (ii) sub-sectors with lower rates of adoption included chemicals (8 percent), leather- Figure 2.21: Manufacturing sub-sector net exports (USD million, current prices) footwear (8 percent) and beverages (9 percent).

Điện tử May Figure 2.23: Firms’ usage of cloud computing by sub-sectors (%) Da giầy Đồ gỗ Chế biến thực phẩm và đồ uống MechanicsCơ khí 26.3 Sản phẩm gỗ tre (không bao gồm đồ gỗ) ElectricalThiết equipment bị điện 23.2 Phương tiện giao thông khác Sản phẩmElectronics điện tử 22.7 Sản phẩm khoáng sản phi kim loại Điện, khíMotor đốt, vehicles nước 18.6 Thuốc lá 18.6 Xuất bản và in ấn Tàu,Machines, thuyền, equipment xe lửa 18.6 Công cụ y tế, chính xác và quang học Sản xuất xe có độngMean cơ 17.1 Than Coke, các sản phẩm hóa dầu, nhiên liệu hạt nhân Máymóc,thiếtFood bị 14.8 Thiết bị điện Chế biếnRubber, thực plastics phẩm 14.0 Giấy Cao su,Median nhựa 13.9 Cao su và nhựa Chế biến kim loại Giấy, sản phẩmPaper giấy 13.2 Xe có động cơ TextileDệt 13.2 Dệt ApparelMay 12.9 Kim loại cơ bản Ngành Otherkhác 10.8 Sản phẩm chế tạo khác Sản xuấtMetal kimproducts loại 9.9 Hóa chất và sản phẩm hóa chất Sản xuất đồBeverage uống 8.9 (20,000.00) (15,000.00) (10,000.00) (5,000.00) - 5,000.00 10,000.00 15,000.00 20,000.00 25,000.00 30,000.00 FootwearDa giày 8.3 Hóa chất, sản phẩmChemical hóa product chất 7.8 2016 2011 2005 0 5 1 0 1 5 2 0 2 5 3 0

Source: Authors’ calculation, UN Comtrade Source: MOIT-VASS-UNDP (2018) FDI linkages with domestic firms

As a result of sharply rising inflows of FDI into Viet Nam with the major share (70 percent) into

13 As trade agreements require more origin countries’ inputs in exported products, demand for made-in-Viet Nam textile products manufacturing, FDI’s share in manufacturing employment, revenue and value addition have grown (as (inputs to wearing apparel – Viet Nam manufacturing’s second largest export sub-sector) increases. This explains: (i) the volume of discussed above). Despite rapid rises in exports and FDI’s contribution to Viet Nam’s GDP, exports and textile exports was the third highest in 2016, but the sub-sector net export was negative, (ii) increased FDI inflows to Viet Nam in this sub-sector and suggests (trade negotiations and industrial policies may not have worked in tandem and resulted in the fact that) domestic firms may not have effectively captured the increased demand created by trade agreements and negatively affected the local content of value added in this sub-sector.

48 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 49 job creation14, FDI’s forward and backward linkages15 to the rest of the economy remain modest, although rising as shown in Figures 2.24 and 2.2516. FDI linkages with domestic firms are particularly weak in high-tech manufacturing sub-sectors (such as electronics and motor vehicles), in which FDI firms focus mainly on assembling (imported) components and packaging final products for export (electronics) or local market (motor vehicles). In resource-based manufacturing sub-sectors (basic metals, chemicals and textiles) FDI tends to exhibit stronger linkages. Besides FDI using natural resources delivered by domestic firms, the longer presence of FDI in resource-based manufacturing sub-sectors (underpinning narrower gaps between domestic and FDI firms, or alternatively higher absorptive capacity of domestic firms) may also explain examples of stronger linkages between FDI and domestic firms.

Weak linkages between FDI and domestic firms indicate that Viet Nam’s integration into the global economy in general and global value chains (GVCs) in particular through the FDI channel is still shallow. Weak linkages also present the biggest obstacles to realization of FDI spillover effects through which technologies, labour skills and managerial experience are transferred to domestic firms as expected by Viet Nam’s policy-makers through FDI policies as is the case in other nations. This result is consistent with previous studies on spillover effects of foreign capital in Viet Nam, such as by Chuc et al. (2008), CIEM (2012), Porter (2010) and Tran Van Tho (2005).

Figure 2.24: Backward-linkages between FDI and domestic firms in Viet Nam’s manufacturing industry (%, max=100%)

Repair and installationSửa of chữa machinery và lắp and đặt equipment thiết bị Các sản phẩm chế biếnOther chếmanufacturing tạo khác FurnitureĐồ gỗ Phương tiện giaoOther thông vehicles khác Motor vehicles, trailers,Xe cósemi-trailers động cơ Máy mócMachinery phụ tùng and chưa equipment phân n.e.c. loại Máy mócElectrical thiết equipment bị điện Manufacture of computer, electronic andĐiện optical tử FabricatedChế biếnmetal kimproducts loại Kim loạiBasic cơ metals bản Sản phẩmNon-metallic khoáng sản mineral phi kim products loại Rubber andCao plastics su và products nhựa Sản Medical,phẩm y precision tế, quang and học optical và chínhinstruments xác HóaChemicals chất và sảnand chemicalphẩm hóa products chất Coke, sản Coke,phẩm refined hóa dầu petroleum và nguyên products, liệu nuclear hạt nhân fuel PrintingXuất andbản publishing và in ấn Paper and paper productsGiấy SảnWood, phẩm bamboo gỗ (không products bao (excl. gồm furniture) đồ gỗ) Leather, leather products and footwearDa giày Wearing apparel,May fur TextilesDệt TobaccoThuốc products lá beveragesĐồ uống Chế biến thực phẩmFood 0.00 10.00 20.00 30.00 40.00 50.00 60.00 70.00 80.00 90.00 100.00 2007-2010 2011-2015

Source: Authors’ calculation, 2017 Enterprise Census

14 FDI has made increasing contributions of around 20 percent of Viet Nam’s GDP (from 15.2 percent in 2005), 72 percent of the country’s exports (from 57 percent in 2005), 18 percent of government revenue and creating 3.7 million jobs for Vietnamese workers in 2017. 15 The idea of linkages grew out of Hirschman's theory of unbalanced growth and describes the relationships that exist between parties involved along the supply chain. Backward linkages describe the process of how a company in a given sector purchases its goods, products, or supplies from a company in a different sector; these are called inputs. Forward linkages describe the process of how a company in a given sector sells its goods, products, or supplies to a company in a different sector; these are called outputs. 16 The backward and forward linkage indexes are calculated to 2007-2010 and 2011-2015. They receive the value from 0-100 percent: 0 percent means “no linkage” and the value of the index approaching 100 percent means a stronger linkage. Details on how backward and forward linkages of the FDI sector with the rest of the economy are calculated are given in Section A.1.4 of Appendix 1.

50 Productivity and Competitiveness of Viet Nam’s Enterprises job creation14, FDI’s forward and backward linkages15 to the rest of the economy remain modest, although Figure 2.25: Forward-linkages between FDI and domestic firms in Viet Nam’s manufacturing rising as shown in Figures 2.24 and 2.2516. FDI linkages with domestic firms are particularly weak in industry (%, max = 100%) high-tech manufacturing sub-sectors (such as electronics and motor vehicles), in which FDI firms focus Repair and installationSửa of chữa machinery và lắp and đặt equipment thiết bị Các sản phẩm chế biếnOther chế manufacturing tạo khác mainly on assembling (imported) components and packaging final products for export (electronics) or FurnitureĐồ gỗ Phương tiện giao Otherthông vehicles khác local market (motor vehicles). In resource-based manufacturing sub-sectors (basic metals, chemicals Motor vehicles, trailers,Xe có semi-trailers động cơ Máy mócMachinery phụ tùng and chưa equipment phân n.e.c. loại and textiles) FDI tends to exhibit stronger linkages. Besides FDI using natural resources delivered by Máy mócElectrical thiết equipment bị điện Manufacture of computer, electronic andĐiện optical tử FabricatedChế biến metal kim products loại domestic firms, the longer presence of FDI in resource-based manufacturing sub-sectors (underpinning Kim loạiBasic cơ metals bản Sản phẩmNon-metallic khoáng sản mineral phi kim products loại narrower gaps between domestic and FDI firms, or alternatively higher absorptive capacity of domestic Rubber andCao plastics su và products nhựa Sản Medical,phẩm y precision tế, quang and học optical và chínhinstruments xác firms) may also explain examples of stronger linkages between FDI and domestic firms. HóaChemicals chất và sảnand chemicalphẩm hóa products chất Coke, sản phẩmCoke, refined hóa dầu petroleum và nguyên products, liệu nuclear hạt nhân fuel PrintingXuất andbản publishing và in ấn Paper and paper productsGiấy Weak linkages between FDI and domestic firms indicate that Viet Nam’s integration into the global SảnWood, phẩm bamboo gỗ (không products bao (excl. gồm furniture) đồ gỗ) Leather, leather products and Dafootwear giày economy in general and global value chains (GVCs) in particular through the FDI channel is still shallow. Wearing apparel,May fur TextilesDệt TobaccoThuốc products lá Weak linkages also present the biggest obstacles to realization of FDI spillover effects through which Đồbeverages uống technologies, labour skills and managerial experience are transferred to domestic firms as expected by Chế biến thực phẩmFood Viet Nam’s policy-makers through FDI policies as is the case in other nations. This result is consistent 0.00 10.00 20.00 30.00 40.00 50.00 60.00 70.00 80.00 90.00 100.00 with previous studies on spillover effects of foreign capital in Viet Nam, such as by Chuc et al. (2008), 2007-2010 2011-2015 CIEM (2012), Porter (2010) and Tran Van Tho (2005). Source: Authors’ calculation, 2017 Enterprise Census

Figure 2.24: Backward-linkages between FDI and domestic firms in Viet Nam’s manufacturing Summary industry (%, max=100%) Key characteristics of manufacturing sub-sectors are summarized in Table 2.3 (besides low IR4.0 Sửa chữa và lắp đặt thiết bị Các sản phẩm chế biến chế tạo khác readiness and FDI’s backward-forward linkages across sub-sectors with degrees of variation as mentioned Đồ gỗ Phương tiện giao thông khác above). To more easily digest characteristics that feed the analyses of productivity and competitiveness Xe có động cơ Máy móc phụ tùng chưa phân loại performance in the following section, the sub-sectors’ importance (in terms of employment, revenue Máy móc thiết bị điện Điện tử Chế biến kim loại and exports) and the FDI, SOE, domestic private and large firms’ and SMEs’ participation levels in sub- Kim loại cơ bản Sản phẩm khoáng sản phi kim loại sectors are divided into the following categories (with thresholds defined based on 2017 Enterprise Cao su và nhựa Sản phẩm y tế, quang học và chính xác Census data) to present the later analysis on performance of enterprises more systematically: Hóa chất và sản phẩm hóa chất Coke, sản phẩm hóa dầu và nguyên liệu hạt nhân Xuất bản và in ấn Giấy Sub-sector sizes (measured by employment, revenue and VA shares in manufacturing totals) are divided Sản phẩm gỗ (không bao gồm đồ gỗ) Da giày into three groups: (i) large (sub-sector shares in total manufacturing employment, revenue and VA are May Dệt above 4 percent), (ii) medium (sub-sector share in total manufacturing employment, revenue and VA are Thuốc lá Đồ uống Chế biến thực phẩm between 2-4 percent) and (iii) small (sub-sector shares in total manufacturing employment, revenue and 0.00 10.00 20.00 30.00 40.00 50.00 60.00 70.00 80.00 90.00 100.00 VA are less than 2 percent) 2007-2010 2011-2015 Participation level of FDI, SOEs and domestic private firms in sub-sectors (measured by employment, Source: Authors’ calculation, 2017 Enterprise Census revenue and VA shares in manufacturing sub-sector totals) are divided into three groups: (i) high (FDI, SOE and domestic private firm shares in sub-sector employment, revenue and VA are above 45 percent), (ii) medium (FDI, SOE and domestic private firm shares in sub-sector employment, revenue and VA are 20-45 percent) and (iii) low (FDI, SOE and domestic private firm shares in sub-sector employment, revenue and VA are less than 20 percent)

Sub-sector’s importance in terms of exports (measured by net export value) is divided into three groups: (i) large positive net export (net export > USD5 billion/year), (ii) medium positive export (0

50 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 51 Table 2.3: Summary of key sub-sector characteristics

Source: Authors’ calculation, 2017 Enterprise Census

Productivity and competitiveness performance at sub-sector level

In this section, productivity and competitiveness of sub-sectors at (VSIC) 2-digit level will be assessed using different measurements, such as LP, RCA, domestic contents, value addition (VA/output and profit/ revenue).

Labour productivity performance at sub-sector level

International comparison:

With a certain degree of similarity to the overall trend of Viet Nam’s and comparator countries’ manufacturing labour productivity (Figure 2.4), Table 2.4 (using UNIDO database, ISIC Rev.317) shows, on the other hand, interesting nuances and variations at sub-sector level, namely:

• Viet Nam’s labour productivity as a share of labour productivity of Japan and RoK remained very low in all sub-sectors: just above 15 percent in textiles, coke-refined petroleum products-nuclear fuel and basic metals, while in the remaining sub-sectors it was around 10 percent. However, almost all of Viet Nam’s sub-sectors have narrowed labour productivity gaps with Japan and RoK at high speeds. Viet Nam’s LP as a share of Japan’s and RoK’s LP increased by more than 70 percent between 2005-

17 Some inconsistencies were observed by the authors while examining the UNIDO database, especially data related to: (i) some sub-sectors such as tobacco and coke-petroleum products-nuclear fuel and (ii) Indonesia. Careful treatment of the analysis using such data is advised.

52 Productivity and Competitiveness of Viet Nam’s Enterprises Table 2.3: Summary of key sub-sector characteristics 2015. Viet Nam’s LP increased with less high speed in tobacco products compared to Japan and RoK and motor vehicles-trailers-semitrailers compared to Japan.

• In almost all sub-sectors, Viet Nam’s LP remained lower than China’s and Indonesia’s and the LP gaps widened. Few sub-sectors narrowed LP gaps (at low speed below 50 percent between 2005-2015) compared with: (a) China (wood excluding furniture, machinery and equipment n.e.c and electronics) and (b) Indonesia (wood products excluding furniture and furniture), the LP gap narrowed by more than 70 percent and Viet Nam’s LP is “within reach” of Indonesia’s. Regarding electronics, Viet Nam’s LP was 85.4 percent of Indonesia’s and the gap narrowed, while the gap closed at a lower speed with respect to paper and paper products and non-metallic mineral products.

• During 2005-2015, Viet Nam closed LP gaps with (and within reach to LP levels of) Malaysia in almost all sub-sectors, except tobacco, wearing apparel and fabricated metal where LP gaps widened. Non-metallic mineral products, electrical equipment and motor vehicle sub-sectors’ LP gaps with Malaysia narrowed at slow speeds. Viet Nam’s LP exceeded and was “within reach” of Philippines’ LP in numerous sub-sectors, while LP gaps widened only in tobacco, non-metallic mineral products and motor vehicles. LP gaps in wearing apparel, wood (excluding furniture), furniture and other manufacturing n.e.c. and chemicals narrowed slowly.

• Viet Nam’s LP performance, compared to Bangladesh and India, is remarkable. Viet Nam outperformed the pair in many manufacturing sub-sectors. However, Viet Nam’s LP was lower and slowly catching India’s (and the Philippines, yet widened with China, Indonesia and Malaysia) in apparel, which plays an important role in Viet Nam’s exports and job creation.

Together with the information on sub-sector characteristics, Table 2.4 shows:

Relative to comparator countries, the best performing sub-sectors (i.e. Viet Nam’s sub-sector LP either already higher than those of some comparator countries or the LP gap between Viet Nam and comparator countries had reduced fast18) included: food (big sub-sector dominated by large and domestic private firms, and with robust positive net exports), beverages (a medium sub-sector, FDI and domestic private have medium participation levels, while FDI has high VA), textiles (large sub-sector, FDI dominated with Source: Authors’ calculation, 2017 Enterprise Census negative net exports), printing (small, domestic private sector-led, with negative net exports), chemicals (medium-sized sub-sector, FDI dominated with negative net exports), coke-refined petroleum products- 19 Productivity and competitiveness performance at sub-sector level nuclear fuel (small-sized, SOE-led, negative net exports), electronics (large, FDI dominated sub-sector with large positive net exports), basic metal (large sub-sector with FDI dominating VA share and negative In this section, productivity and competitiveness of sub-sectors at (VSIC) 2-digit level will be assessed net exports) and other transport equipment (medium-sized sub-sector, led by FDI and medium positive using different measurements, such as LP, RCA, domestic contents, value addition (VA/output and profit/ net exports). revenue).

Labour productivity performance at sub-sector level

International comparison:

18 With the exception of comparisons with China and Indonesia, the LP gaps between which and Viet Nam widened in 2005-2015. 17 Some inconsistencies were observed by the authors while examining the UNIDO database, especially data related to: (i) some 19 Which includes: medical precision-optical equipment, office accounting and computing machinery, radio-television and sub-sectors such as tobacco and coke-petroleum products-nuclear fuel and (ii) Indonesia. Careful treatment of the analysis using communication products. Since 2011, the UNIDO database combined the last two subsections into one. For ease of comparison, in such data is advised. this section all three subsections are included in one sub-sector “electronics”.

52 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 53 Table 2.4: Labour productivity of Viet Nam relative to comparator countries (%)

Bangladesh China India Indonesia Malaysia Philippines ROK Japan ISOC Rev.3 Code Subsector Name 2005 2011 2005 2015 2005 2015 2005 2015 2005 2015 2005 2015 2005 2015 2005 2015 15 Food and beverages 94.3 138.5 30.8 39.2 97.2 170.9 46.6 44.5 24.2 49.6 28.6 51.8 4.2 13.5 5.1 18.8 16 Tobacco products --- 683.9 14.1 5.2 669.6 434.4 119.0 76.6 108.6 10.9 45.4 8.2 2.8 2.5 1.4 1.7 17 Textiles 222.8 216.5 53.3 27.0 86.9 155.1 74.2 67.6 39.2 51.2 60.5 167.4 6.1 19.3 6.1 17.9 18 Wearing apparel, fur --- 60.3 29.5 11.2 63.1 64.2 58.4 27.5 27.8 23.9 53.5 52.5 2.7 4.1 4.6 8.1 19 Leather and footwear 8.4 47.0 22.3 14.9 33.7 61.9 26.9 26.5 14.1 21.0 21.7 42.2 1.9 4.6 1.7 6.0 20 Wood products (excl. furniture) 90.0 137.8 19.0 29.7 36.3 103.6 25.1 84.0 16.4 54.7 37.5 52.7 2.2 13.2 2.0 11.1 21 Paper and paper products 121.1 146.9 27.5 13.7 43.9 89.2 14.7 20.8 23.7 43.1 36.8 70.9 3.4 9.8 2.6 9.1 22 Publishing and printing 187.9 192.2 59.8 30.6 57.5 66.5 48.3 42.6 35.7 45.6 82.7 100.0 6.6 20.7 5.4 9.3 23 Coke, refined petroleum products, nuclear fuel 8.7 2024.3 65.4 82.9 16.8 114.3 156.4 553.7 2.4 43.0 1.9 83.5 1.9 17.9 7.6 96.2 24 Chemicals and chemical products 65.5 182.4 41.0 37.9 40.6 79.1 30.2 22.7 9.9 31.4 33.0 38.4 3.1 7.3 2.0 6.8 25 Rubber and plastics products 36.6 135.2 36.3 28.5 51.4 63.7 47.9 37.9 29.9 44.1 44.6 79.1 4.4 9.7 3.2 8.5 26 Non-metallic mineral products 67.0 260.2 59.1 39.1 76.0 114.2 40.0 45.1 28.0 38.0 26.0 19.6 3.9 8.5 3.9 10.2 27 Basic metals 1.1 97.2 32.9 27.2 38.5 112.1 27.1 25.7 36.0 98.1 43.7 106.0 3.2 16.3 2.6 17.0 28 Fabricated metal products 107.9 91.2 52.3 21.9 76.7 86.2 64.0 34.4 39.0 29.7 71.1 77.3 7.3 12.5 5.2 9.7 29 Machinery and equipment n.e.c. 105.4 156.8 21.2 23.3 23.5 57.4 25.0 26.4 12.5 38.9 24.5 42.2 2.7 10.6 1.9 8.2 30-32 Electronics 308.3 80.4 50.4 60.3 58.4 115.3 77.5 88.9 43.0 55.8 77.3 109.2 5.2 9.7 6.0 17.9 31 Electrical machinery and apparatus 90.0 59.0 45.7 41.4 44.5 78.7 42.5 15.6 42.4 41.9 47.7 56.8 6.8 11.7 5.0 9.3 33 Medical, precision and optical instruments 70.1 --- 34.2 --- 31.9 --- 77.9 --- 22.1 --- 17.4 --- 4.8 --- 2.8 --- 34 Motor vehicles, trailers, semi-trailers 2.4 45.8 62.5 43.8 61.7 100.6 17.2 12.6 55.2 42.9 62.7 34.5 8.0 10.9 6.3 10.3 35 Other transport equipment 25.1 188.7 59.9 42.7 44.0 116.2 21.8 37.2 34.4 67.4 33.4 89.2 5.3 17.2 4.4 15.1 36 Furniture and manufacturing n.e.c. 4.6 71.6 31.4 27.3 31.1 46.1 57.4 81.2 22.6 35.5 42.7 57.6 3.1 10.3 2.2 6.1 Total Manufacturing 103.6 202.6 27.5 22.1 34.9 63.5 33.5 30.5 17.2 27.3 26.6 36.4 2.9 7.2 2.6 7.8 Note: Electronics in this table includes: Office, accounting and computing machinery; Radio,television and communication equipment. Since 2011 UNIDO database does not provide data on Medical, precision and optical instruments, watches and clocks. Color coding: Good performance: Viet Nam's LP exceeds (red numbers) or "within reach" i.e. Viet Nam's LP is of 70% or more of comparator countries's LP and LP gaps reduced between 2005 and 2016 Promissing performance: Viet Nam's LP is below 70% of comparator countries' LP in 2016 but the LP gaps have been reduced faster than 50% between 2005 and 2016 "Needing improvement" performance: Viet Nam's LP is below 70% of comparator countries' LP in 2016 but the LP gaps have been reduced less than 50% between 2005 and 2016 Poor performance: Viet Nam's LP is below 70% of comparator countries' LP in 2016 but the LP gaps with comparator countries' LPs have been widened Source: Authors’ calculation from UNIDO data

The sub-sectors with promising performances (LP gap between Viet Nam and comparator countries was reduced – see the above footnote 18 on comparisons with China and Indonesia) included leather- footwear (large sector, dominated by FDI with large positive net exports), wood (excluding furniture) and paper (two small sub-sectors led by domestic private firms, wood has medium positive net exports and paper has negative net exports). Rubber-plastics (large sub-sector) and other manufacturing (small sub-sector) are two sub-sectors led by FDI with negative net exports, non-metallic mineral products (large sub-sector) and furniture (medium-sized sub-sector), both led by domestic private firms and with medium and large positive net exports.

Less well-performing sub-sectors (LP gap between Viet Nam and comparator countries had widened or reduced at slow speed) included: (i) tobacco products (small-sized, dominated by SOEs with medium positive net exports), (ii) wearing apparel (LP gaps with China and Indonesia widened and closed slowly with Bangladesh, India, Malaysia, the Philippines and Thailand, large sub-sector, FDI-led with large positive net exports), (iii) fabricated metals (large sub-sector with negative net exports led by FDI), electrical equipment and motor vehicles (medium-sized sub-sectors with negative net exports, dominated by FDI in which LP gaps with some countries widened and/or slowly narrowed).

Sub-sector LP analysis using 2017 Enterprise Census data20:

The GSO’s 2017 Enterprise Census data shows that all manufacturing sub-sectors experienced increased labour productivity during 2011-2016, except apparel, printing and publishing which experienced LP reductions by 12 and 7 percent, respectively (Figure 2.26). Sub-sectors with the highest LP increases included basic metals (by more than five-fold), non-metallic mineral products, rubber-plastics products, electrical equipment, motor vehicles-trailers-semitrailers (by more than 100 percent). Fabricated metal products, computer-electronic and optical and paper products also experienced significant LP rises by 89, 75 and 68

20 Noting that 2011 data for coke-refined petroleum products-nuclear fuel was excluded as the value in 2011 was unexplainably inconsistent. Importantly, as in the case of VA, due to the estimation methods and data issues, it is better to use the LP estimated in the report for analysis on structure and trends rather than in absolute values.

54 Productivity and Competitiveness of Viet Nam’s Enterprises Table 2.4: Labour productivity of Viet Nam relative to comparator countries (%) percent, respectively. Sub-sectors with low LP increases included furniture (3 percent), chemicals (6 percent), wood products excluding furniture (12 percent), leather-footwear (19 percent) and beverages (20 percent). Bangladesh China India Indonesia Malaysia Philippines ROK Japan ISOC Rev.3 Code Subsector Name 2005 2011 2005 2015 2005 2015 2005 2015 2005 2015 2005 2015 2005 2015 2005 2015 Figure 2.27a shows changes in sub-sectors’ LP relative to manufacturing LP, underlining the basic 15 Food and beverages 94.3 138.5 30.8 39.2 97.2 170.9 46.6 44.5 24.2 49.6 28.6 51.8 4.2 13.5 5.1 18.8 16 Tobacco products --- 683.9 14.1 5.2 669.6 434.4 119.0 76.6 108.6 10.9 45.4 8.2 2.8 2.5 1.4 1.7 metals sub-sector’s exceptional productivity performance. Its labour productivity relative to average 17 Textiles 222.8 216.5 53.3 27.0 86.9 155.1 74.2 67.6 39.2 51.2 60.5 167.4 6.1 19.3 6.1 17.9 manufacturing LP more than tripled, rising 2.52-fold in 2011 to 8.18-fold in 2016. It, thus, overtook 18 Wearing apparel, fur --- 60.3 29.5 11.2 63.1 64.2 58.4 27.5 27.8 23.9 53.5 52.5 2.7 4.1 4.6 8.1 beverages to become the most productive manufacturing sub-sector in 2016 (Figure 2.26) and as shown 19 Leather and footwear 8.4 47.0 22.3 14.9 33.7 61.9 26.9 26.5 14.1 21.0 21.7 42.2 1.9 4.6 1.7 6.0 20 Wood products (excl. furniture) 90.0 137.8 19.0 29.7 36.3 103.6 25.1 84.0 16.4 54.7 37.5 52.7 2.2 13.2 2.0 11.1 in Table 2.4, the LP of Viet Nam’s basic metals is rapidly catching comparator countries. 21 Paper and paper products 121.1 146.9 27.5 13.7 43.9 89.2 14.7 20.8 23.7 43.1 36.8 70.9 3.4 9.8 2.6 9.1 22 Publishing and printing 187.9 192.2 59.8 30.6 57.5 66.5 48.3 42.6 35.7 45.6 82.7 100.0 6.6 20.7 5.4 9.3 Figure 2.26: Labour productivity of manufacturing sub-sectors (VND million, 2016 prices) 23 Coke, refined petroleum products, nuclear fuel 8.7 2024.3 65.4 82.9 16.8 114.3 156.4 553.7 2.4 43.0 1.9 83.5 1.9 17.9 7.6 96.2 Coke,Coke, sản refined phẩm hóa petroleum dầu và nguyênproducts, liệu nuclear hạt nhân fuel 24 Chemicals and chemical products 65.5 182.4 41.0 37.9 40.6 79.1 30.2 22.7 9.9 31.4 33.0 38.4 3.1 7.3 2.0 6.8 Ngành chếManufacturing biến chế tạo Kim Basicloại cơ metals bản 25 Rubber and plastics products 36.6 135.2 36.3 28.5 51.4 63.7 47.9 37.9 29.9 44.1 44.6 79.1 4.4 9.7 3.2 8.5 Sản phẩmNon-metallic khoáng mineralsản phi kimproducts loại Rubber and plasticsCao su và products nhựa 26 Non-metallic mineral products 67.0 260.2 59.1 39.1 76.0 114.2 40.0 45.1 28.0 38.0 26.0 19.6 3.9 8.5 3.9 10.2 MáyElectrical móc thiết equipment bị điện 27 Basic metals 1.1 97.2 32.9 27.2 38.5 112.1 27.1 25.7 36.0 98.1 43.7 106.0 3.2 16.3 2.6 17.0 Motor vehicles, trailers,Xe semi-trailers có động cơ FabricatedChế metal biến kimproducts loại 28 Fabricated metal products 107.9 91.2 52.3 21.9 76.7 86.2 64.0 34.4 39.0 29.7 71.1 77.3 7.3 12.5 5.2 9.7 Manufacture of computer, electronic andĐiện optical tử Paper and paper productsGiấy 29 Machinery and equipment n.e.c. 105.4 156.8 21.2 23.3 23.5 57.4 25.0 26.4 12.5 38.9 24.5 42.2 2.7 10.6 1.9 8.2 TextilesDệt Các sản phẩm chếOther biến manufacturing chế tạo khác 30-32 Electronics 308.3 80.4 50.4 60.3 58.4 115.3 77.5 88.9 43.0 55.8 77.3 109.2 5.2 9.7 6.0 17.9 Chế biến thực phẩmFood MáyMachinery móc phụ tùngand equipment chưa phân n.e.c.loại 31 Electrical machinery and apparatus 90.0 59.0 45.7 41.4 44.5 78.7 42.5 15.6 42.4 41.9 47.7 56.8 6.8 11.7 5.0 9.3 Phương tiện giaoOther thông vehicles khác 33 Medical, precision and optical instruments 70.1 --- 34.2 --- 31.9 --- 77.9 --- 22.1 --- 17.4 --- 4.8 --- 2.8 --- Medical,Sản phẩm precision y tế, quang and optical học và instruments chính xác beveragesĐồ uống 34 Motor vehicles, trailers, semi-trailers 2.4 45.8 62.5 43.8 61.7 100.6 17.2 12.6 55.2 42.9 62.7 34.5 8.0 10.9 6.3 10.3 Leather, leather products and footwearDa giày Wood,Sản bamboo phẩm gỗ products (không bao (excl. gồm furniture) đồ gỗ) 35 Other transport equipment 25.1 188.7 59.9 42.7 44.0 116.2 21.8 37.2 34.4 67.4 33.4 89.2 5.3 17.2 4.4 15.1 TobaccoThuốc products lá ChemicalsHóa chất vàand sản chemical phẩm hóa products chất 36 Furniture and manufacturing n.e.c. 4.6 71.6 31.4 27.3 31.1 46.1 57.4 81.2 22.6 35.5 42.7 57.6 3.1 10.3 2.2 6.1 Repair and installation ofSửa machinery chữa và lắp and đặt equipment thiết bị Total Manufacturing 103.6 202.6 27.5 22.1 34.9 63.5 33.5 30.5 17.2 27.3 26.6 36.4 2.9 7.2 2.6 7.8 FurnitureĐồ gỗ PrintingXuất and bản publishing và in ấn Note: Electronics in this table includes: Office, accounting and computing machinery; Radio,television and communication equipment. Since 2011 UNIDO database does not provide data on Medical, precision and optical instruments, watches and clocks. Wearing apparel,May fur Color coding: 0 500500 10001000 15001500 20002000 25002500 Good performance: Viet Nam's LP exceeds (red numbers) or "within reach" i.e. Viet Nam's LP is of 70% or more of comparator countries's LP and LP gaps reduced between 2005 and 2016 Labor productivity (LP) in millionNSLĐ, VND (2016,triệu đồng in 2016 (2016, price) giá 2016) LaborNSLĐ, productivity triệu đồng (LP) (2011, in million giá 2016) VND (2011, in 2016 price) Promissing performance: Viet Nam's LP is below 70% of comparator countries' LP in 2016 but the LP gaps have been reduced faster than 50% between 2005 and 2016 "Needing improvement" performance: Viet Nam's LP is below 70% of comparator countries' LP in 2016 but the LP gaps have been reduced less than 50% between 2005 and 2016 Source: Authors’ calculation, 2017 Enterprise Census Poor performance: Viet Nam's LP is below 70% of comparator countries' LP in 2016 but the LP gaps with comparator countries' LPs have been widened Source: Authors’ calculation from UNIDO data Other fast-growing LP sub-sectors performing well relative to comparator countries included: electronics and other transport equipment. Electrical equipment and motor vehicles experienced high LP growth The sub-sectors with promising performances (LP gap between Viet Nam and comparator countries and remain in the group of sub-sectors that have been “slowly catching” comparator countries as sub- was reduced – see the above footnote 18 on comparisons with China and Indonesia) included leather- sector LP growth in these countries was even higher. footwear (large sector, dominated by FDI with large positive net exports), wood (excluding furniture) and paper (two small sub-sectors led by domestic private firms, wood has medium positive net exports While some sub-sectors performed relatively well compared to comparator countries, their LP growth and paper has negative net exports). Rubber-plastics (large sub-sector) and other manufacturing (small was consistently below Viet Nam’s manufacturing average in 2005 and 2016 (wood excluding furniture, sub-sector) are two sub-sectors led by FDI with negative net exports, non-metallic mineral products machinery and equipment n.e.c., furniture) or just slightly better than Viet Nam’s manufacturing average (large sub-sector) and furniture (medium-sized sub-sector), both led by domestic private firms and with (textiles, paper, food and beverages). Those underperforming (in terms of LP level and growth) were medium and large positive net exports. wearing apparel, leather and printing despite performing slightly better than Viet Nam’s manufacturing average, while motor vehicles, electrical equipment and fabricated metals were in the group of less well- Less well-performing sub-sectors (LP gap between Viet Nam and comparator countries had widened performing relative to comparator countries (Table 2.4). It should be noted that wearing apparel and leather or reduced at slow speed) included: (i) tobacco products (small-sized, dominated by SOEs with medium are important manufacturing sub-sectors in terms of employment, revenue and exports in Viet Nam. positive net exports), (ii) wearing apparel (LP gaps with China and Indonesia widened and closed slowly with Bangladesh, India, Malaysia, the Philippines and Thailand, large sub-sector, FDI-led with Figure 2.27a: Relative labour productivity (manufacturing = 1) large positive net exports), (iii) fabricated metals (large sub-sector with negative net exports led by BeverageĐồ uống Hóa chất và Chemicalsản phẩm products hóa chất MetalKim loại products cơ bản FDI), electrical equipment and motor vehicles (medium-sized sub-sectors with negative net exports, TobaccoThuốc lá PhươngOther tiện means giao of thông transport khác dominated by FDI in which LP gaps with some countries widened and/or slowly narrowed). Electronics, PC, optical productsĐiện tử Sản phẩm y Pharmaceuticaltế, quang học và production chính xác MotorXe có vehicles động cơ 20 Sản phẩm khoángNon-metal sản phi products kim loại Sub-sector LP analysis using 2017 Enterprise Census data : UnclassifiedMáy móc phụ machines, tùng chưa equipment phân loại Sửa chữaRepair, và lắp maintenance đặt thiết bị Chế biến thực phẩmFood The GSO’s 2017 Enterprise Census data shows that all manufacturing sub-sectors experienced increased MáyElectrical móc thiết equipment bị điện PrefabricatedChế biến kim metal loại PaperGiấy labour productivity during 2011-2016, except apparel, printing and publishing which experienced LP TextileDệt CaoRubber, su và plastic nhựa reductions by 12 and 7 percent, respectively (Figure 2.26). Sub-sectors with the highest LP increases included Xuất bảnPrinting và in ấn Sản phẩm gỗ (không bao gồm đồWood gỗ) basic metals (by more than five-fold), non-metallic mineral products, rubber-plastics products, electrical OtherCác sản processing phẩm chế and biến manufacturing chế tạo khác Beds, cabinets, tables, chairsĐồ gỗ ApparelMay equipment, motor vehicles-trailers-semitrailers (by more than 100 percent). Fabricated metal products, LeatherDa giày computer-electronic and optical and paper products also experienced significant LP rises by 89, 75 and 68 0% 100%100% 200% 300% 400% 500% 600% 700% 800% 900%

20 Noting that 2011 data for coke-refined petroleum products-nuclear fuel was excluded as the value in 2011 was unexplainably 2016 2011 NSLĐ tương đối (2016) NSLĐ tương đối (2011) inconsistent. Importantly, as in the case of VA, due to the estimation methods and data issues, it is better to use the LP estimated in the report for analysis on structure and trends rather than in absolute values. Source: Authors’ calculation, 2017 Enterprise Census

54 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 55 Box 2.4: Relative labour productivity using UNIDO 2015 data

Figure 2.27b shows sub-sectors’ relative labour productivity using UNIDO database for 2015 (the latest year for which Viet Nam’s data is available on the UNIDO database). While the trend in ranking sub-sectors’ relative LP is rather similar between Figures 2.17a and 2.17b, a few differences include: (i) in sub-sector classification: UNIDO database combined food and beverages into one sub-sector “food and beverages” as well as “office accounting and computers” and “medical, precision and optical equipment” into the “electronics” sub-sector and did not have “other manufacturing” and “repair and installation of machinery” compared to the VSIC sub-sector list used in this report, (ii) different years of data (2016 of Enterprise Census vs. 2015 of UNIDO database), and (iii) different sub-sector rankings in relative LP: if taking into account the differences in sub-sector classification, textiles ranked four places higher using UNIDO database than using Enterprise Census data, rubber-plastics - five places lower than using Enterprise Census data and non-metallic mineral products - six places lower than using Enterprise Census data. While differences in ranking “food” and “beverages” using Enterprise Census and UNIDO database were simply because of different classifications, the differences in rankings of textiles, rubber-plastics and non-metallic mineral products can be explained by many combined factors (besides the above-mentioned differences in sub-sector classification and year of data) such as methods of VA/LP estimation (as noted in Section 2.1), differences in treating the “outliers observations” and assigning weights to sub- sectors with different sizes, and most importantly the quality of the two datasets.

Figure 2.27b: Relative labour productivity (manufacturing = 1), 2015 (source: Authors’ calculation using UNIDO database)

Coke, refined petroleum products, nuclearThuốc fuel lá TobaccoKim loạiproducts cơ bản Basic metalsĐiện tử Phương tiện giaoElectronics thông khác Other transport equipmentHóa chất Chemicals and chemicalXe cóproducts động cơ Motor vehicles, trailers,Cheế biến semi-trailers thực phẩm Sản phẩm khoángFood and sản beveragesphi kim loại Non-metallic mineralThiết products bị điện Electrical machinery and apparatusDệt TextilesGiấy MáyPaper mó thiết and bịpaper chưa products phân loại Machinery and equipment n.e.c. Fabricated Chếmetal biến products kim loại Rubber and plasticsCao products su nhựa Wood productsGỗ tre (không (excl. furniture)tính đồ gỗ) PublishingXuất and bản printing và in ấn Furniture and manufacturing n.e.c.Đồ gỗ Leather and footwearDa giầy Wearing apparel, furMay NSLĐ tương đối (2015, UNIDO) - - 1.00 1.00 2.00 2.00 3.003.00 4.00

Box 2.5: LP growth and employment size of sub-sectors

Additional information on employment size in the assessment of sub-sector labour productivity shows the link between LP level, LP growth and the importance of manufacturing sub-sectors in terms of job creation in the economy (Figure 2.28). The figure illustrates that manufacturing sub-sectors with large employment sizes tend to sit in the left-bottom quadrant, i.e. with lower productivity levels in 2011. The non-metal products sub-sector stands out, with all three measures exceeding the respective median values.

56 Productivity and Competitiveness of Viet Nam’s Enterprises Figure 2.28: Productivity level (2011), growth rate and average employment size (2011-2016) of Box 2.4: Relative labour productivity using UNIDO 2015 data manufacturing sub-sectors 45% 140% Figure 2.27b shows sub-sectors’ relative labour productivity using UNIDO database for 2015 (the 45% 140% MedianTrung vị value, NSLĐ LP 201 20111 40% kMetalim loại cơ bản latest year for which Viet Nam’s data is available on the UNIDO database). While the trend in ranking 120% sub-sectors’ relative LP is rather similar between Figures 2.17a and 2.17b, a few differences include: 35% (i) in sub-sector classification: UNIDO database combined food and beverages into one sub-sector 100% “food and beverages” as well as “office accounting and computers” and “medical, precision and 30% 80% optical equipment” into the “electronics” sub-sector and did not have “other manufacturing” and 25% Plasticscao su, nhựa “repair and installation of machinery” compared to the VSIC sub-sector list used in this report, (ii) NonPhi ki-metalm loại 20% 60% TB điện different years of data (2016 of Enterprise Census vs. 2015 of UNIDO database), and (iii) different Paper Electrics LeatherDa giầy Giấy Điện tử 15% TextileDệt Electronics sub-sector rankings in relative LP: if taking into account the differences in sub-sector classification, WoodGỗ 40% textiles ranked four places higher using UNIDO database than using Enterprise Census data, VehiclesXe có động cơ MedianGiá trị tr value,ung vị growthăngt trưở rateng 10% OtherKhác Phương tiện giao 2011-20162011-2016 PrefabCơ Khí metal Other transport thông khác 20% rubber-plastics - five places lower than using Enterprise Census data and non-metallic mineral CabinetsĐồ gỗ Unclass.thiết bị kh ácMachines TobaccoThuốc lá BeverageĐồ uống 5% TBPharma y tế ChếFood biến thực phẩm products - six places lower than using Enterprise Census data. While differences in ranking “food” PrintingIn ấn Sửa chữa HóaChemical chất 0% and “beverages” using Enterprise Census and UNIDO database were simply because of different 0% MayApparel Repair classifications, the differences in rankings of textiles, rubber-plastics and non-metallic mineral 0 40 80 120 160 200 240 280 320 360 400 440 480 -5% -20% products can be explained by many combined factors (besides the above-mentioned differences in sub-sector classification and year of data) such as methods of VA/LP estimation (as noted in Note: The size of each bubble reflects the employment size of each respective manufacturing sub-sector. Source: Authors’ calculation, 2017 Enterprise Census Section 2.1), differences in treating the “outliers observations” and assigning weights to sub- sectors with different sizes, and most importantly the quality of the two datasets. Đáng lưu ý là, ở hầu hết các tiểu ngành, năng suất lao động trung bình của doanh nghiệp FDI đạt mức cao nhất, tiếp theo là LP của SOE và doanh nghiệp tư nhân trong nước (Hình 2.29) Figure 2.27b: Relative labour productivity (manufacturing = 1), 2015 (source: Authors’ calculation using UNIDO database) It is noted that average labour productivity of FDI is highest, followed by the LP of SOEs and domestic

Thuốc lá private firms in almost all sub-sectors (Figure 2.29). Kim loại cơ bản Điện tử Phương tiện giao thông khác Hóa chất Figure 2.29: SOE, FDI and domestic private firms’ average labour productivity by sub-sector Xe có động cơ Cheế biến thực phẩm 21 Sản phẩm khoáng sản phi kim loại (VND million, 2016 prices) Thiết bị điện Dệt Giấy Repair and installation of machinerySửa chữa và and lắp đặtequipment thiết bị Máy mó thiết bị chưa phân loại Các sảnOther phẩm manufacturing chế tạo khác Chế biến kim loại FurnitureĐồ gỗ Cao su nhựa Phương tiện giao thông khác Gỗ tre (không tính đồ gỗ) Other vehicles Xuất bản và in ấn Motor vehicles, trailers,Xe semi-trailers có động cơ Đồ gỗ MachineryMáy móc thiếtand equipmentbị chưa phân n.e.c. loại Da giầy ElectricalThiết equipment bị điện May Manufacture of computer, electronic andĐiện optical tử NSLĐ tương đối (2015, UNIDO) - 1.00 2.00 3.00 4.00 FabricatedCheế metal biến products kim loại Kim loại cơ bản Basic metals Non-metallicKhoáng mineral sản phi products kim loại Rubber and plasticsCau productssu, nhựa Medical,Thiết precision bị y tế, and chính optical xác và instruments quang học Chemicals and chemical productsHóa chất Than Coke,coke, sản refined phẩm petroleum hóa dầu và products, năng lượng nuclear nguyên fuel tử PrintingXuất and bản publishing và in ấn Paper and paper productsGiấy Box 2.5: LP growth and employment size of sub-sectors Wood, bamboo productsGỗ tre (không (excl. tình furniture) đồ gỗ) Leather, leather products and footwearDa Giày Wearing apparel,May fur Additional information on employment size in the assessment of sub-sector labour productivity TextilesDệt Tobacco productsThuốc lá shows the link between LP level, LP growth and the importance of manufacturing sub-sectors beveragesĐồ uống Chế biến thực phẩm in terms of job creation in the economy (Figure 2.28). The figure illustrates that manufacturing Food 0 1000 2000 3000 4000 5000 6000 7000 sub-sectors with large employment sizes tend to sit in the left-bottom quadrant, i.e. with lower productivity levels in 2011. The non-metal products sub-sector stands out, with all three measures Năng suất lao động,LP DN FDI FDI LPNăng Domestic suất lao Private động, DNfirm tư nhânLP trong SOE nước Năng suất lao động, DNNN exceeding the respective median values. Source: Authors’ calculation, 2017 Enterprise Census

21 In this figure, the average LP of SOE, FDI and domestic private firms in each sub-sector were calculated by dividing the total value added of each group (SOE, FDI and domestic private) by its number of workers for each sub-sector. This way, the sub-sector average LP of SOE, FDI and domestic private firm groups take into account the number of worker LP differences between the three groups in each sub-sector. If the average manufacturing LP of SOE, FDI and domestic private firms were calculated in the same way (i.e. dividing the total manufacturing value added of each group by its number of workers, which are the average numbers of workers of SOE, FDI and domestic private firms in the manufacturing sector), the results would have underestimated the number of worker differences between the three groups in each sub-sector resulting in the higher manufacturing LP of SOEs compared to FDIs.

56 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 57 Scope for future improvements in manufacturing LP

To examine the scope for structural change (from low to higher LP sub-sectors within manufacturing) to contribute to increased manufacturing LP (Table 2.5), Figure 2.30 looks at variations in LP, which declined between 2011-2016 as both commonly-used measures, coefficient of variation and maximum over the minimum ratio, dropped during this period (Table 2.5). This indicates reduced scope for structural change to contribute to future productivity improvements.

Table 2.5: Measures of variation in labour productivity (2011 and 2016)

2011 2016 Coefficient of variation */ 2,65 1,00 Max/Min 101,98 24,85 */ Coefficient of variation is equal to standard deviation divided by the mean Source: Authors’ calculation, 2017 Enterprise Census

Figure 2.30: Labour productivity level in 2011 and annual growth rate in 2011-2016 (%)

45.00% 600 600 40.00% 500 500 35.00% 30.00% 400 400 25.00% 300 300 20.00% 15.00% 200 200 10.00% 100 100 5.00% .00% - - s s t r … l… nt s ,… e o le el d r i tic tsi tali a e or s ng od gg lá y ry c y ngn t s a n c n ử c ơ p c c nd…ỗ ncen m a c c tai aà Gỗ á pấe ấ ự ucảt ạ m t c ó leá a a ri oẩ r x the x ấ hduct loạ ah db lo mloạe icle h g n F uốn ba M PGai intIni o điệ p khá nsm ồ biế tu ph uốc Te pp gi ea hWoo ar c utica,l pnl ro odu iện chines,… ehng y k Đ c ồ h Aa L Pr ó im u p cơ r ân bị, opticĐ ộ tráa , otab ing hế BeveĐ T To D hín l p kce er l kim teh d C ng đv s intc e c Ha i o s ại l p p equi ôma ó ịm ttạs s a thực c h a o ến aica iết, P b of ế nufa - i rma eta i h thd otor ce mạo a n ế a C m l eta br T ical oe e c ns ch o ,t ế n p Rubb i M ics XM iaết bine iếr M y t ả K ế b chư n iaifi h s Ph n-m h ịef a gs the pr ếbi bị Chemg C o ectr s , ca r h t b m hẩms h c C No Pr El iệnla r p he nRepa iết t đặt n h oán hiế ectr g he ả Ot gà T h t El Unc ắp Bed n K c l Ot c s ó ươn và á àn h C o y m P T á ữa M h a c ử S Năng suất Tốc độ tăng trưởng Năng suất Lao Lao động động hằng năm LPnăm in 2011 Annual Growth rate

Source: Authors’ calculation, 2017 Enterprise Census

It is important to know if sub-sectors with high levels of LP in 2011 also grew fast during 2011-2016. The correlation coefficient between these two variables is -0.71, indicating that sub-sectors with higher productivity in 2011 grew slower during 2011-2016. The correlation coefficient between LP level in 2011 and changes in employment shares for manufacturing sub-sectors is negative at -0.01. So is the correlation coefficient between LP growth rate and changes in employment shares (-0.06). All these reinforce the point made earlier that small, even negative contributions of structural change to LP improvements in manufacturing reiterate the reduced scope for structural change to achieve productivity improvements (implying efforts should focus on “within sub-sector” factors to improve manufacturing LP) in the future.

This is confirmed by results of a more rigorous shift share analysis. Figure 2.31 shows that LP grew by 11.3 percent during 2011-2016. Of a 100 percent change in LP, the ‘reallocation effect’ made up -3.3 percent, i.e. structural change adversely affected LP growth, presumably for the reasons above mentioned. The ‘interaction effect’ was also negative, at -4.3 percent. This means that during 2011-2016, labour tended to move away from sub-sectors with positive LP growth towards sub-sectors with declining LP growth (such as wearing apparel, leather-footwear, furniture). The ‘within sector effect’ was as high as 107.6 percent, which more than offset the adverse effects of structural change and interaction. As such, the results of the shift share analysis for manufacturing are broadly in line with those for the whole economy (as discussed in Section 1 and shown in Figure 1.8 - lower panel).

58 Productivity and Competitiveness of Viet Nam’s Enterprises Scope for future improvements in manufacturing LP Figure 2.31: Decomposition of labour productivity growth in manufacturing (2011-2016)

To examine the scope for structural change (from low to higher LP sub-sectors within manufacturing) to 100% contribute to increased manufacturing LP (Table 2.5), Figure 2.30 looks at variations in LP, which declined 90% between 2011-2016 as both commonly-used measures, coefficient of variation and maximum over the WithinHiệu ứng Effect, nội 80% minimum ratio, dropped during this period (Table 2.5). This indicates reduced scope for structural ngành,108% 108% change to contribute to future productivity improvements. 70% 60% Labour Tăng trưởng LP % Table 2.5: Measures of variation in labour productivity (2011 and 2016) productivity growth, (2011- 2016), 11% 50% % (2011-2016), 11%

2011 2016 40% 2,65 1,00 Coefficient of variation */ 30% 101,98 24,85 Max/Min StructuralHiệu ứng 20% */ Coefficient of variation is equal to standard deviation divided by the mean changethay effect,đổi cơ cấu,-03% -03% Source: Authors’ calculation, 2017 Enterprise Census 10%

0% Figure 2.30: Labour productivity level in 2011 and annual growth rate in 2011-2016 (%) 0% Hiệu ứng Interactiontác động -10% 45.00% 600 effect,qua lại, -04% -04% 40.00% 2011-2016 500 35.00% 30.00% 400 Source: Authors’ calculation, 2017 Enterprise Census 25.00% 300 20.00% 15.00% 200 2.2.3 Firm-Level Determinants of Labour Productivity Growth in Manufacturing 10.00% 100 5.00% .00% - This section examines the determinants of LP growth at firm level, important to LP growth within sub- g á y c y n t i a n i i n ử c ơ c c ỗ n m l a y Gỗ á ấ ấ ự ả ạ t c ó á ẩ à ấ h loạ h b lo loạ h g sectors. The firm-level determinants of LP growth in manufacturing can be established through an uốn M x Gi In điệ khá m ồ biế ph uốc gi h a c n iện ng y k Đ ồ h a ó im u cơ ân bị Đ ộ á o hế Đ T D hín k kim h ng đ c c H i o s ại p iết ô ó ịm tạ thực - h a o ến a b ế econometric analysis of the GSO’s Enterprise Census 2016, the latest dataset with detailed firm-level ế i h th tạo n n p C m l T o e c ết ch ế y t ả i ế b chư ia X i ế K h ị g h ếbi bị s C th information in Viet Nam. Results are displayed in Section A.3.2, Appendix 3. As underlined in this section, h g t b hẩm h c C iện p n iết t đặt n h oán hiế g ả gà T h t ắp n K c l c s the results for manufacturing are broadly consistent with those obtained for the whole economy, ó ươn và á àn h C o y m P T á ữa presented in the first section of this report. M h a c ử S Năng suất Tốc độ tăng trưởng Năng suất Lao Worker-related factors Lao động động hằng năm năm 2011 The presence of foreign workers in a firm is beneficial to productivity. Raising their share by 1 percentage Source: Authors’ calculation, 2017 Enterprise Census point boosts productivity by 91.7 percent(e0.651 - 1), indicating strong spill-over effects from foreign to Vietnamese workers. It is important to know if sub-sectors with high levels of LP in 2011 also grew fast during 2011-2016. The correlation coefficient between these two variables is -0.71, indicating that sub-sectors with higher With regard to human capital of workers as proxied by the level of educational attainment or training, productivity in 2011 grew slower during 2011-2016. The correlation coefficient between LP level in 2011 and compared to the base case of no training, if the ratio of workers with primary vocational certification changes in employment shares for manufacturing sub-sectors is negative at -0.01. So is the correlation rose by 1 percentage point, a firm’s LP increased by 9.7 percent. If the ratio of workers with a secondary coefficient between LP growth rate and changes in employment shares (-0.06). All these reinforce the (vocational) degree or (vocational) college degree and the ratio of workers with other certification point made earlier that small, even negative contributions of structural change to LP improvements in increased by 1 percentage point, LP would jump by 4.5 and 15.7 percent, respectively. This is slightly manufacturing reiterate the reduced scope for structural change to achieve productivity improvements different from the same analysis at economy-wide level, which indicates only short-term education (implying efforts should focus on “within sub-sector” factors to improve manufacturing LP) in the future. courses contributed to LP growth.

This is confirmed by results of a more rigorous shift share analysis. Figure 2.31 shows that LP grew by 11.3 Age of workers also matters. If the ratio of workers aged 60 or over increased by 1 percentage point, LP percent during 2011-2016. Of a 100 percent change in LP, the ‘reallocation effect’ made up -3.3 percent, jumped by 66.4 percent. This may imply that in the absence of quality information on experience, old- i.e. structural change adversely affected LP growth, presumably for the reasons above mentioned. The aged workers employed by firms may have special expertise in addition to experience. Firms with bigger ‘interaction effect’ was also negative, at -4.3 percent. This means that during 2011-2016, labour tended shares of a younger workforce under 30 years of age were more productive than those with workers to move away from sub-sectors with positive LP growth towards sub-sectors with declining LP growth aged 31-60. (such as wearing apparel, leather-footwear, furniture). The ‘within sector effect’ was as high as 107.6 percent, which more than offset the adverse effects of structural change and interaction. As such, the results of the shift share analysis for manufacturing are broadly in line with those for the whole economy (as discussed in Section 1 and shown in Figure 1.8 - lower panel).

58 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 59 Firm-related factors Capital deepening is important:

Raising the ratio of capital over labour, which measures how well workers are equipped, by 1 percent would increase LP by 0.38 percent if all firms having participated in 2012-2017 census surveys were taken into account (0.39 percent, if only firms in the 2012-2017 panel were included).

Firms’ size matters:

A size premium, as measured by the productivity gap between different firm sizes over micro-firms with less than five workers as a reference group, exhibits a concave pattern with the premium diminishing from one size group to the next (Figure 2.32). In other words, as the size of firms became larger their productivity performances relative to micro-firms became greater. However, productivity increases declined as a firm’s size grew.

Figure 2.32: Size premium

100% 90%90% 90% 84%84%

80% 76%76% 69% 70% 58% 60%

50% 42% 40%

30%

20%

10%

0% 1-4 5-9 10-24 25-4925-49 50-9950-99 100-299100-299 300-999300-999

Source: Authors’ calculation, 2017 Enterprise Census

Firms’ managerial capabilities matter

Examining the level of educational attainment of firms’ top managers, if a manager held a master degree or higher, labour productivity increased by 5 percent from the base case, where the manager only had a junior college degree or lower.

Manager’s experience as proxied by his/her age also matters, with experience premium having a non- linear form that declined as age increased.

Digitization makes a difference

Firms that used computers more intensively, have a website and use the internet in various activities were more productive (Figure 2.33). Specifically, firms with their own website, use the internet for operational management or for financial transactions have higher LP, by 5.3, 2.6 and 6.1 percent respectively, than firms that did not, other factors being equal.

60 Productivity and Competitiveness of Viet Nam’s Enterprises Firm-related factors Figure 2.33: Internet usage and labour productivity

7.00%7.00% Capital deepening is important: 6.120% 6.00%6.00% Raising the ratio of capital over labour, which measures how well workers are equipped, by 1 percent 5.253%5.253% would increase LP by 0.38 percent if all firms having participated in 2012-2017 census surveys were 5.00%5.00% taken into account (0.39 percent, if only firms in the 2012-2017 panel were included). 4.00%4.00%

Firms’ size matters: 3.00%3.00% 2.634%2.634%

A size premium, as measured by the productivity gap between different firm sizes over micro-firms with 2.00%2.00% less than five workers as a reference group, exhibits a concave pattern with the premium diminishing 1.00%1.00% from one size group to the next (Figure 2.32). In other words, as the size of firms became larger their productivity performances relative to micro-firms became greater. However, productivity increases .00%.00% declined as a firm’s size grew. FirmDN có has website its DN sử dụngFirm internetuse the internet trong quản lý DN Firmdùng use internet the internet để thanh toán own websit foroperation management forfinancial transaction Source: Authors’ calculation, 2017 Enterprise Census Figure 2.32: Size premium

100% Global market participation helps 90% 90% 84% Firms engaged in export and/or import activities were 12.3 percent more productive than those did not.

76% 80% Level of technological sophistication matters 69% 70% 58% With other factors being equal, firms in high-tech and medium manufacturing were more productive by 60% 17 and 11.2 percent, respectively than those in low-tech manufacturing (Figure 2.34) 50% 42% Figure 2.34: Labour productivity of firms in high- and medium-tech sectors relative to firms in 40% low-tech manufacturing 30% 18% 17% 20% 16% 10% 14% 0% 12% 11% 1-4 5-9 10-24 25-49 50-99 100-299 300-999 10% Source: Authors’ calculation, 2017 Enterprise Census 8% Firms’ managerial capabilities matter 6%

Examining the level of educational attainment of firms’ top managers, if a manager held a master degree 4% or higher, labour productivity increased by 5 percent from the base case, where the manager only had 2% a junior college degree or lower. 0% High-techCông nghệmanufacturing cao Medium-techCông nghệ manufacturing trung bình Manager’s experience as proxied by his/her age also matters, with experience premium having a non- Source: Authors’ calculation, 2017 Enterprise Census linear form that declined as age increased.

Digitization makes a difference

60 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 61 SOEs are most productive

Figure 2.35: Ownership and productivity of companies with different ownership forms (reference: private companies)

15.00% 11.182% 9.955%9.955% 10.00%

5.00%

.00% StateDNNN Comp. CollectiveHợp ForeignDN Nước Comp. Ngoài tác xã -5.00%

-10.00%

--12.278%12.278% -15.00%

Source: Authors’ calculation, 2017 Enterprise Census

SOEs were the most productive, 11.2 percent more than private firms with other factors being equal. They were closely followed by foreign firms, 10 percent more productive than private ones. Enterprises with collective ownership had the lowest level of productivity, 12.3 percent less than private firms (Figure 2.35). It must be noted that the regression result at first glance appears different (and counterintuitive) compared to the higher-than-average results of FDI sub-sector LP against SOEs’ in many sub-sectors as shown in Figure 2.29. This is because Figure 2.29 shows the unconditional sub-sector average LP of SOEs and FDI (which, while working in the same sub-sectors, have different employment sizes, capital concentrations, managerial capabilities), while regression compares SOEs’ and FDIs’ LP differences with conditions that all other factors such as firm size, capital concentration, managerial capability are assumed as equal.

Location matters

Figure 2.36: Labour productivity of firms in different regions (reference: Ho Chi Minh City)

50.0%50.0%

37.7%37.7% 40.0%40.0% 36.9%36.9%

30.0%30.0% 24.7% 21.2%21.2% 20.0%20.0% 15.0% 14.8% Đông nam bộ BắcTrungNorthern bộ ĐB sông (trừ TP Hồ Chí Minh) 10.0%10.0% vàCentral duyên and hải Hồng (trừ Centralmiền trung Coast 1.5% RedHà River Nội Delta South East 0.8% (except Ha Noi) (except Ho ChiMinh 0.0%0.0% ĐBMekong Mê Kông NorthernMiền núi HàHa NộiNoi HoTP Chi Hồ MinhCity Chí Minh RiverDelta -10.0%-10.0% Mountainphía bắc

-20.0%-20.0% --11.8%11.8% --20.1% --21.5% --10.3%10.3% -30.0%-30.0% -3.8% --5.5%

TấtAll cả firmsDN DNPanel trong firms điều tra 2015-2016

Source: Authors’ calculation, 2017 Enterprise Census

62 Productivity and Competitiveness of Viet Nam’s Enterprises SOEs are most productive With other factors being equal, firms in Ho Chi Minh City were the most productive, followed by Mekong River Delta, South East (excluding Ho Chi Minh City) and Ha Noi. Companies in other regions were Figure 2.35: Ownership and productivity of companies with different ownership forms considerably less productive than those in Ho Chi Minh City (Figure 2.36.). (reference: private companies)

15.00% 2.2.4 Global Integration and International Competitiveness of Viet Nam’s Manufacturing 11.182% 9.955% 10.00% While LP is a core element of firms’ (sub-sectors’ and sector’s) international competitiveness, competitiveness can also be measured by firms’ (sub-sectors’ and sector’s) export-import activities, an important integration channel for manufacturing and Viet Nam’s economy more broadly. 5.00% Performance measured by Revealed Comparative Advantage (RCA) .00% The link between exports of manufactured goods and LP and the international competitiveness of DNNN Hợp DN Nước Ngoài tác xã manufacturing sub-sectors is two-way. As indicated in the previous sections on factors determining LP -5.00% in manufacturing and sub-sectors, firms engaged in exports and/or imports were more productive than those not, other factors being equal. On the other hand, Viet Nam’s manufacturing through deepened

-10.00% global integration was aligning its manufacturing sub-sectors with their comparative advantages, measured by RCA, as a frequently-used measurement of international competitiveness. -12.278% -15.00% Analysis of UN Comtrade database found that wearing apparel, leather-footwear, textiles, wood

Source: Authors’ calculation, 2017 Enterprise Census (excluding furniture), food and beverages and furniture were sub-sectors with RCAs higher than unity since 2005 and in 2016, electronics (driven by Samsung Corporation operations in Viet Nam) and non- SOEs were the most productive, 11.2 percent more than private firms with other factors being equal. metallic mineral products joined (while tobacco, rubber-plastics fell out from) this group (Figure 2.37). They were closely followed by foreign firms, 10 percent more productive than private ones. Enterprises with collective ownership had the lowest level of productivity, 12.3 percent less than private firms (Figure Figure 2.37: RCA of manufacturing sub-sectors (2005, 2011 and 2016)

2.35). It must be noted that the regression result at first glance appears different (and counterintuitive) 12.000 compared to the higher-than-average results of FDI sub-sector LP against SOEs’ in many sub-sectors 10.000 as shown in Figure 2.29. This is because Figure 2.29 shows the unconditional sub-sector average LP of 6.914 8.000 SOEs and FDI (which, while working in the same sub-sectors, have different employment sizes, capital 6.000 5.067 concentrations, managerial capabilities), while regression compares SOEs’ and FDIs’ LP differences 3.680 4.000 with conditions that all other factors such as firm size, capital concentration, managerial capability are 2.330 1.863 1.442 1.281 2.000 1.068 0.935 0.826 0.740 0.618 0.606 0.343 0.317 0.301 0.214 0.201 0.192 0.192 assumed as equal. 0.066 0.000 i i i y y ử ỗ n lá a ạ c t Location matters à a t g ự sở ấ M dệt loạ h lo loạ h gi ện ồ điệ cơ khá a i Đ im uốc n ân a c D Đ k bị h u kim ại ng h ó i T o ô H h iết o s p h a iến m l th a Figure 2.36: Labour productivity of firms in different regions (reference: Ho Chi Minh City) m và đồn uống p T C i o ẩ ả ế b K ia xuất bản in ấnchư h ị ph s C g g t b Gỗ (không kể đồ gỗ) iện oán t giấy và xe có động cơ 50.0% thực h g thiế n K c ế ươn ó h ế bi P y m 37.7% h á 40.0% 36.9% C M

30.0% thiết bị y tế-chính xác-quang học-đồng hồ 24.7% Coke-sản phẩm hóa dầu-năng lượng hạt nhân 21.2% Đường=12005 2011 2016 Unity line 20.0% 15.0% 14.8% Đông nam bộ BắcTrung bộ ĐB sông (trừ TP Hồ Chí Minh) Source: Authors’ calculations based on UN Comtrade data 10.0% và duyên hải Hồng (trừ miền trung 1.5% Hà Nội 0.8% Notably, almost all sub-sectors, except electronics (RCA>1) and fabricated metal (RCA<1), experienced 0.0% declining RCA over the past five years (2011-2016). Recalling that RCA represents the ratio of sub- ĐB Mê Kông Miền núi Hà Nội TP Hồ Chí Minh -10.0% phía bắc sector shares in Viet Nam’s total manufacturing exports to the same sub-sector’s share in total global manufacturing exports, declining RCA values of these sub-sectors (despite increased export volumes -20.0% -11.8% as shown in the section on characteristics of manufacturing sub-sectors, and Viet Nam’s manufacturing -20.1% -21.5% -10.3% RCA rises as shown in the section on manufacturing performance) may simply be the result of significant -30.0% -3.8% -5.5% increases in export volumes of electronics, leading to the former sub-sectors’ smaller shares in Viet Tất cả DN DN trong điều tra 2015-2016 Nam manufacturing exports.

Source: Authors’ calculation, 2017 Enterprise Census

62 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 63 The electronics sub-sector experienced the fastest RCA increase since 2011. It presumably benefited from the ‘China Plus One Strategy’ employed by multinational corporations, notably Japanese and Korean in developing countries including Viet Nam, as a response to the rapidly rising costs of labour and other production inputs in China. As a result, FDI (especially by Samsung Corporation) rapidly increased its presence in Viet Nam in this sub-sector (mainly smart phone assembling), as noted in the earlier analysis of this sub-sector’s rapid growth in VA, workers and revenue share in Viet Nam’s manufacturing sector and LP. As the FDI-dominated electronics sub-sector had solid performance in the level of LP and growth of LP but not among the highest in 2016, it may reflect the predominant assembling nature of production (using simple skills).

For medium-tech manufacturing sub-sectors (non-metallic mineral products, electrical equipment, rubber-plastics, fabricated metal, other transport vehicles and basic metals) as well as low-tech sub- sectors, the RCA index was low with no signs of improvement for many years. Besides the electronics sub-sector’s increased export share in Viet Nam’s total manufacturing exports as noted above, FDI enterprises in these sub-sectors were more inward-looking: producing goods for domestic consumption (to substitute imports) rather than for exports. Meanwhile, many domestic market-focused firms in these sub-sectors could not compete in international markets and grew slowly in size and LP (exception was the food and beverage sub-sector with: (i) high level of domestic private firm participation, (ii) substantially positive net exports and (iii) an RCA which declined significantly in the last five years, but remained higher than unity).

Figure 2.37 provides a comparison of Viet Nam’s and comparator countries’ RCAs in some sub-sectors where Viet Nam’s RCA values were higher than unity. Besides the electronics sub-sector’s climbing RCA (highest among comparator countries in 2016), RCA values were consistently high across these sub-sectors than only very high in a few sub-sectors (textiles and wearing apparel in Bangladesh and leather-footwear in Cambodia). This indicates Viet Nam’s export portfolios may be more diverse than Bangladesh’s and Camdodia’s, a positive trend in terms of resilience against shocks in international demand for exports. It also indicates Viet Nam’s lower RCA in each of these sub-sectors may be because its export share in total manufacturing export was smaller due to diversification (similar to Viet Nam’s manufacturing RCA being lower than Bangladesh’s and Cambodia’s due to Viet Nam’s significant exports in other sectors - Box 2.2).

64 Productivity and Competitiveness of Viet Nam’s Enterprises The electronics sub-sector experienced the fastest RCA increase since 2011. It presumably benefited Figure 2.38: Sub-sectors’ RCA of Viet Nam and comparator countries from the ‘China Plus One Strategy’ employed by multinational corporations, notably Japanese and Korean in developing countries including Viet Nam, as a response to the rapidly rising costs of labour and LeatherDa and giày footwear WearingMay apparel RCA other production inputs in China. As a result, FDI (especially by Samsung Corporation) rapidly increased 50.00 35.00 45.00 its presence in Viet Nam in this sub-sector (mainly smart phone assembling), as noted in the earlier 45.00 30.00 analysis of this sub-sector’s rapid growth in VA, workers and revenue share in Viet Nam’s manufacturing 40.00 35.00 25.00 sector and LP. As the FDI-dominated electronics sub-sector had solid performance in the level of LP and 30.00 20.00 growth of LP but not among the highest in 2016, it may reflect the predominant assembling nature of 25.00 15.00 production (using simple skills). 20.00 15.00 15.00 10.00 10.00 For medium-tech manufacturing sub-sectors (non-metallic mineral products, electrical equipment, 5.00 5.00 5.00 rubber-plastics, fabricated metal, other transport vehicles and basic metals) as well as low-tech sub- 0.00 0.00 sectors, the RCA index was low with no signs of improvement for many years. Besides the electronics sub-sector’s increased export share in Viet Nam’s total manufacturing exports as noted above, FDI enterprises in these sub-sectors were more inward-looking: producing goods for domestic consumption (to substitute imports) rather than for exports. Meanwhile, many domestic market-focused firms in 2005 2011 2016 2005 2011 2016 these sub-sectors could not compete in international markets and grew slowly in size and LP (exception was the food and beverage sub-sector with: (i) high level of domestic private firm participation, (ii) substantially positive net exports and (iii) an RCA which declined significantly in the last five years, but ElectronicsĐiện tử RCA TextileDệt RCA remained higher than unity). 5.00 10.00 4.50 9.00 Figure 2.37 provides a comparison of Viet Nam’s and comparator countries’ RCAs in some sub-sectors 4.00 8.00 3.50 7.00 where Viet Nam’s RCA values were higher than unity. Besides the electronics sub-sector’s climbing 3.00 6.00 RCA (highest among comparator countries in 2016), RCA values were consistently high across these 2.50 5.00 sub-sectors than only very high in a few sub-sectors (textiles and wearing apparel in Bangladesh and 2.00 4.00 1.50 3.00 leather-footwear in Cambodia). This indicates Viet Nam’s export portfolios may be more diverse than 1.00 2.00 Bangladesh’s and Camdodia’s, a positive trend in terms of resilience against shocks in international 0.50 1.00 0.00 demand for exports. It also indicates Viet Nam’s lower RCA in each of these sub-sectors may be because 0.00 0.00 its export share in total manufacturing export was smaller due to diversification (similar to Viet Nam’s manufacturing RCA being lower than Bangladesh’s and Cambodia’s due to Viet Nam’s significant exports in other sectors - Box 2.2). 2005 2011 2016 2005 2011 2016

Chế Foodbiến thựcand beverage phẩm và đồRCA uống KhoángNon-metallic sản phi metal kim RCAloại WoodGỗ (excluding (trừ đồ gỗ) furniture) 3.000 3.000 2.000 9.000 2.500 1.800 8.000 1.600 7.000 2.000 1.400 6.000 1.200 5.000 1.500 1.000 5.000 4.000 1.000 0.800 0.600 3.000 0.500 0.400 2.000 0.200 1.000 0.000 0.000 0.000

2005 2011 2016 2005 2011 2016 2005 2011 2016

64 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 65 2.2.5 Participation in Global Value Chains: Domestic Content of Exports22 of Sub- Sectors

Over the last few decades, at global level, dramatic changes have occurred in the nature of international trade, increasingly through GVCs. Production processes have involved sequential, vertical trading chains stretching across many countries with specific ones specializing in particular stages of a good’s production sequence.

In modern times, GVCs shared in approximately two-thirds of global trade, according to recent estimates by David Dollar at the Brookings Institution23. In GVC-participating countries, increased participation in GVCs is associated with the rapid growth of trade in tasks (also termed “trade in value added”). “Vertical specialization”, ‘‘slicing up the value chain’’, ‘‘fragmentation’’ are different terms used for this phenomenon in literature. As participating in GVCs is almost a must for a country or firm to engage in international trade, all participants want increased shares of the GVC pie.

In the context of Viet Nam accelerating its global integration, in addition to export values and RCA, the domestic content of exports can serve as a measure of Vietnamese firms’ participation in GVCs in particular and international trade in general. With data from an input-output table, one can calculate it by using the following formula (De La Cruz et al (2013)):

DVS = Av[I-AD]-1 (2.1) of which: DVS – domestic content (or domestic vertical specialization) of exports AD = n x n domestic coefficient matrix; Av – 1× n vector of each sector j ’ s ratio of value-added to gross output;

Based on Equation (2.1), Pham Minh Thai et. al. (2018) used input-output 2012 to calculate foreign content of Viet Nam’s exports, including manufacturing. Their results were used to derive domestic content of manufacturing exports, presented in Figure 2.38. It shows that variations in domestic content of manufacturing exports across sub-sectors are relatively small at 0.23 (coefficient of variation is 0.23 – meaning that standard deviation is equal to 23 percent of mean value). At the top are food processing and non-metallic mineral product sub-sectors with shares of domestic content in exports exceeding 65 percent. These two sub-sectors also displayed relatively good LP performances (Table 2.4, Figures 2.26 and 2.27), with significant positive net exports and high levels of domestic private firms’ participation. Close to the bottom are electronics (FDI-led, large-sized and top exporting sub-sector), electrical equipment and rubber-plastics (high levels of FDI participation and negative net export values), also in the list of manufacturing sub-sectors with relatively good productivity performances (Table 2.4, Figures 2.26 and 2.27). Leather and wearing apparel present another group of sub-sectors with high levels of domestic content, led by FDI and leading in manufacturing exports, but low LP.

There appears to be no correlation between LP performance and domestic content of manufacturing sub-sector exports, instead the major inputs and level of technology used by firms in the sub-sectors seemed to “predict” the level of domestic content and LP of sub-sectors. In some high- and medium- tech sub-sectors (by Standard International Trade Classification (SITC) such as electronics, electrical equipment, motor vehicles, other vehicles, basic metals), firms tended to focus on assembling and/ or processing imported components/materials in Viet Nam and as a result, the share of local content (mainly labour) in VA tended to be low, while LP was higher than the manufacturing average LP. In low- tech sub-sectors (such as wearing apparel, leather-footwear), the share of domestic content in labour (a

22 “Domestic content” measures Viet Nam’s “participation” in GVCs through the share of (value of) domestic content in Viet Nam’s exports (and through Viet Nam’s share in global exports, it can measure Viet Nam’s domestic share of the global export pie), at the same time it: (i) due to data issues, does not distinguish contributions of foreign firms based in Viet Nam and Vietnamese firms to domestic content and (ii) does not measure (changes in) Vietnamese firms’ “trade in tasks”/“vertical specialization” or functions in GVCs. 23 Source: https://www.brookings.edu/blog/order-from-chaos/2017/07/10/global-value-chains-shed-new-light-on-trade/

66 Productivity and Competitiveness of Viet Nam’s Enterprises 2.2.5 Participation in Global Value Chains: Domestic Content of Exports22 of Sub- major share in sub-sector VA) in VA tended to be high with low LP. In some low-tech sub-sectors using Sectors imported materials (fabricated metal, furniture, rubber-plastics and wood excluding furniture), domestic firms (especially domestic private SMEs which dominate these sub-sectors) focused on labour-intensive Over the last few decades, at global level, dramatic changes have occurred in the nature of international processing and as a result domestic content and LP were low. Lastly, in sub-sectors that use locally- trade, increasingly through GVCs. Production processes have involved sequential, vertical trading supplied materials (food, beverage, non-metallic mineral products), domestic content was high and LP chains stretching across many countries with specific ones specializing in particular stages of a good’s higher than the manufacturing average LP (in these sub-sectors: LP levels were defined by labour or production sequence. capital intensity of firms, such as the LP of the more capital-intensive beverage sub-sector exceeding In modern times, GVCs shared in approximately two-thirds of global trade, according to recent estimates that of the more labour-intensive food processing sub-sector). This is largely in line with the commonly by David Dollar at the Brookings Institution23. In GVC-participating countries, increased participation agreed assessment that Viet Nam’s growth model (and manufacturing, in particular) had been based on in GVCs is associated with the rapid growth of trade in tasks (also termed “trade in value added”). exploiting low cost/simple skilled labour and natural resources. “Vertical specialization”, ‘‘slicing up the value chain’’, ‘‘fragmentation’’ are different terms used for this It should be noted that the correlation between domestic content of exports and share of VA out of phenomenon in literature. As participating in GVCs is almost a must for a country or firm to engage in revenue increased substantially, from 0.14 in 2011 to 0.31 in 2016. This means there is a positive international trade, all participants want increased shares of the GVC pie. association between increasing a firm’s internal value and raising the country’s domestic value addition In the context of Viet Nam accelerating its global integration, in addition to export values and RCA, and this association strengthened substantially during 2011-2016. the domestic content of exports can serve as a measure of Vietnamese firms’ participation in GVCs in It is noted, however, that the analysis on domestic content in this section must be treated with a degree of particular and international trade in general. With data from an input-output table, one can calculate it caution. Firstly, the analysis used the GSO 2012 input-output table which may be outdated, despite being by using the following formula (De La Cruz et al (2013)): the most recent available in Viet Nam. Secondly, data on domestic content did not distinguish between DVS = Av[I-AD]-1 (2.1) and include “domestic” content produced by domestic and Viet Nam-based FDI firms. Therefore, the analysis in this sub-section cannot answer whether localized production of (increased domestic content of which: in) exports meant increased participation of, and domestic content produced by, Vietnamese firms or DVS – domestic content (or domestic vertical specialization) of exports Viet Nam-based FDI firms. The analysis of FDI participation, as shown in in the manufacturing sub- AD = n x n domestic coefficient matrix; sector characteristics section, spotlights: (i) the higher shares of “domestic content” in textiles, wearing apparel, leather and footwear, chemicals, rubber-plastics, electronics, electrical equipment, machinery Av – 1× n vector of each sector j ’ s ratio of value-added to gross output; and equipment n.e.c, motor vehicles, other vehicles, furniture and other manufacturing are produced by Based on Equation (2.1), Pham Minh Thai et. al. (2018) used input-output 2012 to calculate foreign Viet Nam-based FDI firms which dominate these sub-sectors in terms of employment size, revenue and content of Viet Nam’s exports, including manufacturing. Their results were used to derive domestic VA share and (ii) the substantial shares of “domestic content” in beverages, basic metals and fabricated content of manufacturing exports, presented in Figure 2.38. It shows that variations in domestic content metal are produced by Viet Nam-based FDI firms which have medium employment and revenue, and of manufacturing exports across sub-sectors are relatively small at 0.23 (coefficient of variation is 0.23 large VA shares in these three sub-sectors. However, to fully answer this important question for policy – meaning that standard deviation is equal to 23 percent of mean value). At the top are food processing formulation and adjustment, further analysis is needed and especially more granular research at sub- and non-metallic mineral product sub-sectors with shares of domestic content in exports exceeding 65 sector and firm levels, including in the next stage of this study. percent. These two sub-sectors also displayed relatively good LP performances (Table 2.4, Figures 2.26 Figure 2.39: Domestic content of exports of manufacturing sub-sectors and 2.27), with significant positive net exports and high levels of domestic private firms’ participation. Beverage…Đồ uống 70%70% Close to the bottom are electronics (FDI-led, large-sized and top exporting sub-sector), electrical Chế biến thựcFood… phẩm 67%67% Sản phẩmNon-metal khoáng chất products… phi kim loại 65%65% equipment and rubber-plastics (high levels of FDI participation and negative net export values), also in Leather…May 60%60% Apparel…Da giày 60%60% ThiếtPharmaceutical bị y tê, chính xác production… và quang học 58%58% the list of manufacturing sub-sectors with relatively good productivity performances (Table 2.4, Figures SửaRepair, chữa maintenance… và lắp đặt thiết bị 56%56% Xuất bảnPaper… và in ấn 55%55% 2.26 and 2.27). Leather and wearing apparel present another group of sub-sectors with high levels of Printing…Giấy 55%55% Chemical products…Hóa chất 51%51% domestic content, led by FDI and leading in manufacturing exports, but low LP. Other means of transport…Trung bình 49%49% Phương tiện giao thôngMean khác 49%49% MedianThuốc lá 48%48% Tobacco…Trung vị 48%48% There appears to be no correlation between LP performance and domestic content of manufacturing Other processing and manufacturing…Dệt 46%46% Sản phâm chế biến chếTextile… tạo khác 46%46% MotorXe vehicles… có động cơ 45%45% sub-sector exports, instead the major inputs and level of technology used by firms in the sub-sectors MetalKim products… loại cơ bản 44%44% Than coke, sản phẩmCoal, hóa dầu petroleum và năng lượng products… hạt nhân 43%43% seemed to “predict” the level of domestic content and LP of sub-sectors. In some high- and medium- PrefabricatedChế biến metal… kim loại 42%42% Beds, cabinets, tables, chairs…Đồ gỗ 40% tech sub-sectors (by Standard International Trade Classification (SITC) such as electronics, electrical Rubber,Cao plastic… su nhựa 38% Electrical equipment…Thiết bị điện 35% Electronics, PC, optical products…Điện tử 33% equipment, motor vehicles, other vehicles, basic metals), firms tended to focus on assembling and/ UnclassifiedMáy móc machines, và thiết bị equipment… chưa phân loại 32% Wood… or processing imported components/materials in Viet Nam and as a result, the share of local content Gỗ (không kể đồ gỗ) 31% 0% 10% 20% 30% 40% 50% 60% 70%70% 80%80% (mainly labour) in VA tended to be low, while LP was higher than the manufacturing average LP. In low- tech sub-sectors (such as wearing apparel, leather-footwear), the share of domestic content in labour (a Source: Adapted from Pham Minh Thai et al (2018)

66 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 67 Performance measured by VA-to-output ratio

Figure 2.43, based on UNIDO database, shows that electronics, leather-footwear and wearing apparel were the top three in 2015. Other sub-sectors to post significant improvements in VA-to-output ratios between 2011-2015 included printing and motor vehicles and to a lesser extent furniture, chemicals, electrical equipment, wood (excluding furniture), rubber-plastics, fabricated metal and textiles. The remaining sub-sectors experienced VA-to-output ratio declines. Among the five top exporting sub- sectors, four (electronics, wearing apparel, leather-footwear and furniture) had VA-to-output ratios higher than the manufacturing average that substantially increased between 2011-2015. The only sub- sector with large positive net exports, but a VA-to-output ratio lower than the manufacturing average was food and beverages (noting UNIDO and UN Comtrade classification placed food and beverages in one sub-sector).

On the other side, sub-sectors with VA-to-revenue ratios lower than 20 percent included machinery n.e.c, tobacco, textiles, food and beverages, paper, basic metal, fabricated metal rubber-plastics. Chemicals, other vehicles and electrical equipment were also among the sub-sectors with VA-to-revenue ratios lower than the manufacturing average, but higher than 20 percent.

It should be noted that 2016 VA-to-revenue ratios were calculated using 2017 Enterprise Census data, which resulted in VA-to-revenue ratio calculations using two data sources and different estimation methods (described in Section 2.1) remaining consistent. However, there were some differences and thus, the report’s analysis based on this ratio should be treated with caution (Box 2.6).

Figure 2.40: Value added-to-revenue ratio (%), 2011 and 2015

ToànTotal ngành Manufacturing chế biến chế tạo ElectronicsĐiện tử Leather, leather products and footwearDa giày Wearing apparel, furMay Xuất bảnPrinting và in ấn Motor vehicles, trailers, semi-trailersXe có động cơ Non-metallicSản phẩm khoáng mineral chát phiproducts kim loại Furniture; manufacturing n.e.c.Đồ gỗ OtherPhương transport tiện giao equipment thông khác Chemicals and chemical productsHóa chất Electrical machinery and apparatusThiết bị điện Wood products (excl.Gỗ furniture) (trừ đồ gỗ) Paper and paper productsGiấy Rubber and plasticsCao products su và nhựa Fabricated metalChế biến products kim loại Chế biếnFood thực and phẩm beverages và dồ uống BasicKim loại metals cơ bản Coke,refinedThan coke, sản petroleum phẩm hóa products,nucleardầu và nhiên liệu hạt fuelnhân TextilesDệt MachineryMáy móc and thiết equipment bị chưa phân n.e.c loại Tobacco productsThuốc lá 0.0%0.0% 5.0%5.0% 10.0%10.0% 15.0%15.0% 20.0%20.0% 25.0% 30.0% 35.0% 40.0% 45.0% 50.0%

20112011 20152015

Source: Authors’ calculation, UNIDO database

68 Productivity and Competitiveness of Viet Nam’s Enterprises Performance measured by VA-to-output ratio BOX 2.6: Value added-to-revenue ratio calculated using Enterprise Census data Figure 2.43, based on UNIDO database, shows that electronics, leather-footwear and wearing apparel were the top three in 2015. Other sub-sectors to post significant improvements in VA-to-output ratios Figure 2.41 shows the VA-to-revenue ratios of sub-sectors calculated using Enterprise Census between 2011-2015 included printing and motor vehicles and to a lesser extent furniture, chemicals, data. While the key trend of VA improvement in sub-sectors remained similar to ratios calculated electrical equipment, wood (excluding furniture), rubber-plastics, fabricated metal and textiles. The using UNIDO database, some differences were observed. Namely: (i) in many sub-sectors, VA- remaining sub-sectors experienced VA-to-output ratio declines. Among the five top exporting sub- to-revenue ratios using Enterprise Census data were higher than VA-to-output ratios calculated sectors, four (electronics, wearing apparel, leather-footwear and furniture) had VA-to-output ratios using the UNIDO database, (ii) sub-sectors where large differences were observed included higher than the manufacturing average that substantially increased between 2011-2015. The only sub- wearing apparel, non-metallic mineral products, paper, rubber-plastics, basic metal, textiles and sector with large positive net exports, but a VA-to-output ratio lower than the manufacturing average electronics (noting differences in classification mentioned in Box 2.4). was food and beverages (noting UNIDO and UN Comtrade classification placed food and beverages in one sub-sector). Figure 2.41.: 2016 Value added-to-revenue ratio (%), using Enterprise Census data Wearing apparel,May fur KimBasic loại cơmetals bản Sản phẩmNon-metallic khoáng mineralchát phi products kim loại On the other side, sub-sectors with VA-to-revenue ratios lower than 20 percent included machinery n.e.c, Leather, leather products and footwearDa giày Repair and installation ofLắp machinery đặt và sửa and chữa equipment thiết bị Các sản phẩm chếOther biến manufacturing chế tạo khác tobacco, textiles, food and beverages, paper, basic metal, fabricated metal rubber-plastics. Chemicals, Rubber and plasticsCao su products và nhựa FurnitureĐồ gỗ Paper and paper productsGiấy other vehicles and electrical equipment were also among the sub-sectors with VA-to-revenue ratios TextilesDệt PrintingXuất and bản publishing và in ấn beveragesĐồ uống lower than the manufacturing average, but higher than 20 percent. MachineryMáy móc thiết and bịequipment chưa phân n.e.c. loại Trung bình ngànhManufacturing chế biến averagechế tạo Medical, Thiếtprecision bị y tế-chínhand optical xác-quang instruments học ElectricalThiết equipment bị điện Phương tiện giaoOther thông vehicles khác It should be noted that 2016 VA-to-revenue ratios were calculated using 2017 Enterprise Census data, FabricatedChế metal biến products kim loại Wood, bamboo products (excl.Gỗ (trừ furniture) đồ gỗ) which resulted in VA-to-revenue ratio calculations using two data sources and different estimation Chemicals and chemical Hóaproducts chất Motor vehicles, trailers,Xe semi-trailers có động cơ Manufacture of computer, electronic Electronicsand optical methods (described in Section 2.1) remaining consistent. However, there were some differences and Tobacco productsThuốc lá Chế biến thực phẩmFood thus, the report’s analysis based on this ratio should be treated with caution (Box 2.6). Than coke,Coke, sảnrefined phẩm petroleum hóa dầu vàproducts, nhiên liệu nuclear hạt nhân fuel 0.00% 10.00% 20.00% 30.00%30.00% 40.00%40.00% 50.00%50.00% 60.00%60.00%

Figure 2.40: Value added-to-revenue ratio (%), 2011 and 2015 Source: Authors’ calculation from Enterprise Census 2017

Toàn ngành chế biến chế tạo Điện tử The differences are largely attributed to methods of VA estimation as well as other factors (and Da giày May their combined effects), such as different definitions (output and revenue), different years of Xuất bản và in ấn Xe có động cơ data, different classifications, data/level of depreciation and taxes used in VA estimations using Sản phẩm khoáng chát phi kim loại Đồ gỗ Enterprise Census data in this report and the quality of Enterprise Census and UNIDO database. Phương tiện giao thông khác Hóa chất Thiết bị điện Gỗ (trừ đồ gỗ) Giấy Across ownership forms, FDI firms tended to have higher VA to-revenue ratios than SOEs and domestic Cao su và nhựa Chế biến kim loại private firms. Similarly, large firms’ VA-to-revenue ratios exceeded SMEs. The differences resembled Chế biến thực phẩm và dồ uống Kim loại cơ bản those in LP as shown in the section on sub-sector LP performances (Figures 2.26,2.27.). Than coke, sản phẩm hóa dầu và nhiên liệu hạt nhân Dệt Máy móc thiết bị chưa phân loại Figure 2.42a: Value added-to-revenue ratio by ownership (2016) Thuốc lá Than coke, sản phẩm hóa dầu và nhiên liệu hạt nhân 0.0% 5.0% 10.0% 15.0% 20.0% 25.0% 30.0% 35.0% 40.0% 45.0% 50.0% May60.0% Chế biến thực phẩm Kim loại cơSOE bản Private ThuốcF láDI 2011 2015 50.0% Sản phẩm khoáng chát phi kim loại Electronics 40.0% Food Repair and installation of machinery and… 50.00% Beverages OtDahe rgiày manufacturing 4530.0%.00% Tobacco pXeroduc cót sđộng cơ Source: Authors’ calculation, UNIDO database 40.00% Furniture 20.0% Textiles Lắp đặt và sửa chữa thiết bị 35.00% Hóa chất 3010.0%.00% Other vehicles 25.00% Wearing apparel, fur Các sản phẩm chế biến chế tạo khác 200.0%.00% Gỗ (trừ đồ gỗ) Motor vehicles, trailers, semi-trailers 15.00% Leather, leather products and footwear 10.00% Cao su và nhựa 5.00% Chế biến kim loại Machinery and equipment n.e.c. 0.00% Wood, bamboo products (excl. furniture) Đồ gỗ Phương tiện giao thông khác Electrical equipment Paper and paper products Giấy Thiết bị điện

Manufacture of computer, electronic and… Dệt Thiết bị yP tế-chínhrinting an dxác-quang publishing học Xuất bản và in ấn Máy móc thiết bị chưa phân loại Fabricated metal products Đồ uống Coke, refined petroleum products, nuclear…

Basic metals Chemicals and chemical products Non-metallic mineral products Medical, precision and optical instruments Rubber and plastics products FDI DN tư nhân trong nước DNNN

Source: Authors’ calculation from Enterprise Census 2017

68 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 69 Figure 2.42b: Value added-to-revenue ratio by firm size (2016)

Than coke, sản phẩm hóa dầu và nhiên liệu hạt nhân LMayarg0.5e firms Chế biếnSM thựcEs phẩm Kim loại cơ bản 0.45 Thuốc lá 0.4 Sản phẩm khoáng chát phi kim loại Food Electronics Repair and installation of machinery and… 500.35.00% Beverages 0.3 OtDaher giày manufacturing 45.00% Tobacco pXeroduc cót sđộng cơ 400.25.00% Furniture 35.0.200% Textiles Lắp đặt và sửa chữa thiết bị 300.15.00% Hóa chất Other vehicles 25.0.100% Wearing apparel, fur 200.05.00% Các sản phẩm chế biến chế tạo khác 0 Gỗ (trừ đồ gỗ) Motor vehicles, trailers, semi-trailers 15.00% Leather, leather products and footwear 10.00% Cao su và nhựa 5.00% Chế biến kim loại Machinery and equipment n.e.c. 0.00% Wood, bamboo products (excl. furniture)

Đồ gỗ Phương tiện giao thông khác Electrical equipment Paper and paper products Giấy Thiết bị điện Manufacture of computer, electronic and… Printing and publishing Dệt Thiết bị y tế-chính xác-quang học Xuất bản và in ấn Máy móc thiết bị chưa phân loại Fabricated metal products Đồ uống Coke, refined petroleum products, nuclear… Basic metals Chemicals and chemical products Non-metallic mineral products Medical, precision and optical instruments Rubber and plastics products DN lớn DNVVN

Source: Authors’ calculation from Enterprise Census 2017

As international comparisons can shed light on Viet Nam’s sub-sector competitiveness, the 2015 VA-to- output ratios of comparator countries as shares of Viet Nam’s are presented in Figure 2.43 (constructed using data from UNIDO database24, in 2015).

Figure 2.43: Comparator countries’ VA-to-output ratios as percentage of Viet Nam’s

Tỷ suấtComparator VA/đầu ra countries' của các nước VA-to-Output so sánh tính ratios theo as % percentage tỷ suất VA/đầu of Viet ra của Nam's Việt Nam

Toàn ngànhTotal chếManufacturing biến chế tạo FurnitureĐồ gỗ OtherPhuwong transport tiện giao equipment thông khác Motor vehicles, trailers, semi-trailersXe có động cơ Electrical machinery andThiết apparatus bị điện ElectronicsĐiện tử MachineryMáy móc and thiết equipment bị chưa phân n.e.c loại Fabricated Chếmetal biến products kim loại BasicKim loại metals cơ sở SảnNon-metallic phẩm khoáng mineral chát phi products kim loại Rubber and plasticsCao su products và nhựa Chemicals and chemical productsHóa chất ThanCoke,refined coke, sản phẩmpetroleum hóa dầu products,nuclear và nhiên liệu hạt nhân fuel PrintingIn ấn Paper and paper productsGiấy Wood products (excl.Gỗ furniture)(trù đố gỗ) Leather, leather products and footwearDa giày Wearing apparel,May fur TextilesDệt Tobacco productsThuôc lá Chế biến Foodthực phẩm and beveragesvà đồ uống 0.0% 100.0% 200.0% 300.0% 400.0% 500.0% 600.0% 700.0%

HànROK quốc Philippines Malaysia NhậtJapan bản IndonesiaIndonesia ẤnIndia độ TrungChina quốc

Source: Authors’ calculation, UNIDO database

Figure 2.43 shows that Viet Nam’s ratios were higher than China’s and India’s in several sub-sectors (wearing apparel, leather-footwear, wood products excluding furniture, electronics, electrical machinery, motor vehicles, other transport equipment and furniture). While Viet Nam’s electronics sub-sector VA- to-output ratio was higher than most comparator countries and its VA-to-output ratios were “within reach” of comparator countries in almost all sub-sectors, its VA-to-output ratio gaps with comparator countries remained wide in several sub-sectors (food and beverages, tobacco, textiles, wood products excluding furniture, paper, fabricated metal and machinery, equipment n.e.c., and coke-refined petroleum products-nuclear fuel).

24 See the above footnote 9 on differences between revenue and output.

70 Productivity and Competitiveness of Viet Nam’s Enterprises Figure 2.42b: Value added-to-revenue ratio by firm size (2016) Wage Growth and Competitiveness of Manufacturing

Than coke, sản phẩm hóa dầu và nhiên liệu hạt nhân May 0.5 Chế biến thực phẩm A topic of current hot policy debate in Viet Nam is wages and LP growth. Debate was initiated and fuelled Kim loại cơ bản 0.45 Thuốc lá 0.4 Sản phẩm khoáng chát phi kim loại 0.35 Electronics by concern that wage growth is outpacing LP growth in favour of workers that results in erosion of 0.3 Da giày 0.25 Xe có động cơ international competitiveness of (Vietnamese and Viet Nam-based FDI) firms in the manufacturing 0.2 Lắp đặt và sửa chữa thiết bị 0.15 Hóa chất 0.1 sector, which is labour intensive and open to international competition. 0.05 Các sản phẩm chế biến chế tạo khác 0 Gỗ (trừ đồ gỗ) Data from the ILO database indicates that during 2004-2015, real wages grew considerably faster than Cao su và nhựa Chế biến kim loại LP (Table 2.6) in Viet Nam (and modestly in Malaysia and Thailand). Đồ gỗ Phương tiện giao thông khác Giấy Thiết bị điện Table 2.6: Wage and productivity growth in Viet Nam and Asian countries 2004-2015 (average annual real Dệt Thiết bị y tế-chính xác-quang học Xuất bản và in ấn Máy móc thiết bị chưa phân loại wage growth deflated by the CPI, %) Đồ uống Country Productivity Average wage DN lớn DNVVN growth rates growth rates Source: Authors’ calculation from Enterprise Census 2017 China 9.1 8.8

As international comparisons can shed light on Viet Nam’s sub-sector competitiveness, the 2015 VA-to- Indonesia 3.6 2.6 output ratios of comparator countries as shares of Viet Nam’s are presented in Figure 2.43 (constructed Malaysia 2.1 2.5 24 using data from UNIDO database , in 2015). Philippine 2.6 0.4 Figure 2.43: Comparator countries’ VA-to-output ratios as percentage of Viet Nam’s Singapore 1.8 1.2 Thailand 2.7 3.5 Tỷ suất VA/đầu ra của các nước so sánh tính theo % tỷ suất VA/đầu ra của Việt Nam Vietnam 4.4 5.8 Toàn ngành chế biến chế tạo Đồ gỗ Source: Nguyen Duc Thanh (2017), calculation from ILO database Phuwong tiện giao thông khác Xe có động cơ Thiết bị điện Experts blamed minimum wages, adjusted too fast, that resulted in excessive upward pressure on real Điện tử Máy móc thiết bị chưa phân loại wages in Viet Nam (Figure 2.44)25. The World Bank’s international comparison also revealed that Viet Chế biến kim loại Kim loại cơ sở Nam’s private sector minimum wage was high relative to other countries, with the average ratio of Sản phẩm khoáng chát phi kim loại 26 Cao su và nhựa minimum to median wage at 58 percent . Hóa chất Than coke, sản phẩm hóa dầu và nhiên liệu hạt nhân 27 In ấn Figure 2.44: Regional minimum wages , CPI and GDP per capita in Viet Nam, 2009-2016 Giấy Gỗ (trù đố gỗ) (2008=100) Da giày May 600600 Dệt Thuôc lá 500 Chế biến thực phẩm và đồ uống 500 0.0% 100.0% 200.0% 300.0% 400.0% 500.0% 600.0% 700.0% 400400 Hàn quốc Philippines Malaysia Nhật bản Indonesia Ấn độ Trung quốc 300300 Source: Authors’ calculation, UNIDO database 200200

Figure 2.43 shows that Viet Nam’s ratios were higher than China’s and India’s in several sub-sectors 100100

(wearing apparel, leather-footwear, wood products excluding furniture, electronics, electrical machinery, 0 motor vehicles, other transport equipment and furniture). While Viet Nam’s electronics sub-sector VA- 2009 2010 2011 2012 2013 2014 2015 2016 to-output ratio was higher than most comparator countries and its VA-to-output ratios were “within MW category I MW category II MW category III reach” of comparator countries in almost all sub-sectors, its VA-to-output ratio gaps with comparator MW category IV GDP per capita CPI Ghi chú: trước tháng 10 2011, lương tối thiểu vùng được áp dụng cho các doanh nghiệp trong nước. countries remained wide in several sub-sectors (food and beverages, tobacco, textiles, wood products Note: Before October 2011, regional minimum wage qpplicable to domestic enterprises. excluding furniture, paper, fabricated metal and machinery, equipment n.e.c., and coke-refined petroleum Source: VEPR (2017) products-nuclear fuel). However, analysis of the Enterprise Census 2011-2016 dataset finds that during this period, manufacturing productivity growth (11.3 percent per annum) outpaced manufacturing real wage growth (8.7 percent per

25 Source: VEPR (2017) 26 Source: World Bank (2016) 27 In Viet Nam, minimum wages differ across four regions considered to have different costs of living. Furthermore, up until 2011, there were two minimum wages applicable to domestic and foreign enterprises. Since 2007, to implement Viet Nam’s WTO obligations on national treatment which do not allow dual pricing to discriminate against foreign firms, the minimum wage for domestic firms was 24 See the above footnote 9 on differences between revenue and output. adjusted up aggressively and the two have been unified since 2011.

70 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 71 annum). Only less than half of manufacturing sub-sectors have annual rate of wage growth greater than that of LP growth (Figure 2.45, left of the vertical line), and amongst them, coke-petroleum products- nuclear fuel, printing, repair and installation of machinery were less open to international competition. The coefficient of correlation between these two variables is 0.4. Among the sub-sectors with annual LP growth outpacing wage growth, the largest gap was observed in the basic metals sub-sector (33 percentage points), followed by non-metallic minerals (15.78 percentage points).

Figure 2.45: Viet Nam’s productivity (LPG) and wage growth in manufacturing, 2011-2016 (%)

550.0%0

440.0%0 Tăng trưởng NSLĐ>tăng trưởng lương Tăng trưởng NSLĐWG 33.109 330.0%0

220.0%0 15.878 10 10.0% 4.364.918 5.176 6.458 2 l… ,… 3.871 … t

s 2.592.736 2.836 1

s 2.305 s s .64.47241.137 nt 0 .188 s r e -.44-.7446nd… o -.657 r tal o r -.875

-1.133 ts tic l e e le 0.0% el -2.45-2.116 9 g

nce -2.988 ng d p c uc t r s od uc t t i c a a p e -6.159 m d a th e o icl e duc t d m e x utica l a n

n s p chines,… F o -10 optic a int i b a ing a , ta b P , r od u r o pp a e a r e h Wo o T e , p l c e s

-10.0% pr o C L p tr a P A t s s m a te d

T o v

l p int e l e r Beve r l p r , P a equ i a d

-20 c e of rm a c um ic a et a m bin e -20.0% oto r ic s et a l e ca l , ifi e b r n s n Rub b M s pr o i r a M Ph a r

-30 s e tr o , c a n- m

-31.872 ef a s l a Chem i ectr i

-30.0% h e m P r p e N o r ectr o E l Rep a l , -40 O t Be d E l Un c -40.0% h e Co a O t Tăng trưởngLP NSLĐ growth RTăngeal trưởngwage lương gro wthựcth tế LP-RealNSLĐ wage -lương thực tế Source: Authors’ calculation, 2017 Enterprise Census

Furthermore, there should not necessarily be a one-to-one relationship between: (i) LP and wage growth and (ii) maintaining competitiveness (i.e. wage growth higher than LP should not necessarily lead to lower competitiveness). This nexus can be analyzed more rigorously through a simple economic model, which will show this relationship depends on technological change. Specifically, one can assume the production function has the Cobb-Douglas form with time varying technology . Under the assumption of constant returns to scale (CRS), one have αt+βt=1.

It can be shown (details are given in Section A.1.5, Appendix 1) that if workers’ wage growth in real terms is proportional, by factor , to LP growth, wage growth can be said to be competitiveness-neutral, as it is in line with technological change in the economy or sector. If the rate of wage growth is greater (smaller) than , then wage growth is competitiveness-hurting (competitiveness-enhancing), both may not be sustainable in the long term, because they either hurt the owner of capital (the former case), or worker (the latter case).

To calculate , one can estimate Cobb-Douglas production function with contant returns to scale (Table 2.7, columns 1, 2 and 3). This table shows that wearing apparel is the only sub-sector in the list where wage growth can be characterized as competitiveness-damaging, as the actual wage growth rate exceeded the competitiveness-neutral one by 11.8 percent (column 7). In the leather-footwear sub- sector, although real wage grew faster than LP by 0.45 percentage points (column 9), it is still seen as competitiveness-enhancing, as such a gap is more than justified by technological change in favour of labour28. In other sub-sectors electronics and textiles as well as in average manufacturing, wage growth is lower than productivity growth.

28 In this simple production function, L integrates both quantity and quality of labour and therefore increases in labour share over time (i.e. βt+1>βt) may well be explained by the improvements of skills.

72 Productivity and Competitiveness of Viet Nam’s Enterprises annum). Only less than half of manufacturing sub-sectors have annual rate of wage growth greater than Table 2.7: Productivity and wage growth and competitiveness in selected manufacturing sub- that of LP growth (Figure 2.45, left of the vertical line), and amongst them, coke-petroleum products- sectors nuclear fuel, printing, repair and installation of machinery were less open to international competition. The coefficient of correlation between these two variables is 0.4. Among the sub-sectors with annual LP growth outpacing wage growth, the largest gap was observed in the basic metals sub-sector (33 percentage points), followed by non-metallic minerals (15.78 percentage points).

Figure 2.45: Viet Nam’s productivity (LPG) and wage growth in manufacturing, 2011-2016 (%)

50.0%

40.0% Tăng trưởng NSLĐ>tăng trưởng lương Tăng trưởng NSLĐ

72 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 73 Table 2.8: Summary of manufacturing sub-sector characteristics and performance

Large Firm SOE Domestic Private VSIC paricipation level SME paricipation FDI participation participation participation level Labor Productivity(2016, VND Domestic VA-to-output ratio (%), Wage growth vs. LP growth (LP code Subsectors Subsector size Net export (2016) (2016) level (2016) level (2016) level (2016) (2016) million, 2016 price) RCA (2016) contents (%) 2015 (UNIDO databse) growth minus wage growth) 283 (medium level among low (medium in VN manufacturing 10 Food Large High employment) Medium Low High subsectors) 67% Positive 664 (high level among VN manufacturing subsectors) - food and beverage: is 17.4% low level, big Medium (low in Large positive low (medium in Medium (high in Medium (high catching up with gaps with Negative but competitiveness 11 Beverages employment) net export High employment) VA) Low inemployment) comparator countries 1.281 70% comparators enhancing 407 (high level among VN manufacturing subsectors) - 12.6% low level, big Medium positive Medium (low in gaps with comparators gaps with 12 Tobacco products Small net export High low employment) High Low widened Less than unity 48% comparators Positive 263 (medium level among VN manufacturing 2.33 (within 14.8% low level, big Large (medium Negative net Medium (low in subsectors), catching up reach to gaps with 13 Textiles in revenue) export High revenue) High Low Medium with comparator countries comparators) 46% comparators Positive 81 (low level among VN manufacturing subsectors) - 35.1% - high level, Large net gaps with comparators within reach to Negative and competitiveness 14 Wearing apparel, fur Large positive export High Low High Low Medium widened/slowly narrowed 5.067 60% comparators damaging 87(low level among VN manufacturing subsectors) - large gaps with 38.3% high level, Leather, leather products and Large net comparators higher/ within reach Negative, ompetitiveness 15 footwear Large positive export High Low High Low Low (medium in VA) widened/slowly narrowed 6.941 60% to comparators enhancing 141 (low level among VN manufacturing subsectors) - 20.3% medium level, Wood, bamboo products (excl. Small (medium Medium positive gaps with comparators fast narrow gaps with 16 furniture) in employment) net export Medium High Low Low High narrowing 1.863 31% comparators Positive 310 medium level among VN's manufacturing subsectors, (large gaps 19.5% low (high EC Small (medium Negative net High (medium in Medium (high in Medium (high in with comparators data), big gaps with 17 Paper and paper products in VA) export employment) employment) VA) Low High narrowing fast) Less than unity 55% comparators Positive 146 - low level among VN's manufacturing subsector - 34.1%, high level, Medium positive gaps with comparators narrow gaps with Negative but competitiveness 18 Printing and publishing Small net export Medium High Medium Low High slowly narrowing Less than unity 55% comparators enhancing 998 - high level among VN's High Medium (high in manufacturing subsectors 15.2% low level, big Coke, refined petroleum products, Negative net Low (medium in (medium in employment and (within reach to gaps with 19 nuclear fuel Small export High employment) Low employment) low in revenue) comparators) Less than unity 43% comparators Negative

457 - high level among VN's Medium (low manufacturing subsectors, 21% medium level, inemployment, Negative net High (medium in Medium (high in large gap with comparators big gaps with Negative but competitiveness 20 Chemicals and chemical products high in revenue) export High Medium employment) Low employment) narrowing fast Less than unity 51% comparators enhancing 311 medium level among VN's manufacturing Medical, precision and optical Negative net Low (medium in subsectors, NA int. 21 instruments Small export High employment) Medium Low High comparisons Less than unity 58% NA Positive 328 medium level among VN's manufacturing 18.6% low level (high - Negative net subsectors (large gaps EC data), big gaps Negative but competitiveness 22 Rubber and plastics products Large export High Medium High Low High narrowing slowly) Less than unity 38% with comparators enhancing 460 - high level among VN's 25.1% medium level manufacturing subsectors, (high EC data), big Medium positive Medium (low in Medium (low in (large gaps narrowing gaps with 23 Non-metallic mineral products Large net export High VA) employment) Low High slowly) 1.068 65% comparators Positive 2031 High level among VN's 17.1% low level (high - manufacturing subsectors EC data), medium Large (low in Negative net Low (medium in Medium (high in (within reach to gaps with 24 Basic metals employment) export High employment) VA) Low High (low in VA) comparators) Less than unity 42% comparators Positive 285 medium level among VN's manufacturing subsectors, (largte gaps 17.7% low level, big Negative net High (medium in High (medium in Medium (high in with comparators gaps with 25 Fabricated metal products Large export employment) revenue) VA) Low High narrowed slowly) Less than unity 44% comparators Positive 451 high level among VN's 44.4% high level (low manufacturing subsectors EC data), higher or Large net (within reach to within reach to 26 Electronics Large positive export High Low High Low Low comparators) 3.68 33% comparators Positive 340 medium level amon VN's manufacturing subsectors, (large gaps 20.3% medium level, Negative net Low (medium in with comparators medium gaps with 27 Electrical equipment Medium export High VA) High Low Medium narrowed slowly) Less than unity 35% comparators Positive 278 medium level among VN's manufacturing subsectors, Large gaps 14.6% low level, big Negative net Medium (high in Medium (high in with comparators gaps with 28 Machinery and equipment n.e.c. Small export High employment) High Low employmment) narrowing fast Less than unity 32% comparators Positive 404 high level among VN's manufacturing subsectors, (largte gaps with 31.5% high level, Motor vehicles, trailers, semi- Medium (high in Negative net Medium (low in comparators narrowed within reach to 29 trailers revenue) export High Low High Low employmment) slowly) Less than unity 45% comparators Positive 406 High level among VN's manufacturing subsectors, small gaps with 22.6% medium level, Medium (low in Medium positive Low (medium in comparators narrowed medium gaps with Negative but competitiveness 30 Other vehicles employment) net export High Low High Low empployment) fast) Less than unity 49% comparators enhancing 112 low level among VN's 25.1% medium level, Medium (high in Large net manufacturing subsectors medium gaps with Negative but competitiveness 31 Furniture employment) positive export High Medium High Low High (gaps narrowing fast. Less than unity 40% comparators enhancing 156 low level among VN's Small (medium Low (medium in Low (medium in manufacturing subsectors, 32 Other manufacturing in employment) NA High revenue) High Low employment) NA int. comparison 1.442 46% NA Positive 212 medium level amongf VN's manufacturing Repair,installation of machinery Medium High (medium in subsectors, NA int. 33 and equipment Small NA Medium High Low (high in VA) VA) comparison Less than unity 56% NA Negative Low (medium in 248 (large gaps narrowing 23.4% medium gaps Manufacturing NA NA High employment) High Low Medium slowly) 1.02 49% with comparators Positive Notes: Subsector size, FDI/SOE/Domestic firm participation level are measured by their employment, revenue and VA shares in manufacturing/subsectors. Color coding large/high/fast: (1) subsector share in total manufacturing employment, revenue and VA above 4%, (2) FDI, SOE and domestic private firm share in subsector employment, revenue and VA above 45%, (3) LP higher than VND350million; (4) Domestic contents>60%; (5) VA/output>30%, (6) net export>USD5 billion Medium: (1) subsector share in total manufacturing employment, revenue and VA 2-4%; (2) FDI, SOE and domestic private firm share in subsector employment, revenue and VA 20-45%; (3) LP >200 and <350 VND million; (4) Domestic contents>40% and <60%; (5) VA/ouput<30% and >20%, (6)0

Manufacturing sub-sectors differ by importance to the economy

Manufacturing sub-sectors differ in size and contributions to exports, measured by sub-sector employment, revenue and VA shares in the manufacturing sector and net export values indicating levels of importance to the manufacturing sector and economy as a whole.

74 Productivity and Competitiveness of Viet Nam’s Enterprises Table 2.8: Summary of manufacturing sub-sector characteristics and performance • Food and beverages, furniture (medium-tech), textiles and wearing apparel29, leather-footwear (low- tech) and electronics (high-tech) are large-sized sub-sectors (except beverages and furniture are

Large Firm SOE Domestic Private VSIC paricipation level SME paricipation FDI participation participation participation level Labor Productivity(2016, VND Domestic VA-to-output ratio (%), Wage growth vs. LP growth (LP medium-sized) making important contributions to the manufacturing sector and economy in terms code Subsectors Subsector size Net export (2016) (2016) level (2016) level (2016) level (2016) (2016) million, 2016 price) RCA (2016) contents (%) 2015 (UNIDO databse) growth minus wage growth) 283 (medium level among low (medium in VN manufacturing of job creation, revenue, value addition and exports. 10 Food Large High employment) Medium Low High subsectors) 67% Positive 664 (high level among VN manufacturing subsectors) - food and beverage: is 17.4% low level, big Medium (low in Large positive low (medium in Medium (high in Medium (high catching up with gaps with Negative but competitiveness • Wood (excluding furniture), printing and tobacco are small-sized (and low-tech), other vehicles 11 Beverages employment) net export High employment) VA) Low inemployment) comparator countries 1.281 70% comparators enhancing 407 (high level among VN manufacturing subsectors) - 12.6% low level, big (high-tech) is medium-sized and non-metallic mineral products is (medium-tech) a large-sized sub- Medium positive Medium (low in gaps with comparators gaps with 12 Tobacco products Small net export High low employment) High Low widened Less than unity 48% comparators Positive sector contributing to manufacturing exports (with RCA>1). 263 (medium level among VN manufacturing 2.33 (within 14.8% low level, big Large (medium Negative net Medium (low in subsectors), catching up reach to gaps with 13 Textiles in revenue) export High revenue) High Low Medium with comparator countries comparators) 46% comparators Positive 81 (low level among VN • Sub-sectors with high and medium technology, large and medium-sized, negative net exports manufacturing subsectors) - 35.1% - high level, Large net gaps with comparators within reach to Negative and competitiveness 14 Wearing apparel, fur Large positive export High Low High Low Medium widened/slowly narrowed 5.067 60% comparators damaging (import substitution) and RCA<1 include: (rubber-plastics, basic metal, fabricated metal), large- 87(low level among VN manufacturing subsectors) - large gaps with 38.3% high level, sized sub-sectors (chemicals, electrical equipment and motor vehicles), medium-sized sub-sectors Leather, leather products and Large net comparators higher/ within reach Negative, ompetitiveness 15 footwear Large positive export High Low High Low Low (medium in VA) widened/slowly narrowed 6.941 60% to comparators enhancing 141 (low level among VN (coke-petroleum-nuclear fuel, paper products, medical precision-optical equipment, machinery manufacturing subsectors) - 20.3% medium level, Wood, bamboo products (excl. Small (medium Medium positive gaps with comparators fast narrow gaps with 16 furniture) in employment) net export Medium High Low Low High narrowing 1.863 31% comparators Positive and equipment n.e.c.) and small-sized sub-sectors (repair and installation of machinery and other 310 medium level among VN's manufacturing subsectors, (large gaps 19.5% low (high EC manufacturing). Small (medium Negative net High (medium in Medium (high in Medium (high in with comparators data), big gaps with 17 Paper and paper products in VA) export employment) employment) VA) Low High narrowing fast) Less than unity 55% comparators Positive 146 - low level among VN's manufacturing subsector - 34.1%, high level, Medium positive gaps with comparators narrow gaps with Negative but competitiveness SOEs, FDI and domestic private firms differ in sub-sector participation levels, sizes and 18 Printing and publishing Small net export Medium High Medium Low High slowly narrowing Less than unity 55% comparators enhancing 998 - high level among VN's High Medium (high in manufacturing subsectors 15.2% low level, big operate with weak linkages Coke, refined petroleum products, Negative net Low (medium in (medium in employment and (within reach to gaps with 19 nuclear fuel Small export High employment) Low employment) low in revenue) comparators) Less than unity 43% comparators Negative

457 - high level among VN's FDI is the biggest player in manufacturing, dominating the majority of sub-sectors with large sizes Medium (low manufacturing subsectors, 21% medium level, inemployment, Negative net High (medium in Medium (high in large gap with comparators big gaps with Negative but competitiveness 20 Chemicals and chemical products high in revenue) export High Medium employment) Low employment) narrowing fast Less than unity 51% comparators enhancing and net export values as well as in high- and medium-technology import substitution sub-sectors, 311 medium level among VN's manufacturing Medical, precision and optical Negative net Low (medium in subsectors, NA int. but has weak linkages to domestic firms. 21 instruments Small export High employment) Medium Low High comparisons Less than unity 58% NA Positive 328 medium level among VN's manufacturing 18.6% low level (high - Negative net subsectors (large gaps EC data), big gaps Negative but competitiveness 22 Rubber and plastics products Large export High Medium High Low High narrowing slowly) Less than unity 38% with comparators enhancing FDI was the biggest employer (hiring 55.52 percent of workers) in manufacturing, had the largest 460 - high level among VN's 25.1% medium level manufacturing subsectors, (high EC data), big share in manufacturing revenue (58.46 percent) and MVA (64.44 percent) in 2016. FDI participation Medium positive Medium (low in Medium (low in (large gaps narrowing gaps with 23 Non-metallic mineral products Large net export High VA) employment) Low High slowly) 1.068 65% comparators Positive 2031 High level among VN's 17.1% low level (high - levels (measured by employment, revenue and VA shares in manufacturing) was high in 12 out of 24 manufacturing subsectors EC data), medium Large (low in Negative net Low (medium in Medium (high in (within reach to gaps with 24 Basic metals employment) export High employment) VA) Low High (low in VA) comparators) Less than unity 42% comparators Positive manufacturing sub-sectors (taking into account sub-sectors where FDI’s VA share was high, the number 285 medium level among VN's manufacturing subsectors, (largte gaps 17.7% low level, big of sub-sectors FDI dominated was 16 out of 24). FDI also had a high participation level in four out of Negative net High (medium in High (medium in Medium (high in with comparators gaps with 25 Fabricated metal products Large export employment) revenue) VA) Low High narrowed slowly) Less than unity 44% comparators Positive 451 high level among VN's 44.4% high level (low the top five manufacturing exporting sub-sectors (textiles and wearing apparel, leather-footwear, manufacturing subsectors EC data), higher or Large net (within reach to within reach to 26 Electronics Large positive export High Low High Low Low comparators) 3.68 33% comparators Positive electronics and furniture), and a medium participation level in food and beverages - the fifth-largest 340 medium level amon VN's manufacturing exporting manufacturing sub-sector (FDI had a high VA share in beverages, while domestic private firms subsectors, (large gaps 20.3% medium level, Negative net Low (medium in with comparators medium gaps with 27 Electrical equipment Medium export High VA) High Low Medium narrowed slowly) Less than unity 35% comparators Positive have high participation levels in food processing and medium levels in beverages). FDI had high and 278 medium level among VN's manufacturing subsectors, Large gaps 14.6% low level, big medium participation levels in other vehicles and non-metallic mineral products, respectively – two of Negative net Medium (high in Medium (high in with comparators gaps with 28 Machinery and equipment n.e.c. Small export High employment) High Low employmment) narrowing fast Less than unity 32% comparators Positive 404 high level among VN's the remaining four sub-sectors with positive net exports. Among import substitution (with negative net manufacturing subsectors, (largte gaps with 31.5% high level, Motor vehicles, trailers, semi- Medium (high in Negative net Medium (low in comparators narrowed within reach to exports) sub-sectors, FDI had high and medium participation levels in large sub-sectors (rubber-plastics, 29 trailers revenue) export High Low High Low employmment) slowly) Less than unity 45% comparators Positive 406 High level among VN's manufacturing subsectors, basic metals and fabricated metals), medium-sized sub-sectors (electrical equipment, motor vehicles, small gaps with 22.6% medium level, Medium (low in Medium positive Low (medium in comparators narrowed medium gaps with Negative but competitiveness chemicals) and small-sized sub-sectors (medical precision-optical equipment, machinery n.e.c, other 30 Other vehicles employment) net export High Low High Low empployment) fast) Less than unity 49% comparators enhancing 112 low level among VN's 25.1% medium level, Medium (high in Large net manufacturing subsectors medium gaps with Negative but competitiveness manufacturing, paper, printing). FDI only had low participation in three small-sized sub-sectors: wood 31 Furniture employment) positive export High Medium High Low High (gaps narrowing fast. Less than unity 40% comparators enhancing 156 low level among VN's Small (medium Low (medium in Low (medium in manufacturing subsectors, and bamboo products (excluding furniture), coke-petroleum-nuclear fuel and repair and installation of 32 Other manufacturing in employment) NA High revenue) High Low employment) NA int. comparison 1.442 46% NA Positive 212 medium level amongf VN's manufacturing machinery. Repair,installation of machinery Medium High (medium in subsectors, NA int. 33 and equipment Small NA Medium High Low (high in VA) VA) comparison Less than unity 56% NA Negative Low (medium in 248 (large gaps narrowing 23.4% medium gaps Manufacturing NA NA High employment) High Low Medium slowly) 1.02 49% with comparators Positive It is important to note that despite FDI’s major employment, revenue and VA shares in many sub-sectors, Notes: Subsector size, FDI/SOE/Domestic firm participation level are measured by their employment, revenue and VA shares in manufacturing/subsectors. Color coding its linkages with domestic firms remain weak and especially so in high and medium-tech sub-sectors. large/high/fast: (1) subsector share in total manufacturing employment, revenue and VA above 4%, (2) FDI, SOE and domestic private firm share in subsector employment, revenue and VA above 45%, (3) LP higher than VND350million; (4) Domestic contents>60%; (5) VA/output>30%, (6) net export>USD5 billion Medium: (1) subsector share in total manufacturing employment, revenue and VA 2-4%; (2) FDI, SOE and domestic private firm share in subsector employment, revenue and VA 20-45%; (3) LP >200 and <350 VND million; (4) Domestic contents>40% and <60%; (5) VA/ouput<30% and >20%, (6)0

29 The textile sub-sector had negative net exports, but as noted earlier it produced inputs for wearing apparel’s exports so taking textiles and wearing apparel together in one “sub-sector”, the combined sub-sector was large-sized and a positive net exporter.

74 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 75 Domestic private firms (commonly SMEs) are the second biggest player in manufacturing, dominating in two sub-sectors with large positive net export values, high participation levels in several sub-sectors with large-sized and positive net exports and limited participation in export- led sub-sectors where FDI and SOEs dominate.

Domestic private firms are the second biggest manufacturing employer (40.88 percent), with the second highest revenue shares in manufacturing revenue (35.47 percent) and MVA (30.79 percent). Food and beverages and furniture are only two (of five top exporting) sub-sectors that domestic private firms have high levels of participation in (medium participation and high employment in beverages). Domestic firms also dominate in non-metallic mineral products (large-sized with medium positive net exports), wood and bamboo (excluding furniture) and printing (small-sized and medium positive net exports) and medium participation in wearing apparel and textiles (large-sized and positive net exports). In the remaining large and medium-sized sub-sectors with positive net exports where FDI (and SOEs) dominate, domestic firms only have low participation. In import substitution sub-sectors, domestic firms have high participation levels in basic metals (with low VA), fabricated metal, rubber-plastics, medical precision- optical equipment, repair and installation of machinery. This contrasts with FDI firms and SOEs that tend to be large-sized, while domestic private firms are often SMEs.

Productivity and competitiveness performance varies substantially across manufacturing sub-sectors Top exporting sub-sectors

• Electronics

Is the (high-tech) sub-sector with the highest VA and revenue shares, large employment share and positive net exports30 among Viet Nam’s manufacturing sub-sectors. Its high RCA has risen exponentially since 2010 and reached 3.68 in 2016, higher than electronics RCA values of almost all comparator countries. Electronics’ LP - the fourth-highest among Viet Nam manufacturing sub-sectors in 2016 with rises between 2011-2016 (while wage growth was lower than LP growth) - was assessed as “within reach” of comparator countries’ levels. The VA-to-output ratio was 44.4 percent (UNIDO database) (17.29 percent - EC data), the highest compared to Viet Nam’s other manufacturing sub-sectors and higher/“within reach” of comparator countries, while LP growth outstripped wages.

It is noteworthy that: (i) with FDI’s VA share at 98.41 percent, most electronics output and exports were generated by big foreign corporations largely from Japan and RoK, such as Canon, LG, Nidec, Panasonic and Samsung (attracted by Viet Nam’s cheap labour and preferential policies to set up assembly plants), (ii) electronics FDI firms almost exclusively focused on the assembling component stage (mostly imported by FDI enterprises and the remainder produced by other FDI enterprises or self-produced, World Bank, 2017) and packaging final products in Viet Nam, with low backward-forward linkages. Electronics exports had low domestic content (mainly labour) levels of 33 percent among the lowest compared to other manufacturing sub-sectors despite signs of improvement (Box 2.7) and (iii) key products contributed most to the sub-sector’s revenue and exports, including smartphones and computers. TVs and other electronics products were also for the domestic market.

30 The electronics sub-sector’s export value reached USD22.9 billion in 2012 and doubled to USD57 billion by 2015, accounting for 30.7 percent of Viet Nam's total exports. Viet Nam became the 12th largest electronics exporter in the world and ranked third in ASEAN within five years (DBS Research Group: 2015). In particular, Viet Nam is becoming a hub for electronics production and the second largest exporter of smartphones in the world behind China (BDG, 2016). Currently, computers and electronic products exported from Viet Nam are presented in more than 30 countries and regions. The EU remains the largest importer of this product group from Viet Nam, followed by the US. China and RoK are also two major export markets for electronic goods in Viet Nam. Because FDI firms almost exclusively focus on the stage of assembling imported components and packaging the final products in Viet Nam, exports go hand- in-hand with imports in the electronics sub-sector. According to the General Department of Customs, the value of electronic goods imports has increased steadily and by 2015, the import value of the electronics industry was estimated at USD36.39 billion, about 22 percent of Viet Nam's total imports. Imports of parts and components mainly came from major countries such as RoK (29.1 percent), China (22.5 percent) and ASEAN (15.4 percent), Japan, Taiwan and the US.

76 Productivity and Competitiveness of Viet Nam’s Enterprises Domestic private firms (commonly SMEs) are the second biggest player in manufacturing, dominating in two sub-sectors with large positive net export values, high participation levels in Box 2.7: Signs of strengthening linkages between Samsung and domestic firms several sub-sectors with large-sized and positive net exports and limited participation in export- led sub-sectors where FDI and SOEs dominate. As Samsung’s presence in Viet Nam grows it allows domestic firms to accumulate experience and build necessary capacity to meet the stringent requirements of Samsung. Its Vietnamese Domestic private firms are the second biggest manufacturing employer (40.88 percent), with the second suppliers have grown from four first-tier vendors in 2014 to 29 first-tier vendors in 201731. Similarly, highest revenue shares in manufacturing revenue (35.47 percent) and MVA (30.79 percent). Food and the number of second-tier suppliers has also increased to nearly 300 enterprises32. Samsung plans beverages and furniture are only two (of five top exporting) sub-sectors that domestic private firms to increase the number of Vietnamese suppliers to 500 by 2020 (Minh Duc, 2017). Samsung Viet have high levels of participation in (medium participation and high employment in beverages). Domestic Nam made great strides in increasing its localization rate, from 35 percent in 2014 to 51 percent in firms also dominate in non-metallic mineral products (large-sized with medium positive net exports), 201633. Domestic supplier-firms such as Minh Nguyen company (established in 2015 and in 2018 wood and bamboo (excluding furniture) and printing (small-sized and medium positive net exports) became a first-tier Samsung supplier of plastic components), Goldsun (started cooperation with and medium participation in wearing apparel and textiles (large-sized and positive net exports). In the Samsung in 2010 and supplies 20 percent of Samsung smartphone cases) and others engaged remaining large and medium-sized sub-sectors with positive net exports where FDI (and SOEs) dominate, in producing Samsung smartphone components (such as 3D lenses, metallic cases, displays and domestic firms only have low participation. In import substitution sub-sectors, domestic firms have high batteries) have contributed to Samsung’s localization rate of 58 percent in 201834. This presumably participation levels in basic metals (with low VA), fabricated metal, rubber-plastics, medical precision- provides an important explanation for the substantial reduction in Viet Nam’s trade deficit with optical equipment, repair and installation of machinery. This contrasts with FDI firms and SOEs that tend China, as the large flows of exports and imports between the two countries are through the to be large-sized, while domestic private firms are often SMEs. Samsung-led GVC.

Productivity and competitiveness performance varies substantially across manufacturing Such encouraging developments are largely attributed to joint efforts of the Government of Viet sub-sectors Nam and Samsung to strengthen the corporation’s linkages with Vietnamese firms. Specifically, the MOIT and Samsung Viet Nam signed a memorandum of understanding in March 2018 on Top exporting sub-sectors training 200 qualified Vietnamese consultants to advise and train Vietnamese supporting • Electronics enterprises to participate more deeply in Samsung’s GVC. As part of this programme, on 10 July 2018, Samsung Viet Nam in cooperation with the MOIT completed the First Supporting Industry Is the (high-tech) sub-sector with the highest VA and revenue shares, large employment share and positive Consultant Training Course, in which the first 25 Vietnamese experts spent three months with net exports30 among Viet Nam’s manufacturing sub-sectors. Its high RCA has risen exponentially since Korean experts in the field of manufacturing innovation and quality improvement35. 2010 and reached 3.68 in 2016, higher than electronics RCA values of almost all comparator countries. Electronics’ LP - the fourth-highest among Viet Nam manufacturing sub-sectors in 2016 with rises Vietnamese firms, however, face challenges when attempting to move up the value chain to upgrade between 2011-2016 (while wage growth was lower than LP growth) - was assessed as “within reach” of technological capabilities, increase value addition and sustain long-term growth. Interviews comparator countries’ levels. The VA-to-output ratio was 44.4 percent (UNIDO database) (17.29 percent with Vietnamese firms supplying parts and components or receiving technical assistance from - EC data), the highest compared to Viet Nam’s other manufacturing sub-sectors and higher/“within Samsung as potential suppliers revealed that although working with Samsung was rewarding, it reach” of comparator countries, while LP growth outstripped wages. posed considerable risks and barriers which many Vietnamese firms found difficult to overcome. Notably, Samsung’s orders are in large volume with short lead times, among the biggest challenges It is noteworthy that: (i) with FDI’s VA share at 98.41 percent, most electronics output and exports were for Vietnamese firms. This contrasts with Japanese firms that often place orders well in advance, generated by big foreign corporations largely from Japan and RoK, such as Canon, LG, Nidec, Panasonic sometimes a year before actual delivery. Besides “traditional” barriers such as access to capital and and Samsung (attracted by Viet Nam’s cheap labour and preferential policies to set up assembly plants), land, weak linkages between local (often SME) firms, weak capacity of the ecosystem to support (ii) electronics FDI firms almost exclusively focused on the assembling component stage (mostly imported local firms in application of new technology and innovation (especially in high-quality technology by FDI enterprises and the remainder produced by other FDI enterprises or self-produced, World Bank, universities and research centres to support domestic firms with necessary human resources), 2017) and packaging final products in Viet Nam, with low backward-forward linkages. Electronics exports R&D, technology and innovation-related inputs remain key challenges to strengthen local firms’ had low domestic content (mainly labour) levels of 33 percent among the lowest compared to other participation in (Samsung-led in the case of Samsung smartphones and in general) GVCs. manufacturing sub-sectors despite signs of improvement (Box 2.7) and (iii) key products contributed most to the sub-sector’s revenue and exports, including smartphones and computers. TVs and other 3132333435 electronics products were also for the domestic market. Overall, Viet Nam’s (FDI-dominated) electronics sub-sector (especially computers and smartphones) can be assessed as “competitive” at the stage of final product assembling with a “promising” increase 30 The electronics sub-sector’s export value reached USD22.9 billion in 2012 and doubled to USD57 billion by 2015, accounting for in domestic suppliers of components, as long as: (i) foreign firms can maintain competitiveness of 30.7 percent of Viet Nam's total exports. Viet Nam became the 12th largest electronics exporter in the world and ranked third in ASEAN products, (ii) Viet Nam’s sub-sector LP and wages remain competitive amid high risks of losing repetitive within five years (DBS Research Group: 2015). In particular, Viet Nam is becoming a hub for electronics production and the second largest exporter of smartphones in the world behind China (BDG, 2016). Currently, computers and electronic products exported from assembling jobs to automation and (iii) Viet Nam’s domestic firms can accelerate participation as major Viet Nam are presented in more than 30 countries and regions. The EU remains the largest importer of this product group from Viet Nam, followed by the US. China and RoK are also two major export markets for electronic goods in Viet Nam. Because FDI firms almost exclusively focus on the stage of assembling imported components and packaging the final products in Viet Nam, exports go hand- 31 https://news.samsung.com/vn/29-doanh-nghiep-viet-la-nha-cung-cap-cap-1-cho-samsung in-hand with imports in the electronics sub-sector. According to the General Department of Customs, the value of electronic goods 32 http://nhipcaudautu.vn/thuong-truong/samsung-nham-toi-muc-tieu-nang-ty-le-noi-dia-hoa-len-57-3324291/ imports has increased steadily and by 2015, the import value of the electronics industry was estimated at USD36.39 billion, about 22 33 https://news.samsung.com/vn/khao-sat-doanh-nghiep-viet-nam-tao-co-hoi-tham-gia-chuoi-cung-ung-toan-cau-cua-samsu percent of Viet Nam's total imports. Imports of parts and components mainly came from major countries such as RoK (29.1 percent), 34 http://cafef.vn/hanh-trinh-cua-thuong-hieu-viet-tien-vao-chuoi-gia-tri-samsung-mitsubishi-vi-tu-ai-20190202090256359.ch China (22.5 percent) and ASEAN (15.4 percent), Japan, Taiwan and the US. 35 https://news.samsung.com/vn/be-giang-khoa-dao-tao-chuyen-gia-tu-van-cong-nghiep-ho-tro-lan-thu-nhat

76 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 77 suppliers in domestic and global value chains. Electronics sub-sector industries that focus on assembling electronic home appliances for the domestic market (to substitute imports) also risk FDI firms moving assembling plants to other countries if Viet Nam is no longer competitive compared to other countries as a result of trade agreements. Looking forward, strengthening backward-forward linkages between FDI and domestic firms, accelerating domestic firms’ linkages in GVCs and moving to higher value chain stages will need prioritization.

• Wearing apparel and textiles, leather-footwear

Labour intensive and low-tech wearing apparel is Viet Nam’s biggest sub-sector employer, with the fifth largest revenue and VA shares among manufacturing sub-sectors, while the more capital- intensive textiles ranked ninth in employment and 11th in VA. The labour intensive and low-tech leather- footwear was the second-largest sub-sector employer, yet ranked seventh and sixth in revenue and VA, respectively. The wearing apparel and leather-footwear sub-sectors were the second and third largest exporting manufacturing sub-sectors, while textiles had negative net exports and provided important inputs for the export-oriented wearing apparel sub-sector. All three sub-sectors were among the few manufacturing sub-sectors that have RCAs higher than unity: leather-footwear had the highest RCA 6.94 (only lower than Cambodia’s), followed by wearing apparel at 5.07 (only lower than Bangladesh’s and textiles ranked fourth) after electronics with an RCA of 2.33 (lower than Bangladesh’s, China’s and India’s) in 2016.

All three sub-sectors have high participation levels of FDI firms, the VA share of which in leather-footwear was 82.4 percent, textiles (71.42 percent) and wearing apparel (58.5 percent), while SOE participation was low and domestic private firms’ participation was medium. Domestic private sector firms: (i) consisted mainly of SMEs, the largest employment share in wearing apparel, a sub-sector with productivity less sensitive to firm size and (ii) are small employers, despite the medium VA share in leather-footwear. In these sub-sectors, like in many others, FDI firms’ backward-forward linkages with domestic firms were very weak: backward linkages in wearing apparel were only 7 (of 100) and almost zero in leather- footwear, while higher in textiles (56.8). Forward linkages in leather-footwear were (30), apparel (24) and textiles (13).

Despite being leaders in terms of exports, employment, revenue, VA and dominated by FDI, the LP of wearing apparel, leather-footwear were the lowest among Viet Nam’s manufacturing sub-sectors and LP gaps with comparator countries either widened or narrowed very slowly. More capital-intensive textiles had medium LP and the gap with comparator countries had narrowed, its VA-to-output ratio was low and gaps with comparator countries were wide. The VA-to-output ratios of leather-footwear, and wearing apparel were high and gaps with comparator countries were closing fast. It should be noted that in leather-footwear, and wearing apparel wage growth was higher than LP growth (while the gap between LP and wage growth remained “competitiveness enhancing” in leather-footwear, in wearing apparel it was assessed as “competitiveness damaging” reflecting the stronger competition for simple skilled and young workers within Viet Nam and weaker international competitiveness of the sub-sector). Textiles had LP growth higher than wage growth.

Overall, Viet Nam’s wearing apparel, textiles and leather-footwear sub-sectors mainly focused on the final stage of physical production in value chains (producing final products based on foreign firms’ orders) and can be assessed as competitive in the short term. However, this “competitiveness” was weakening based on widening or slowly-narrowing LP gaps with comparator countries (except textiles) and wearing apparel’s damaging competitiveness wage growth in addition to its long struggle to move vertically up value chains (Box 2.8). These sub-sectors’ future competitiveness depends on several factors: (i) foreign firms’ ability to maintain competitiveness of products, (ii) Viet Nam’s sub-sector LP and VA-to-output ratio’s ability to continue growing and (iii) how the high risk of losing repetitive jobs to automation will unfold. Given these sub-sectors’ large sizes and importance to Viet Nam’s economy in terms of GDP, exports and employment, the impacts of not effectively increasing productivity and competitiveness

78 Productivity and Competitiveness of Viet Nam’s Enterprises suppliers in domestic and global value chains. Electronics sub-sector industries that focus on assembling as well as managing automation risks to jobs will be highly significant for Viet Nam’s socio-economic electronic home appliances for the domestic market (to substitute imports) also risk FDI firms moving development. In the short-term, the leather-footwear, textiles and wearing apparel sub-sectors will assembling plants to other countries if Viet Nam is no longer competitive compared to other countries benefit from CPTTP, which will: (i) allow a 95-98 percent tax reduction on Viet Nam’s exports and (ii) as a result of trade agreements. Looking forward, strengthening backward-forward linkages between have stricter product of origin requirements for Viet Nam’s garment and leather-footwear exports, FDI and domestic firms, accelerating domestic firms’ linkages in GVCs and moving to higher value chain which in principle will lead to higher demand for made-in-Viet Nam products. As such, CPTTP will offer stages will need prioritization. significant opportunities for leather-footwear, textiles and wearing apparel sub-sectors for growth and development of local value chains, including local firms to climb value chains. However, as noted • Wearing apparel and textiles, leather-footwear above, challenges in realizing such opportunities remain significant. Viet Nam’s wearing apparel and Labour intensive and low-tech wearing apparel is Viet Nam’s biggest sub-sector employer, with leather-footwear sub-sectors heavily depend on textile inputs imported from China and India, the key the fifth largest revenue and VA shares among manufacturing sub-sectors, while the more capital- question is how domestic textile firms (large and capital intensive, and especially producing synthetic intensive textiles ranked ninth in employment and 11th in VA. The labour intensive and low-tech leather- fibers) will grow in number and size to seize the new demand. If Viet Nam’s domestic firms fail to grasp footwear was the second-largest sub-sector employer, yet ranked seventh and sixth in revenue and VA, these opportunities, Viet Nam-based FDI firms will swoop and build on their current VA dominance in the respectively. The wearing apparel and leather-footwear sub-sectors were the second and third largest textile sub-sector (as experienced in wearing apparel and leather-footwear sub-sectors, many FDI firms exporting manufacturing sub-sectors, while textiles had negative net exports and provided important have already increased investment in Viet Nam-located textile plants in anticipation of CPTTP approval). inputs for the export-oriented wearing apparel sub-sector. All three sub-sectors were among the few manufacturing sub-sectors that have RCAs higher than unity: leather-footwear had the highest RCA Box 2.8: Increasing local content and VA-to-output ratio via development of local value 6.94 (only lower than Cambodia’s), followed by wearing apparel at 5.07 (only lower than Bangladesh’s chain – motorcycle and garment and textile cases and textiles ranked fourth) after electronics with an RCA of 2.33 (lower than Bangladesh’s, China’s and India’s) in 2016. “Local value chain” is defined as the chain of production that involves a network of interconnected local enterprises that generate domestic value within the chain. The policy aim is to develop a value All three sub-sectors have high participation levels of FDI firms, the VA share of which in leather-footwear chain that improves local firms’ technical capabilities, competitiveness and market integration was 82.4 percent, textiles (71.42 percent) and wearing apparel (58.5 percent), while SOE participation was within and across local industries first in the local market and then in the global market. It is low and domestic private firms’ participation was medium. Domestic private sector firms: (i) consisted expected that as local firms move along the value chain, more value is added to their component mainly of SMEs, the largest employment share in wearing apparel, a sub-sector with productivity less production and final outputs. Table 2.9 provides typical stages of textile and garment (upper panel) sensitive to firm size and (ii) are small employers, despite the medium VA share in leather-footwear. and motor vehicles (lower panel) industries that can also be generally applied for other industries In these sub-sectors, like in many others, FDI firms’ backward-forward linkages with domestic firms such as electronics, food and beverages, and electrical equipment. were very weak: backward linkages in wearing apparel were only 7 (of 100) and almost zero in leather- footwear, while higher in textiles (56.8). Forward linkages in leather-footwear were (30), apparel (24) and Table 2.9: Value chain stages (textiles and garments (upper panel) and motor vehicles (lower textiles (13). panel)) Despite being leaders in terms of exports, employment, revenue, VA and dominated by FDI, the LP of Stage 1: Raw Stage 2: Textile Stage 3: Clothing Stage 4: Stage 5: Retailers wearing apparel, leather-footwear were the lowest among Viet Nam’s manufacturing sub-sectors and LP material suppliers companies manufacturers Wholesalers (marketing, sales) gaps with comparator countries either widened or narrowed very slowly. More capital-intensive textiles (cotton, wool, silk, (thread, yarn, (cutting, and exporters had medium LP and the gap with comparator countries had narrowed, its VA-to-output ratio was low and hemp, oil/gas) fabric, polymers, assembling, (labelling, gaps with comparator countries were wide. The VA-to-output ratios of leather-footwear, and wearing synthetic fibres, finishing) packaging, apparel were high and gaps with comparator countries were closing fast. It should be noted that in cloth) shipping) leather-footwear, and wearing apparel wage growth was higher than LP growth (while the gap between LP and wage growth remained “competitiveness enhancing” in leather-footwear, in wearing apparel it Stage 1: Local Stage 2: Suppliers Stage 3: Suppliers Stage 4: Suppliers Stage 5: Local was assessed as “competitiveness damaging” reflecting the stronger competition for simple skilled and assembling to local or foreign to first-tier foreign to foreign-led lead firms with young workers within Viet Nam and weaker international competitiveness of the sub-sector). Textiles assemblers suppliers firms domestic-branded had LP growth higher than wage growth. products

Overall, Viet Nam’s wearing apparel, textiles and leather-footwear sub-sectors mainly focused on the final Development processes require capability development of local firms, especially from a stage of physical production in value chains (producing final products based on foreign firms’ orders) technological standpoint. Capability building of local firms takes time and goes through stages and can be assessed as competitive in the short term. However, this “competitiveness” was weakening within the value chain. based on widening or slowly-narrowing LP gaps with comparator countries (except textiles) and wearing apparel’s damaging competitiveness wage growth in addition to its long struggle to move vertically up In Stage 1, foreign lead firms relocate assembling facilities to developing countries and hire local value chains (Box 2.8). These sub-sectors’ future competitiveness depends on several factors: (i) foreign workers to assemble all components imported from abroad (or cutting and assembling cloth) into firms’ ability to maintain competitiveness of products, (ii) Viet Nam’s sub-sector LP and VA-to-output final products with foreign bands. There is little technology diffusion at this stage, as assembling ratio’s ability to continue growing and (iii) how the high risk of losing repetitive jobs to automation will does not require a transfer of technology from advanced to developing countries. In addition, a unfold. Given these sub-sectors’ large sizes and importance to Viet Nam’s economy in terms of GDP, number of foreign component suppliers, especially those with established relationships with lead exports and employment, the impacts of not effectively increasing productivity and competitiveness firms in their home countries, move production to developing countries to provide components for foreign lead firms.

78 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 79 These suppliers are regarded as first-tier suppliers, implying they sell components directly to lead firms. In Stage 2, local firms move to produce basic and low-tech components for local and foreign assemblers. In Stage 3, local firms become second-tier suppliers. They establish relationships and supply specific and more complex components to first-tier suppliers. The local second-tier suppliers benefit from collaboration with first-tier suppliers and lead firms. At this stage, many foreign firms transfer technology and expertise to support technical learning to these local suppliers.

Technology adoption and technical learning also speed up, corresponding to the component’s level of complexity. In the final two stages, local firms become first-tier suppliers, supplying complex parts directly to foreign lead firms (Stage 4) and become the lead firms themselves (Stage 5). As lead firms, they produce Vietnamese-branded products. From a developing country’s perspective, Stages 4 and 5 indicate high levels of technological upgrading, especially when local firms become capable of designing, manufacturing, and branding final products using local components and labour. This local value chain does not distinguish whether local firms supply for local industries or foreign buyers.

In the motorcycle industry, during 1995-2000, Vietnamese firms mainly engaged in the first stage, and Viet Nam failed to achieve significant technology transfers. Although there were numerous local firms (mainly SOEs that entered into joint ventures with Japanese firms) engaged in the production of replacement parts, these firms were largely outside the procurement networks of foreign lead firms. As the result of “China shock” (cheap Chinese motorcycles penetrating and capturing the low-end Vietnamese market), local content and movement of (domestic and foreign Viet Nam-based firms) firms to Stages 2, 3 and 4 accelerated. As a result: prior to 2005, this industry accounted for 3.1 percent of the total industrial production value of the country, which grew to 23.9 percent by 2007 and starting in 2010, the major foreign motorcycle-makers in Viet Nam started to export production surplus to other markets (in 2011, Honda exported 300,000 motorcycles to the Philippines, Cambodia, Laos and Afghanistan) and by 2012 the local content ratio, measuring the percentage of domestic parts in each motorcycle, was between 70 and 95 percent (for instance, Honda’s local content ratio reached nearly 95 percent on some of its models). The motorcycle industry provides an exemplary case study of industrial upgrading predominantly driven by FDI starting in the early 1990s. This industry also offers an example of how market competition among Japanese and Chinese manufacturers led to significant technical learning and capability building for local firms, despite the Vietnamese Government’s failure to effectively implement policies.

In the (labour intensive and less economy-of-scale sensitive) garment and (more capital intensive and economy-of-scale sensitive) textile industry, despite the fast-growing volume of garment exports, the average profitability was only 5-8 percent and the rate of value addition was about 20 percent in 2010 as Vietnamese firms, largely engaged in Stage 3 – clothing manufacturing, failed to pursue upgrading strategies to move up the value chain due to the sector’s heavy dependence on imported materials (as supply of raw materials and textile production shrank) and production and local abilities in design, branding, and product differentiation are still very limited.

Sources: “Local Value Chain Development in Viet Nam: Motorcycles, Technical Learning and Rents Management”, Christine Ngoc Ngo, JOURNAL OF CONTEMPORARY ASIA, 2017, VOL. 47, NO. 1, 1–26, http://dx.doi.org/10.1080/0 0472336.2016.1214744, and Ngo, Ngoc Thai Hong (2013) Technology adoption in rent seeking economies: the case of Viet Nam. PhD Thesis. SOAS, University of London, http://eprints.soas.ac.uk/17838.

80 Productivity and Competitiveness of Viet Nam’s Enterprises These suppliers are regarded as first-tier suppliers, implying they sell components directly to lead Recently, with Vinfast’s USD3.5 billion project becoming operational, Viet Nam’s motor vehicle firms. In Stage 2, local firms move to produce basic and low-tech components for local and foreign industry for the first time has a local lead firm with domestic-branded cars and electric bikes. assemblers. In Stage 3, local firms become second-tier suppliers. They establish relationships Learning from past experience, it is important for policy-makers to work closely with Vinfast and and supply specific and more complex components to first-tier suppliers. The local second-tier Vietnamese firms to develop local value chains by providing support to improve Vietnamese firms’ suppliers benefit from collaboration with first-tier suppliers and lead firms. At this stage, many technical capabilities, competitiveness and market integration within and across local industries, foreign firms transfer technology and expertise to support technical learning to these local first in the local market and then in the global market. suppliers. More recently, the garment industry experienced a significant increase in VA-to-output ratio (23 Technology adoption and technical learning also speed up, corresponding to the component’s percent in 2011 to more than 35 percent in 2015 – UNIDO database). This can be attributed to: (i) level of complexity. In the final two stages, local firms become first-tier suppliers, supplying efforts of Vietnamese producers in active marketing and sales (Stage 5) largely in the domestic complex parts directly to foreign lead firms (Stage 4) and become the lead firms themselves market, especially in promoting medium- to high-end apparels, a stepping-stone to eventually (Stage 5). As lead firms, they produce Vietnamese-branded products. From a developing country’s penetrating the international market, (ii) some improvement in the textile industry (Stages 1 and perspective, Stages 4 and 5 indicate high levels of technological upgrading, especially when 2 in garment and textile value chain in Table 2.9) resulting in less dependency by the garment local firms become capable of designing, manufacturing, and branding final products using local industry on imported materials and (iii) some technical process and product upgrading undertaken components and labour. This local value chain does not distinguish whether local firms supply for by garment firms (while functional upgrading, moving up value chains remains modest: out of local industries or foreign buyers. 108 firms surveyed in 2018 only one had functionally upgraded) source: Economic Upgrading in global value chains – opportunity for social upgrading: a case study of Viet Nam’s apparel In the motorcycle industry, during 1995-2000, Vietnamese firms mainly engaged in the first stage, and electronics sectors – CAF/VASS and JustJobs network, November 2018 (source: CAF-VASS and Viet Nam failed to achieve significant technology transfers. Although there were numerous 2018). It is noted, however, that: (i) the garment sub-sector’s LP remains low and slowly catching local firms (mainly SOEs that entered into joint ventures with Japanese firms) engaged in the comparator countries, (ii) despite improvements, the textile sub-sector has negative net exports production of replacement parts, these firms were largely outside the procurement networks of (with a high level of (Viet Nam-based) FDI participation and low VA-to-output ratio) and (iii) the foreign lead firms. As the result of “China shock” (cheap Chinese motorcycles penetrating and garment sub-sector still heavily depends on imported materials and firms are still largely engaged capturing the low-end Vietnamese market), local content and movement of (domestic and foreign in Stage 3 (manufacturing cloth using designs and materials supplied by foreign partners). This Viet Nam-based firms) firms to Stages 2, 3 and 4 accelerated. As a result: prior to 2005, this industry indicates more effort is needed to accelerate Vietnamese firms into higher stages of the local accounted for 3.1 percent of the total industrial production value of the country, which grew to 23.9 value chain, such as in supplying textile materials and developing and marketing Vietnamese percent by 2007 and starting in 2010, the major foreign motorcycle-makers in Viet Nam started to brands for local and export markets (extremely difficult due to Vietnamese firms’ ability to design export production surplus to other markets (in 2011, Honda exported 300,000 motorcycles to the and brand remain limited). Philippines, Cambodia, Laos and Afghanistan) and by 2012 the local content ratio, measuring the percentage of domestic parts in each motorcycle, was between 70 and 95 percent (for instance, Honda’s local content ratio reached nearly 95 percent on some of its models). The motorcycle • Food and beverages, and furniture industry provides an exemplary case study of industrial upgrading predominantly driven by FDI starting in the early 1990s. This industry also offers an example of how market competition among (Low-tech) furniture, food processing and (medium-tech) beverages are the fourth and fifth largest of Japanese and Chinese manufacturers led to significant technical learning and capability building Viet Nam’s manufacturing export sub-sectors, noting that food processing contributes most to “food for local firms, despite the Vietnamese Government’s failure to effectively implement policies. and beverage” exports (while beverages mainly serve the domestic market), with RCA values larger than unity at 1.44 and 1.28, respectively. Food processing ranked second in revenue share, third in VA and In the (labour intensive and less economy-of-scale sensitive) garment and (more capital intensive fourth in employment, while beverages (a medium-sized sub-sector) had a low rank in employment and economy-of-scale sensitive) textile industry, despite the fast-growing volume of garment (fifth from bottom), revenue and VA (10th from bottom). Furniture is a medium-sized manufacturing sub- exports, the average profitability was only 5-8 percent and the rate of value addition was about 20 sector with a lower-middle ranking in revenue and VA shares, while fifth in employment. percent in 2010 as Vietnamese firms, largely engaged in Stage 3 – clothing manufacturing, failed to pursue upgrading strategies to move up the value chain due to the sector’s heavy dependence FDI firms dominated in beverages and furniture with VA shares of 57 and 58.3 percent respectively, on imported materials (as supply of raw materials and textile production shrank) and production (noting that recently a foreign company took a controlling stake from an SOE in Sabeco, Viet Nam’s and local abilities in design, branding, and product differentiation are still very limited. largest beverage company. The new board decided to remove the 49 percent cap on foreigners’ share of company capital, hence FDI domination in beverages will increase further). Notably, backward linkages Sources: “Local Value Chain Development in Viet Nam: Motorcycles, Technical Learning and Rents Management”, Christine Ngoc Ngo, JOURNAL OF CONTEMPORARY ASIA, 2017, VOL. 47, NO. 1, 1–26, http://dx.doi.org/10.1080/0 between FDI and domestic firms are very low in furniture (12.5 percent), beverages (10.4 percent) and 0472336.2016.1214744, and Ngo, Ngoc Thai Hong (2013) Technology adoption in rent seeking economies: the food processing (2 percent). case of Viet Nam. PhD Thesis. SOAS, University of London, http://eprints.soas.ac.uk/17838. While SOEs have low participation levels, domestic private firms have medium VA and high employment shares in beverages and rather high VA, employment and revenue shares in furniture. Food processing was the only large-sized and large positive net export sub-sector in which domestic firms dominated with a VA share of 62.2 percent (FDI participation was at medium level).

80 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 81 Labour productivity of food processing was at medium level, while high in beverages and low in furniture. The LP gaps with comparator countries were: (i) medium and narrowing in food processing, (ii) small and narrowing in beverages and (iii) large and slowly narrowing in furniture. VA-to-output ratios were low and gaps with comparator countries were large in food processing and beverages, while furniture featured a medium ratio and narrower gap. Wage growth was lower than LP growth in food processing, but higher (though remaining competitiveness enhancing) in beverages and furniture.

Overall, Viet Nam’s food processing sub-sector was assessed as “competitive” based on performances measured by indicators used by this report and especially due to the comparative advantages (and good performance) of Viet Nam’s agriculture production in many commodities (cashews, catfish, coffee, pepper, rice and shrimp). Domestic private firms (mostly SMEs and applying labour intensive and simple labour skills) dominating this sub-sector were export-oriented and aggressively expanded (due to recent fast production growth and exports of some agriculture produce such as tropical fruits and vegetables) their shares in global markets and diversified export markets to move up GVCs. Accelerating capacity in branding, marketing and playing lead roles in GVCs (given Viet Nam is a big global player in many agriculture products) and especially in local value chains (by increasing the economy of scale for higher efficiency of agriculture production and ensuring international standards in quality, food safety and environment, promoting organic farming/green production methods and application of IR4.0 technologies) will be key to enhancing this sub-sector’s productivity and competitiveness.

Box 2.9: Deeper look at food processing industries’ labour productivity

Figure 2.46: Labour productivity of food processing industries

% 25

20 Oil & fat

Meat Flour Vegie 10

-100 0 100 200 300 400 500 600 700 800 Other food Dairy Mil. VND Aqua 5

Note: Horizontal axis – value of labour productivity in 2011, vertical axis – growth rate of labour productivity in 2011-2016, bubble size – number of labour involved.

Source: Authors’ calculation using Enterprise Census data

82 Productivity and Competitiveness of Viet Nam’s Enterprises Labour productivity of food processing was at medium level, while high in beverages and low in furniture. The LP gaps with comparator countries were: (i) medium and narrowing in food processing, (ii) small and Within the food processing sub-sector, production of oil, animal fat, meat and vegetables narrowing in beverages and (iii) large and slowly narrowing in furniture. VA-to-output ratios were low and experienced higher than the average LP growth, but industries were modest in employment and gaps with comparator countries were large in food processing and beverages, while furniture featured a LP of vegetables and meat processing were low. Milk and dairy products had exceptionally high LP, medium ratio and narrower gap. Wage growth was lower than LP growth in food processing, but higher though with modest growth and employment. Processing and preservation of aquatic products (though remaining competitiveness enhancing) in beverages and furniture. experienced relatively lower LP growth and LP levels were lower, but (given the predominant use of labour intensive technologies in the aquatic product processing industry) the industry Overall, Viet Nam’s food processing sub-sector was assessed as “competitive” based on performances was the biggest employer. Noting the important contribution of aquatic products (and recently measured by indicators used by this report and especially due to the comparative advantages (and vegetables and fruits) to Viet Nam’s exports and job creation improving its competitiveness and good performance) of Viet Nam’s agriculture production in many commodities (cashews, catfish, productivity performance would contribute to broad-based growth. In general, given important coffee, pepper, rice and shrimp). Domestic private firms (mostly SMEs and applying labour intensive and roles in improving value added (and enhancing food safety) of agricultural products, leading simple labour skills) dominating this sub-sector were export-oriented and aggressively expanded (due the local value chains and connecting Viet Nam’s farmers to GVCs, food processing industries to recent fast production growth and exports of some agriculture produce such as tropical fruits and (with high participation levels of domestic private firms) should be a prioritized sub-sector for vegetables) their shares in global markets and diversified export markets to move up GVCs. Accelerating policy interventions to enhance productivity and competitiveness. A more detailed assessment capacity in branding, marketing and playing lead roles in GVCs (given Viet Nam is a big global player in of current technologies and foresight of IR4.0 technology progress is needed to guide domestic many agriculture products) and especially in local value chains (by increasing the economy of scale firms for technology upgrading. for higher efficiency of agriculture production and ensuring international standards in quality, food safety and environment, promoting organic farming/green production methods and application of IR4.0 technologies) will be key to enhancing this sub-sector’s productivity and competitiveness. Furniture and beverage sub-sectors were assessed as “competitive” with risks. The furniture sub-sector’s low LP (as production is based on intensive labour and simple skills) and widening/slowly narrowing LP gaps with comparator countries and dependency on imports of wood (while CPTPP requires higher Box 2.9: Deeper look at food processing industries’ labour productivity export content from origin countries) present key risks. FDI domination in the sub-sector (while domestic firms and SOEs only have low levels of participation), negative (though still competitiveness enhancing) Figure 2.46: Labour productivity of food processing industries wage-labour productivity growth and the risk of automation swallowing up simple and repetitive jobs heighten the likelihood of FDI relocating production elsewhere, resulting in significant impacts to Viet % 25 Nam’s exports and socio-economic development.

20 In the beverage sub-sector, which mainly serves the domestic market while FDI (with its global brands) Oil & fat increases its participation level (also in the retail sub-sector), smaller-sized and fragmented domestic 15 firms will face stronger competition in the near future. It is important to note that this sub-sector’s FDI

Meat Flour attraction policy has been very “open”: anecdotally Viet Nam has a higher number of beer brands than Vegie 10 many countries and this has contributed to a fragmented market (small-sized firms have slimmer market

-100 0 100 200 300 400 500 600 700 800 shares than nations with less brands). Within this context, some domestic firms’ efforts to sharpen their Other food Dairy Mil. VND Aqua 5 competitive edge, develop brands and new products is remarkable and should be recognized. While demand from wealthier Vietnamese will grow, public health concerns related to consumption of alcohol36

0 and sugar-rich drinks may eat into demand. This together with stronger competition from FDI firms will demand domestic firms make extra efforts, including increasing linkages with domestic suppliers to Note: Horizontal axis – value of labour productivity in 2011, vertical axis – growth rate of labour productivity in enhance productivity and competitiveness in the domestic market in the next five years, greater efforts 2011-2016, bubble size – number of labour involved. to penetrate international markets. Source: Authors’ calculation using Enterprise Census data Other export contributing sub-sectors

Though not top-five manufacturing exporters, the net exports of wood (excluding furniture), tobacco, other vehicles and non-metallic mineral products was positive.

Other vehicles, the only high-tech (medium-sized in terms of employment, revenue and VA) sub- sector in this group was dominated by FDI firms with a VA share of 82.21 percent and (similar to sub- sectors where FDI firms dominate) almost zero backward linkages and low (17 percent) forward linkages

36 According to the Minister of Health, in a presentation of the draft Law on Prevention of Negative Impacts of Alcohol Drinks to the National Assembly on 9 November 2018, Vietnamese people in 2017 consumed around 305 million litres of liquor and 4.1 billion litres of beer (with spending on beer of USD4 billion). An average Vietnamese consumes 42 litres of beer annually, making Viet Nam the highest beer consumer in ASEAN and third in Asia (after Japan and China). In Viet Nam, the estimated economic cost of six cancers to which alcohol consumption is among the leading causes was 0.25 percent of GDP and the cost of traffic accidents caused by alcohol consumption was 0.5 percent GDP in 2017.

82 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 83 with domestic firms. Labour productivity was high compared to other manufacturing sub-sectors in Viet Nam and “within reach” of comparator countries. With a medium-level VA-to-output ratio, gaps with comparator countries have narrowed. Notably within the sub-sector, the motorcycle industry experienced significant improvements in “local” value chains and very high levels of “domestic” content (not necessarily produced by Vietnamese firms, but Viet Nam-based FDI ones). The strong penetration of “cheap” Chinese motorcycles into Viet Nam during the 1990s could be a key reason for other foreign (mainly Japanese) firms to increase “domestic” content as a successful strategy to remain competitive and expand export production (Box 2.8). The sub-sector, mainly motorcycles (as well as production of bicycles/parts), was assessed as “competitive” with scope for enhancement and stronger linkages between FDI and Vietnamese firms to lift domestic content and allow the latter to progressively move up to stages (2, 3 and 4) in the local value chain. With Vinfast’s car and electric bicycle plant in Dinh Vu – Cat Hai Economic Zone having become operational in late 2018, Viet Nam for the first time has a leading firm to produce branded electric bicycles (Stage 5 in the local value chain). Vinfast’s strong linkages with domestic suppliers and robust domestic market competition to build export capacity will be the key to Vinfast’s and the local electric bicycle industry’s success.

Other sub-sectors in this group: wood (excluding furniture), printing and tobacco (small-sized and low- tech) and non-metallic mineral products (large-sized and medium-tech) are dominated by domestic private firms (tobacco by SOEs). The RCA of wood (excluding furniture) and non-metallic mineral products were higher than unity (1.86 and 1.07, respectively), while RCA for the remainder was less than unity. Gaps between Viet Nam and comparator countries in VA-to-output ratios of wood (excluding furniture) and printing narrowed, while remained large in tobacco and non-metallic mineral products. Labour productivity gaps between Viet Nam and comparator countries were high and slowly narrowing. These sub-sectors were assessed as having “low competitiveness”. Within these sub-sectors, given they are dominated by domestic firms and large domestic markets, scope for policy interventions to enhance productivity and competitiveness remain significant. Applying stricter permission of foreign brands, control of smuggling, marketing as well as a higher tax on tobacco (similar to alcoholic drinks in beverage sub-sector) may be helpful. Tailored support to build stronger linkages between wood (excluding furniture) and local firms in the oil refinery and chemicals sub-sectors as well as suppliers of bamboo, rattan and other related materials available in Viet Nam and for sub-sector firms to upgrade technologies, branding and marketing capabilities must be explored.

Sub-sectors with negative net exports

The last group of sub-sectors had high and medium technology with negative net exports and RCA<1 (classified as import-substitution), namely: basic metals, fabricated metals, rubber-plastics (large- sized sub-sectors), chemicals, electrical equipment, motor vehicles (medium-sized sub-sectors), other manufacturing, paper products, machines and equipment n.e.c., coke- petroleum-nuclear fuel, repair and installation of machinery and equipment, and medical precision-optical equipment (small-sized sub-sectors).

In 2016, FDI firms had high VA shares in almost all sub-sectors in this group: basic metals (83.6 percent), electrical equipment (62.4 percent) and motor vehicles (73.7 percent), other manufacturing (82.7 percent), paper products (57.7 percent), machines and equipment n.e.c. (69.15 percent), chemicals (49.4 percent, domestic private firms’ VA share was 32.5 percent) and fabricated metals (49.52 percent, domestic private firms’ VA share was comparable: 48.57 percent). Besides the substantial VA share in chemicals and fabricated metal, domestic private firms also have significant VA shares in rubber-plastics (50.82 percent, while FDI’s VA share was 46.63 percent) and medical precision-optical equipment (74.53 percent) in 2016. SOEs had medium employment and high VA shares in coke-petroleum-nuclear fuel and repair and installation of machinery and equipment. As noted earlier, FDI in resource-based industries tended to have higher backward-forward linkages with domestic firms, with backward linkages in basic metals and chemicals at 96 percent and 62 percent, respectively and backward-forward linkages in rubber and plastics at 25 percent and 24 percent respectively. Linkages in other sub-sectors were small

84 Productivity and Competitiveness of Viet Nam’s Enterprises with domestic firms. Labour productivity was high compared to other manufacturing sub-sectors in across 2011-2015. Viet Nam and “within reach” of comparator countries. With a medium-level VA-to-output ratio, gaps with comparator countries have narrowed. Notably within the sub-sector, the motorcycle industry Basic metals, paper, chemicals, coke-petroleum-nuclear fuel, motor vehicles sub-sectors have high LP experienced significant improvements in “local” value chains and very high levels of “domestic” content with narrowing gaps with comparator countries. However, the performance measured by VA-to-output (not necessarily produced by Vietnamese firms, but Viet Nam-based FDI ones). The strong penetration revenue showed only motor vehicles had a high ratio and declining gaps with comparator countries, while of “cheap” Chinese motorcycles into Viet Nam during the 1990s could be a key reason for other foreign paper, coke-petroleum-nuclear fuel, rubber-plastics, chemicals and basic metals have low/medium VA- (mainly Japanese) firms to increase “domestic” content as a successful strategy to remain competitive to-output ratios and large gaps to other countries. and expand export production (Box 2.8). The sub-sector, mainly motorcycles (as well as production Motor vehicle sub-sector was assessed as “competitive” as long as FDI car-assembling plants in Viet of bicycles/parts), was assessed as “competitive” with scope for enhancement and stronger linkages Nam remained competitive. The key risk involved trade agreements (including ASEAN) rewarding tax between FDI and Vietnamese firms to lift domestic content and allow the latter to progressively move incentives for products with higher (20 percent and more) local content and fluid movement of goods up to stages (2, 3 and 4) in the local value chain. With Vinfast’s car and electric bicycle plant in Dinh Vu – that may encourage FDI firms to relocate car-assembling plants abroad if Vietnamese firms fail to supply Cat Hai Economic Zone having become operational in late 2018, Viet Nam for the first time has a leading leading FDI firms. As noted, with Vinfast operational in late 2018 the development of local value chains firm to produce branded electric bicycles (Stage 5 in the local value chain). Vinfast’s strong linkages with and its ability to compete in local markets will raise the sub-sector’s productivity and competitiveness. domestic suppliers and robust domestic market competition to build export capacity will be the key to Vinfast’s and the local electric bicycle industry’s success. Basic metals, paper, rubber-plastics, chemicals, coke-petroleum-nuclear fuel sub-sectors were assessed as having “medium competitiveness” if the supply of local resource-based (as well as electricity supplies Other sub-sectors in this group: wood (excluding furniture), printing and tobacco (small-sized and low- for energy-intensive basic metal sub-sector) inputs continued. Key challenges included: (i) accelerating tech) and non-metallic mineral products (large-sized and medium-tech) are dominated by domestic LP and VA-to-revenue ratios, (ii) improving logistics and linkages with other sub-sectors in local value private firms (tobacco by SOEs). The RCA of wood (excluding furniture) and non-metallic mineral chains and (iii) applying higher environmental standards and changing competition rules towards products were higher than unity (1.86 and 1.07, respectively), while RCA for the remainder was less than application of more environment-friendly technologies. Moving up value chains in plastics (with oil unity. Gaps between Viet Nam and comparator countries in VA-to-output ratios of wood (excluding refineries supplying the plastics industry and fibres for textile industry) and the rubber industry moving furniture) and printing narrowed, while remained large in tobacco and non-metallic mineral products. away from exporting rubber raw materials to producing high-quality plastics and rubber products for Labour productivity gaps between Viet Nam and comparator countries were high and slowly narrowing. other sub-sectors (cars, bikes and motorcycles and electronics or high-quality rubber-plastics products These sub-sectors were assessed as having “low competitiveness”. Within these sub-sectors, given for healthcare) is necessary for the rubber industry to become more competitive. Notably, within the they are dominated by domestic firms and large domestic markets, scope for policy interventions to chemicals sub-sector, some local firms producing washing detergent and toothpaste have maintained enhance productivity and competitiveness remain significant. Applying stricter permission of foreign brands and competed with FDI giants such as P&G and Unilever (especially in markets for lower-end brands, control of smuggling, marketing as well as a higher tax on tobacco (similar to alcoholic drinks products)37 in Viet Nam. Also within chemicals, Viet Nam’s pharmaceutical industry is fragmented with in beverage sub-sector) may be helpful. Tailored support to build stronger linkages between wood a high number of SMEs operating under management of local governments and decentralized drug- (excluding furniture) and local firms in the oil refinery and chemicals sub-sectors as well as suppliers procurement systems. A more detailed case study on these chemicals and pharmaceutical industries of bamboo, rattan and other related materials available in Viet Nam and for sub-sector firms to upgrade may be needed to identify tailored support to enhance productivity and competitiveness of local firms. technologies, branding and marketing capabilities must be explored. Fabricated metal, electrical equipment, other manufacturing, machinery and equipment n.e.c., medical- Sub-sectors with negative net exports precision equipment and installation of machinery have medium or low LP and VA-to-output ratios, with The last group of sub-sectors had high and medium technology with negative net exports and RCA<1 large LP and VA-to-output ratio gaps with comparator countries. These sub-sectors were assessed (classified as import-substitution), namely: basic metals, fabricated metals, rubber-plastics (large- as “low competitive”. More in-depth analysis at ISIC/VSIC 3- and 4-digit level is required to define sized sub-sectors), chemicals, electrical equipment, motor vehicles (medium-sized sub-sectors), other opportunities and challenges of industries within these sub-sectors. manufacturing, paper products, machines and equipment n.e.c., coke- petroleum-nuclear fuel, repair Recommendations and installation of machinery and equipment, and medical precision-optical equipment (small-sized sub-sectors). • Enhancing Vietnamese firms’ productivity and competitiveness must be at the heart of Viet Nam’s growth strategy and achieved through implementing a wide range of integrated and concerted In 2016, FDI firms had high VA shares in almost all sub-sectors in this group: basic metals (83.6 percent), policy actions electrical equipment (62.4 percent) and motor vehicles (73.7 percent), other manufacturing (82.7 percent), paper products (57.7 percent), machines and equipment n.e.c. (69.15 percent), chemicals As lower middle-income Viet Nam, in its next stage of development, pursues an inclusive growth pathway (49.4 percent, domestic private firms’ VA share was 32.5 percent) and fabricated metals (49.52 percent, to generate more productive jobs for its workers, enhance productivity and competitiveness, Vietnamese domestic private firms’ VA share was comparable: 48.57 percent). Besides the substantial VA share in firms in manufacturing must be at the centre of Viet Nam’s growth strategy. This should be a common chemicals and fabricated metal, domestic private firms also have significant VA shares in rubber-plastics prioritized goal within a wide range of national policies, from industrialization, SOE reforms and private (50.82 percent, while FDI’s VA share was 46.63 percent) and medical precision-optical equipment (74.53 sector/enterprise development to trade, FDI attraction, R&D, skills training and public investment. percent) in 2016. SOEs had medium employment and high VA shares in coke-petroleum-nuclear fuel and Learning from the past and international experiences, these policies must be formulated and implemented repair and installation of machinery and equipment. As noted earlier, FDI in resource-based industries in an integrated and coordinated manner with optimum synergies to achieve this common goal. Clear tended to have higher backward-forward linkages with domestic firms, with backward linkages in basic metals and chemicals at 96 percent and 62 percent, respectively and backward-forward linkages in 37 Given the low-tech and import substitution nature of production of these commodities, many countries often apply measures rubber and plastics at 25 percent and 24 percent respectively. Linkages in other sub-sectors were small to protect local firms. As FDI companies already operate in Viet Nam, support could be provided to local firms (especially SMEs) in R&D, employee training, branding and marketing (still allowed by trade agreements) as well as eliminate price transfers by FDI firms.

84 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 85 targets within these policies must be aligned towards achieving improved capacity of domestic firms in terms of: (i) linkages and upward movements in local and global value chains, (ii) increased productivity, value addition and (iii) local market shares and especially export volumes and values.

• Trade policies and domestic enterprise development

The commonly understood weakness of limited linkages between trade negotiations, industry policies and programmes supporting enterprise development require urgent remedial action. The assessments of weaknesses and potential of Vietnamese firms could underpin (i) industrial policies and programmes building firms’ capacity and competitiveness and (ii) international trade negotiations. New opportunities for Vietnamese firms created by trade negotiations/agreements, such as access to new markets with lower tariffs (raising demand for products and services with important incentives for firms to expand production capacity and competitiveness), legitimate protection and time for firms to become compliant with new trade rules, must be accompanied with appropriate policies and action to help firms build production capacity, connect with local and global value chains and better compete in domestic and international markets.

• SOE reforms, public investment and domestic enterprise development

SOE reform efforts must focus on: (i) enhancing SOEs’ (especially those at higher stages or leading local value chains with relatively high productivity and competitiveness) effectiveness and efficiencies, plus backward-forward linkages with domestic firms and (ii) accelerated equitization of SOEs in industries/ sub-sectors where domestic private firms have capacity and readiness to take over leading roles from SOEs (with equitization going hand-in-hand with building capacity of domestic private firms).

Public investment should aim to crowd-in domestic private sector investment. Public investment could help create demand for domestic private firms’ products and services, and thus incentives to invest in business and technology upgrades. Public investment could also provide domestic private firms with opportunities to build capacity, including through learning-by-doing and receiving technology transfers through public investment from ODA loan projects.

Public investment in development of IT/telecommunications infrastructure (cloud computing, network and data security as well as e-commerce platforms, including for intermediate goods), e-payments and e-banking (similar to e-tax, e-customs and e-government payments), will not only help private firms (especially SMEs) improve IR4.0 readiness and efficiency, but also value chain connections. While public services (R&D and skills training) will benefit SMEs, special services such as testing and certification (and perhaps radiation and cold storage) have strong potential to enhance access and competitiveness of food and agriculture processing firms in global markets.

• Attracting higher-quality FDI and enhancing linkages between FDI and domestic firms

Within the overarching goals of sustainable development and creation of productive jobs for Vietnamese workers, Viet Nam must shift its focus from quantity to attraction of higher-quality FDI firms with technological sophistication, that are green, energy efficient and leaders in GVCs. The most important element of higher-quality FDI firms is long-term partnerships with local firms (as key players in production chains) as the core of their international competition strategy. Following this strategy, FDI firms must go beyond exploiting natural resources and cheap labour to establish strong and long-term linkages with domestic firms by using higher levels of local content (less expensive than imported components for simple assembling) and utilization of local workers’ skills and innovations in products. Such FDI firms, in “win-win” cooperation with government programmes to support capacity development of domestic firms, would be better positioned to engage in technology transfers with domestic firms and connect with them as first-tier suppliers to leading FDI firms and GVCs.

To attract higher-quality FDI, attraction policies must be formulated and implemented as an integral part of Viet Nam’s socio-economic development strategy, industrial policies, economic restructuring,

86 Productivity and Competitiveness of Viet Nam’s Enterprises targets within these policies must be aligned towards achieving improved capacity of domestic firms in accelerated domestic private sector development and the country’s and its firms’ IR4.0 readiness. Within terms of: (i) linkages and upward movements in local and global value chains, (ii) increased productivity, this integrated framework, it is important to establish clear international standards and requirements value addition and (iii) local market shares and especially export volumes and values. in terms of technology requirements, domestic content, technology transfers to domestic firms and linkages to GVCs, together with compliance requirements for stricter energy efficiency and environmental • Trade policies and domestic enterprise development safety standards, work conditions and social protection. Strengthening the legal framework, institutional The commonly understood weakness of limited linkages between trade negotiations, industry policies capacity and systems for rigorous screening, appraising and approval of FDI projects is important to and programmes supporting enterprise development require urgent remedial action. The assessments ensure adherence to international standards and requirements. of weaknesses and potential of Vietnamese firms could underpin (i) industrial policies and programmes It is also essential for Viet Nam to limit harmful competition between provinces and nationally in use building firms’ capacity and competitiveness and (ii) international trade negotiations. New opportunities of tax and other incentives to attract FDI38. Instead, Viet Nam should focus on creating other (more for Vietnamese firms created by trade negotiations/agreements, such as access to new markets with fundamental) incentives to capture high-quality and long-term FDI, namely: high capacity and skilled lower tariffs (raising demand for products and services with important incentives for firms to expand human resources, large population purchasing power, investment regulation stability, consistent production capacity and competitiveness), legitimate protection and time for firms to become compliant application of the rule of law, political stability, quality infrastructure (notably transportation and utilities) with new trade rules, must be accompanied with appropriate policies and action to help firms build and competitive domestic support services and supplies. Government and FDI firms, those planning to production capacity, connect with local and global value chains and better compete in domestic and locate and present in Viet Nam, should work together with domestic firms to: (i) define domestic support international markets. services and supplies for FDI to underpin international competition strategies and (ii) formulate and • SOE reforms, public investment and domestic enterprise development implement “win-win” plans to build domestic firms’ capacity to become suppliers in (FDI-led) GVCs. Such a mutually beneficial approach could be applied in high and medium technology and export-led sub- SOE reform efforts must focus on: (i) enhancing SOEs’ (especially those at higher stages or leading local sectors/industries where FDI firms have high participation levels (motorcycles, electronics, furniture value chains with relatively high productivity and competitiveness) effectiveness and efficiencies, plus and non-metallic mineral products). It could also be applied in large domestic market-focussed sub- backward-forward linkages with domestic firms and (ii) accelerated equitization of SOEs in industries/ sectors with high FDI participation levels (rubber-plastics, electrical equipment, fabricated metals and sub-sectors where domestic private firms have capacity and readiness to take over leading roles from textiles) for improved linkages between FDI and domestic firms in local value chains to mutually enhance SOEs (with equitization going hand-in-hand with building capacity of domestic private firms). productivity, competitiveness and export production.

Public investment should aim to crowd-in domestic private sector investment. Public investment could • Accelerate development of local value chains and capacity of domestic private firms to achieve help create demand for domestic private firms’ products and services, and thus incentives to invest in upward movement in value chains business and technology upgrades. Public investment could also provide domestic private firms with opportunities to build capacity, including through learning-by-doing and receiving technology transfers Given that domestic private investment in Viet Nam is low compared to other sources (FDI and public through public investment from ODA loan projects. investment) and ASEAN countries (source: UNDP Development Finance Assessment report, UNDP 2018) as well as policy-makers’ expectation the private sector will be a “growth engine” in the country’s Public investment in development of IT/telecommunications infrastructure (cloud computing, network next development stage, domestic sector development must be prioritized in the Socio-Economic and data security as well as e-commerce platforms, including for intermediate goods), e-payments and Development Strategy (2021-2030) and Plan (2021-2025). Key policy objectives should support e-banking (similar to e-tax, e-customs and e-government payments), will not only help private firms domestic private firms to grow in size, accelerate transition to formalization and enhance productivity (especially SMEs) improve IR4.0 readiness and efficiency, but also value chain connections. While public and competitiveness through development of local value chains, improved linkages within and upward services (R&D and skills training) will benefit SMEs, special services such as testing and certification movement in domestic and global value chains. (and perhaps radiation and cold storage) have strong potential to enhance access and competitiveness of food and agriculture processing firms in global markets. It is important to continue creating a level playing field for the domestic private sector by removing obstacles (access to credit, land or tax reductions and exemptions) for it to compete on an equal footing • Attracting higher-quality FDI and enhancing linkages between FDI and domestic firms with SOEs and FDI enterprises. As noted, reforms to SOEs and FDI policies should facilitate private firms to enter markets and enhance linkages in domestic and global value chains, create “win-win” policies Within the overarching goals of sustainable development and creation of productive jobs for Vietnamese and incentives for FDI enterprises to generate more spillovers in terms of connecting domestic firms to workers, Viet Nam must shift its focus from quantity to attraction of higher-quality FDI firms with GVCs and achieving technology transfers. technological sophistication, that are green, energy efficient and leaders in GVCs. The most important element of higher-quality FDI firms is long-term partnerships with local firms (as key players in production In addition to continued efforts to nuance the business environment and support domestic private chains) as the core of their international competition strategy. Following this strategy, FDI firms must go enterprises to access land and credit, more tailored support is needed for SMEs to elevate their: (a) capacity beyond exploiting natural resources and cheap labour to establish strong and long-term linkages with for business management and marketing, (b) linkages in domestic and international value chains and (c) domestic firms by using higher levels of local content (less expensive than imported components for technical capacity to adopt new technologies and readiness to grasp opportunities unlocked by IR4.0 simple assembling) and utilization of local workers’ skills and innovations in products. Such FDI firms, and a new wave of higher quality FDI. Establishment of independent (para-governmental) institutions in “win-win” cooperation with government programmes to support capacity development of domestic (similar to Fraunhofer Foundation, Germany) specialized in providing training and R&D support to firms, would be better positioned to engage in technology transfers with domestic firms and connect Vietnamese firms (impeded by their small-sizes, limited linkages in value chains and inability to afford with them as first-tier suppliers to leading FDI firms and GVCs. 38 Given the strong competition between countries attracting FDI, it is important Viet Nam strengthens its active participation To attract higher-quality FDI, attraction policies must be formulated and implemented as an integral in international initiatives (such as Tax Inspectors Without Borders and Base Erosion and Profit Shifting Inclusive Framework) that part of Viet Nam’s socio-economic development strategy, industrial policies, economic restructuring, provide opportunities for countries to jointly develop and apply codes of conduct to address harmful tax practices related to attracting FDI as well as “price transfers” and other tax avoidance practices.

86 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 87 investment in R&D and training) is necessary. As domestic private firms improve integration and upward movement in local and global value chains, technology upgrades must follow. Besides access to credit, guidance for technology upgrades (based on assessment of sub-sector/firm level current technologies and foresight of future IR4.0 technological progress) will be necessary for domestic firms (especially SMEs) to make appropriate investment decisions.

The emergence of big domestic firms to lead local value chains and become significant players in GVCs must be facilitated. Domestic food processing firms’ growth based on Viet Nam’s comparative advantages in producing key agriculture commodities for export (catfish, coffee, pepper and shrimp) not only adds value and contributes to exports, they lead development of local value chains (connecting micro, small, medium-sized farming units in value chains, increasing application of quality/food safety standards, facilitating branding and marketing, expanding access to international markets). Learning from the past, further action is needed to accelerate development of more (newly-emerging agriculture export commodities such as fruit and vegetables) as well as bigger and longer value chains in food and food processing – the fifth largest exporting manufacturing sub-sector (with large VA, employment and revenue shares) that domestic (private) firms dominate. The government’s facilitation, incentives (including development of e-commerce platforms for sale of agriculture, intermediate products and connecting value chain players) and especially support to (small) farmers is needed to: (i) apply production practices that meet standards of food processing/export firms leading local value chains and (ii) organize farmers in cooperatives or other formal forms to gain greater efficiency from higher economies of scale.

The recent entry of Vingroup into Viet Nam’s automobile and electric bike industry and a planned smartphone production project present an encouraging case for developing local value chains. For the first time in these industries, there is a Vietnamese firm at the highest value chain stage to be an aspirational leader. Similar to facilitating FDI and domestic firm linkages, with support of the Government of Viet Nam, Vingroup should work with domestic firms to: (i) define domestic support services and suppliers for Vinfast according to its (international) competition strategy and (ii) formulate and implement “win-win” plans to build capacity of domestic firms to become suppliers in Vingroup-led value chains.

88 Productivity and Competitiveness of Viet Nam’s Enterprises investment in R&D and training) is necessary. As domestic private firms improve integration and upward 5. Conclusions movement in local and global value chains, technology upgrades must follow. Besides access to credit, guidance for technology upgrades (based on assessment of sub-sector/firm level current technologies Viet Nam is entering its next development stage as a lower middle-income country in the context of and foresight of future IR4.0 technological progress) will be necessary for domestic firms (especially IR4.0’s acceleration, significant changes in GVCs and international trade. Embarking on a more inclusive SMEs) to make appropriate investment decisions. development pathway and a new growth model based on higher productivity is essential for Viet Nam to create greater volumes of decent work for its people and achievement of the SDGs. The emergence of big domestic firms to lead local value chains and become significant players in GVCs must be facilitated. Domestic food processing firms’ growth based on Viet Nam’s comparative The assessment of productivity and competitiveness performance of the manufacturing sector and its advantages in producing key agriculture commodities for export (catfish, coffee, pepper and shrimp) sub-sectors presented in this report identified challenges and opportunities for Viet Nam to greatly not only adds value and contributes to exports, they lead development of local value chains (connecting enhance domestic firms’ competitive edge through capturing more value added from a bigger local and micro, small, medium-sized farming units in value chains, increasing application of quality/food safety global value chain footprint. With the recommended enhancement of domestic firms’ productivity and standards, facilitating branding and marketing, expanding access to international markets). Learning competitiveness as a central tenet in Viet Nam’s growth strategy for its next development stage, the from the past, further action is needed to accelerate development of more (newly-emerging agriculture interlinked policy actions outlined in this section of the report should be formulated and implemented export commodities such as fruit and vegetables) as well as bigger and longer value chains in food within an integrated policy framework with concerted efforts by different stakeholders from government and food processing – the fifth largest exporting manufacturing sub-sector (with large VA, employment and business sectors. and revenue shares) that domestic (private) firms dominate. The government’s facilitation, incentives Now is the time for all stakeholders to act in a concerted and synergized manner on the identified (including development of e-commerce platforms for sale of agriculture, intermediate products challenges to realize Viet Nam’s aspirations to achieve its ambitious SDG agenda and lift human and connecting value chain players) and especially support to (small) farmers is needed to: (i) apply development to new heights. production practices that meet standards of food processing/export firms leading local value chains and (ii) organize farmers in cooperatives or other formal forms to gain greater efficiency from higher economies of scale.

88 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 89 References

To be revised 1. Blomstrom, M., and Kokko, Ari., (1998). Multinational Corporations and Spillovers. Centre for Economic Policy Research, CEPR Discuss Paper: No.1365. 2. Bùi Trinh và Nguyễn Việt Phong (2011), ‘Định hướng lại cơ cấu ngành, vùng’. Thời báo Kinh tế 3. Sài Gòn Online. Trang web: [http://www.thesaigontimes.vn/home/diendan/ykien/68264/]. 4. Business Monitor International. (2010). Viet Nam Autos Report Q4 2010. 5. CIEM (2012), Năng lực cạnh tranh và công nghệ ở cấp độ doanh nghiệp tại Việt Nam. Nhà xuất 6. bản Lao động 7. Chuc, et al. (2008), FDI Horizontal and Vertical Effects on Local Firm Technical Efficiency . Working Paper, No.17. DEPOCEN. 8. DBS Research Group. (2015). Viet Nam: Asia’s latest electronics spark. DBS Research Group. 9. GSO (2017), Điều tra lao động việc làm hàng năm. 10. Hollweg et al., (2017), Viet Nam at a Crossroads - Engaging in the Next Generation of Global Value Chains. World Bank Group. 11. ILO (2016), Wage and productivity in the garment sector in Asia and the Pacific and the Arab States. 12. MOIT (2016). Studying the Proposal to Develop Supporting Industries via Promoting Linkage and Supply to Multinational Corporations Operating in the Electronics Sector of Viet Nam. 13. Nguyen, T. B. (2008). The Vietnamese Automobile Industry and Globalization. The Fourth VDF-Tokyo Conference on the development of Viet Nam, (pp. 197-214). Tokyo. 14. Porter, Michael E. (2010), Báo cáo năng lực cạnh tranh Việt Nam 2010. CIEM và Lee Kuan Yew School of Public Policy. 15. Takii, S. and D. A. Narjoko (2012), ‘FDI Forward Linkage Effect and Local Input Procurement- Evidence from Indonesian Manufacturing-’, in Hahn, C. H. And D. A. Narjoko (eds.), Dynamics of Firm Selection Process in Globalized Economies. ERIA Research Project 16. Tran, D.-H., & Ngo, D.-T. (2014). Performance of the Vietnamese Automobile Industry: A Measurement using DEA. Asian Journal of Business and Management. 17. Report 2011 no.3, pp.111-146. 18. Trần Văn Thọ (2005), Biến động kinh tế Đông Á và con đường công nghiệp hóa Việt Nam, Nhà xuất bản Chính trị Quốc gia, Hà Nội 19. UNFPA (2011), Già hóa dân số và người cao tuổi ở Việt Nam. Qũy dân số Liên hợp quốc tại Việt Nam. 20. Vahter, P. (2004). ‘The Effect of Foreign Direct Investment on Labour Productivity. ‘ Working Paper No. 32 edn. University of Tartu. 21. World Bank (2017), Việt Nam: Tăng cường năng lực cạnh tranh và liên kết của doanh nghiệp vừa và nhỏ - Bài học kinh nghiệm trong nước và quốc tế 22. Vietnamese Ministry of Planning and Investment. (2007). Report on investment in automotive industry. Hanoi. 23. Linh and Huyền (2018), Một số giải pháp nâng cao hiệu quả sản xuất kinh doanh của doanh nghiệp thép. http://tapchitaichinh.vn/nghien-cuu-trao-doi/nghien-cuu-dieu-tra/mot-so-giai-phap-nang- cao-hieu-qua-san-xuat-kinh-doanh-cua-doanh-nghiep-thep-137533.html. 24. Nguyễn Đình Tài (2017), Nhận dạng các cụm liên kết ngành và một số đề xuất chính sách tại Việt Nam. Tạp chí Tài chính, kỳ 1. Thanh Nien Newspaper (2018), serious shortage of technical labour in electronics. https://thanhnien.

90 Productivity and Competitiveness of Viet Nam’s Enterprises vn/giao-duc/thieu-hut-tram-trong-lao-dong-ky-thuat-nganh-dien-tu-935343.html. References 25. Business Forum (2017), Many worries for Viet Nam steel industry. http://enternews.vn/nhieu-moi-lo- cho-nganh-thep-viet-109154.html. 26. Huynh The Du, Do Thien Anh Tuan and Dinh Cong Khai (2014), Tương lai nào cho ngành thép Việt Nam. Institute of Public Policy.Minh Đức (2017), Hỗ trợ doanh nghiệp tham gia chuỗi cung ứng toàn cầu. http://www.sggp.org.vn/ho-tro-doanh-nghiep-tham-gia-chuoi-cung-ung-toan-cau-479580. To be revised html. 1. Blomstrom, M., and Kokko, Ari., (1998). Multinational Corporations and Spillovers. Centre for Economic 27. Phạm Đình Hạnh (2016), Đầu tư phát triển hạ tầng giao thông: Thực trạng và giải pháp. Tạp chí cộng Policy Research, CEPR Discuss Paper: No.1365. sản. http://www.tapchicongsan.org.vn/Home/Viet-nam-tren-duong-doi-moi/2016/38469/Dau-tu- 2. Bùi Trinh và Nguyễn Việt Phong (2011), ‘Định hướng lại cơ cấu ngành, vùng’. Thời báo Kinh tế phat-trien-ha-tang-giao-thong-Thuc-trang-va-giai.aspx, 3. Sài Gòn Online. Trang web: [http://www.thesaigontimes.vn/home/diendan/ykien/68264/]. 28. Guillermo Arenas (2017). Viet Nam in the ASEAN Motor Vehicle Industry. Chapter 11 in “Viet Nam at a 4. Business Monitor International. (2010). Viet Nam Autos Report Q4 2010. Crossroads Engaging in the Next Generation of Global Value Chains”, World Bank Group. 5. CIEM (2012), Năng lực cạnh tranh và công nghệ ở cấp độ doanh nghiệp tại Việt Nam. Nhà xuất 29. Aggrey, N., Eliab, L., & Joseph, S. (2010). The Relationship between Firm Size and Technical Efficiency 6. bản Lao động in East Africa Manufacturing Firms. Journal of Sustainable Development in Africa, 12(4), 226–236. 7. Chuc, et al. (2008), FDI Horizontal and Vertical Effects on Local Firm Technical Efficiency . Working 30. Alvarez, R., & Crespi, G. (2003). Determinants of Technical Efficiency in Small Firms. Small Business Paper, No.17. DEPOCEN. Economics, 20(3), 233–244. 8. DBS Research Group. (2015). Viet Nam: Asia’s latest electronics spark. DBS Research Group. 31. Anos-Casero, P., & Udomsaph, C. (2009). What Drives Firm Productivity Growth? (Policy Research Working Paper No. 4841). The World Bank. 9. GSO (2017), Điều tra lao động việc làm hàng năm. 32. Casson, M. (2005). Entrepreneurship and the theory of the firm. Jounal of Economic Behavior and 10. Hollweg et al., (2017), Viet Nam at a Crossroads - Engaging in the Next Generation of Global Value Organization, 58, 327–348. Chains. World Bank Group. 33. Ganotakis, P. (2012). Founder’s human capital and the performance of UK new technology based 11. ILO (2016), Wage and productivity in the garment sector in Asia and the Pacific and the Arab States. firms. Small Business Economics, 39(2), 495–515. 12. MOIT (2016). Studying the Proposal to Develop Supporting Industries via Promoting Linkage and 34. Kim, S. (2003). Identifying and estimating sources of technical inefficiency in Korean manufacturing Supply to Multinational Corporations Operating in the Electronics Sector of Viet Nam. industries. Contemporary Economic Policy, 21(1), 132–144. 13. Nguyen, T. B. (2008). The Vietnamese Automobile Industry and Globalization. The Fourth VDF-Tokyo 35. MPI, & UNDP. (2005). Services Sector Development: A key to Viet Nam’s Sustainable Growth. Conference on the development of Viet Nam, (pp. 197-214). Tokyo. 36. Pagano, P., & Schivardi, F. (2001). Firm size distribution and growth (Temi di discussione No. 394). 14. Porter, Michael E. (2010), Báo cáo năng lực cạnh tranh Việt Nam 2010. CIEM và Lee Kuan Yew School of Public Policy. 37. Saignaleuth, S. (2013). Characteristics of FDI in Lao PDR and its effect on growth enterprise performance (Doctoral thesis). Kobe University. 15. Takii, S. and D. A. Narjoko (2012), ‘FDI Forward Linkage Effect and Local Input Procurement- Evidence from Indonesian Manufacturing-’, in Hahn, C. H. And D. A. Narjoko (eds.), Dynamics of Firm Selection 38. Sanguinetti, M. J. S., & Fruentes, A. (2012). An Analysis of Productivity Performance in Spain Before Process in Globalized Economies. ERIA Research Project and During the Crisis: Exploring the Role of Institutions (OECD Economics Department Working Papers No. 973). OECD. 16. Tran, D.-H., & Ngo, D.-T. (2014). Performance of the Vietnamese Automobile Industry: A Measurement using DEA. Asian Journal of Business and Management. 39. Sinani, E., Jones, D. C., & Mygind, N. (2003). Determinants of firm-level technical efficiency: Evidence using Stochastic Frontier Approach. Presentation presented at the The 2nd Hellenic Workshop on 17. Report 2011 no.3, pp.111-146. Measurement of Efficiency and Productivity Analysis, Patras, Greece. 18. Trần Văn Thọ (2005), Biến động kinh tế Đông Á và con đường công nghiệp hóa Việt Nam, Nhà xuất 40. Taymaz, E. (2005). Are small firms really less productive? Small Business Economics, 25, 429–445. bản Chính trị Quốc gia, Hà Nội 41. Yang, C.-H., & Chen, K.-H. (2009). Are small firms less efficient? Small Business Economics, 32, 375– 19. UNFPA (2011), Già hóa dân số và người cao tuổi ở Việt Nam. Qũy dân số Liên hợp quốc tại Việt Nam. 395. 20. Vahter, P. (2004). ‘The Effect of Foreign Direct Investment on Labour Productivity. ‘ Working Paper 42. GSO (2018), Statistical yearbooks and survey data No. 32 edn. University of Tartu. 43. IPSARD (2014), Factors that reduce the growth rate of agriculture and propose measures to boost 21. World Bank (2017), Việt Nam: Tăng cường năng lực cạnh tranh và liên kết của doanh nghiệp vừa và agricultural growth in the coming time, Ministerial-level research project. nhỏ - Bài học kinh nghiệm trong nước và quốc tế 44. MARD (2013), Agriculture Restructuring Programme in Viet Nam 22. Vietnamese Ministry of Planning and Investment. (2007). Report on investment in automotive industry. Hanoi. 45. World Bank and IPSARD (2017), Report on jobs diagnostic in agri-food sector. 23. Linh and Huyền (2018), Một số giải pháp nâng cao hiệu quả sản xuất kinh doanh của doanh nghiệp thép. http://tapchitaichinh.vn/nghien-cuu-trao-doi/nghien-cuu-dieu-tra/mot-so-giai-phap-nang- cao-hieu-qua-san-xuat-kinh-doanh-cua-doanh-nghiep-thep-137533.html. 24. Nguyễn Đình Tài (2017), Nhận dạng các cụm liên kết ngành và một số đề xuất chính sách tại Việt Nam. Tạp chí Tài chính, kỳ 1. Thanh Nien Newspaper (2018), serious shortage of technical labour in electronics. https://thanhnien.

90 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 91 Appendix 1. Technical Notes on Productivity Analysis

A.1.1. Labour productivity: Concept and international comparison Concept

Among other productivity measures such as multi-factor productivity or capital productivity, labour productivity is particularly important in the economic and statistical analysis of a country. Labour productivity is equal to the ratio between a volume measure of output and a measure of input use (OECD, 2011). The volume measure of output, which reflects the goods and services produced by the workforce, is normally measured by gross domestic production (GDP). The measure of input use, which reflects the time, effort and skills of the workforce, is measured either by the total number of hours worked of all persons employed or total employment (head count). There are both advantages and disadvantages associated with the different input measures that are used in the calculation of labour productivity. It is generally accepted that the total number of hours worked is the most appropriate measure of labour input because a simple headcount of employed persons can hide changes in average hours worked, caused by the evolution of part-time work or the effect of variations in overtime, absence from work or shifts in normal hours. In contrast, total employment is easier to measure than the total number of hours worked.

It is noticeable that labour productivity changes reflect the joint influence of changes in capital, as well as technical, organizational and efficiency change within and between firms, the influence of economies of scale, varying degrees of capacity utilization and measurement errors. Labour productivity only partially reflects the productivity of labour in terms of the personal capacities of workers or the intensity of their effort. The ratio between output and labour input depends to a large degree on the presence of other inputs, as mentioned above. International comparison

The comparison of labour productivity across countries requires Purchasing Power Parity (PPP) data for GDP. A purchasing-power-parity exchange rate is an exchange rate estimated on the assumption that the same set of (international) prices prevail for the same goods, quality adjusted, in all economies. While it is reasonable to assume that with free trade, the prices of tradable goods should have a tendency to converge, the same cannot be said of non-tradable goods and services. Therefore, the GDP data will be converted to international dollars using PPP for GDP rates. A.1.2. Sources of Labour Productivity: Shift Share Analysis

One way to analyze sources of productivity growth is to study the impact of the difference in sectoral structure on aggregate labour productivity growth in the different countries (Wolff 2000; Rao and Tang 2001). Sectoral differences could be due to either differences in labour productivity growth or changes in relative size, which is decomposed by a shift-share analysis. This analysis is widely used to measure the contribution of different sectors to aggregate productivity growth (see, for example, Dekle and Vandenbroucke 2006, IMF 2006, and Usui 2011). The underlying assumption of this formula is that real output is calculated in constant prices using fixed-base Laspeyres quantity and Paasche price indexes at both the aggregate and sectoral levels.

The economy’s real ouput is equal to the sum of sectoral real ouput where i = 1, … N denotes the industry and t = 1, … T denotes the time period. Defining aggregate labour productivity as and sectoral productivity where and represent labour input used in the aggregate economy and in sector i(respectively) such that . In this case, since real output is additive:

92 Productivity and Competitiveness of Viet Nam’s Enterprises Appendix 1. Technical Notes on Productivity Analysis

where .

A.1.1. Labour productivity: Concept and international comparison Sectoral contributions to aggregate labour productivity growth can be computed by looking at Concept productivity changes between two periods of time:

Among other productivity measures such as multi-factor productivity or capital productivity, labour productivity is particularly important in the economic and statistical analysis of a country. Labour productivity is equal to the ratio between a volume measure of output and a measure of input use (OECD, 2011). The volume measure of output, which reflects the goods and services produced by the workforce, is normally measured by gross domestic production (GDP). The measure of input use, which reflects the time, effort and skills of the workforce, is measured either by the total number of hours worked of Defining , so that: all persons employed or total employment (head count). There are both advantages and disadvantages associated with the different input measures that are used in the calculation of labour productivity. It is Therefore, (A1.2.1) generally accepted that the total number of hours worked is the most appropriate measure of labour input According to this formula, sectoral contributions to aggregate productivity growth can be broken down because a simple headcount of employed persons can hide changes in average hours worked, caused by the evolution of part-time work or the effect of variations in overtime, absence from work or shifts in into three effects. The first term of the equation (1) represents the ‘within sector effect’. If sectorallabour normal hours. In contrast, total employment is easier to measure than the total number of hours worked. shares remain unchanged over time ( ), the second and third terms of the equation equal 0 and

It is noticeable that labour productivity changes reflect the joint influence of changes in capital, as well the contribution of each sector collapses to the first term, which is the sectorallabour productivity as technical, organizational and efficiency change within and between firms, the influence of economies growth weighted by the sector’s real share in aggregate real output. of scale, varying degrees of capacity utilization and measurement errors. Labour productivity only The second term of the equation (A2.1) captures the ‘structural change effect’. Aggregate labour partially reflects the productivity of labour in terms of the personal capacities of workers or the intensity productivity can increase even when sectoral labour productivity remains constant, as long as labour of their effort. The ratio between output and labour input depends to a large degree on the presence of moves from sectors with below average labour productivity levels towards sectors with above average other inputs, as mentioned above. labour productivity levels (Denison, 1962). This effect is positive when and the magnitude of International comparison the effect is scaled by the ratio between the sector’s labour productivity level and the aggregate labour The comparison of labour productivity across countries requires Purchasing Power Parity (PPP) data for productivity level. GDP. A purchasing-power-parity exchange rate is an exchange rate estimated on the assumption that the same set of (international) prices prevail for the same goods, quality adjusted, in all economies. While The third term of the equation (A2.1) is the ‘interaction effect’. This effect will be positive either when it is reasonable to assume that with free trade, the prices of tradable goods should have a tendency to labour has moved toward a sector with positive labour productivity growth ( and ) or converge, the same cannot be said of non-tradable goods and services. Therefore, the GDP data will be when labour has moved away from a sector with negative labour productivity growth ( and converted to international dollars using PPP for GDP rates. ) (see Baumol 1967 and Baumol et al. 1985). The magnitude of this effect is also scaled by the A.1.2. Sources of Labour Productivity: Shift Share Analysis ratio between the sector’s labour productivity level and the aggregate labour productivity level.

One way to analyze sources of productivity growth is to study the impact of the difference in sectoral A.1.3. Sources of Total Factor Productivity Growth: Decomposition Exercise structure on aggregate labour productivity growth in the different countries (Wolff 2000; Rao and Tang Stochastic frontier production function for a sector has the following form: 2001). Sectoral differences could be due to either differences in labour productivity growth or changes in relative size, which is decomposed by a shift-share analysis. This analysis is widely used to measure (A.1.3.1) the contribution of different sectors to aggregate productivity growth (see, for example, Dekle and Vandenbroucke 2006, IMF 2006, and Usui 2011). The underlying assumption of this formula is that real of which output is calculated in constant prices using fixed-base Laspeyres quantity and Paasche price indexes i denotes the sector ; at both the aggregate and sectoral levels. t time; Y – output; The economy’s real ouput is equal to the sum of sectoral real ouput where i = 1, … N denotes the X – capital and labour inputs; industry and t = 1, … T denotes the time period. Defining aggregate labour productivity as and ε – error term ∼ iid N (0,1); sectoral productivity where and represent labour input used in the aggregate economy f(.) - production frontier); and in sector i(respectively) such that . In this case, since real output is additive: e-u measure technical efficiency (TE), u ≥ 0, normal distribution (truncated), TE takes value in range (0.1].

92 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 93 Taking logarithm of (A.3.1):

(A.1.3.2) of which: j = 1, 2 – capital and labour inputs;

– elasticity of output with respect to input j;

– technological progress

– technical efficiency change

Therefore: (A.1.3.3)

TFP is defined as (A1.3.4)

of which , witj price of factor j in sector i at time t. sitj – share of factor j in total cost39

From (A.1.3.3) and (A.1.3.4):

(A.1.3.5) of which: returns to scale in sector i, therefore share of factor j or elasticity of output with respect to factor j in the case of constant returns to scale (CRS).

Under the CRS assumption, the third term in equation (A.3.5) expresses the degree of productivity growth induced by raising scale of sector i, or alternatively, scale efficiency (SEC).

Therefore, from equation (A.1.3.5), TFP growth can be decomposed into 4 components:

Estimation

Stochastic frontier production function in translog form can be expressed as follows: lnYit = β0 + tt + 0.5βttt2 + βKlnKit + βLlnLit + 0.5βKK(lnKit)2 + 0.5βLL(lnLit)2 + βKLlnKitlnLit + βtKtlnKit+ βtLtlnLit – uit + εit (A.1.3.6)

Assume: uit = uie-ηt ~ N+(µ, σu2), εit ~ N(0, σε2), cov(uit, εit) = 0

Equation (A.1.3.6) cannot be estimated by OLS because of the endogeneity problem (output depends on capital and labour in the production function, but capital and labour are determined by output in factor demand functions. Therefore, IV estimation will be applied, with lagged values of capital and labour being used as instruments. These instruments are available thanks to the availability of panel data of Enterprise Census.

Technological progress (TC) and technical efficiency change (TEC) can be calculated as follows :

TCit = βt + βttt + βtKlnKit+ βtLlnLit

39 Total cost of capital and labour is the sum of total wage bill and depreciation

94 Productivity and Competitiveness of Viet Nam’s Enterprises Taking logarithm of (A.3.1): TECit = ηuie-ηt = ηuit

(A.1.3.2) Change in scale efficiency and change in allocative efficiency can be calculated as follows: of which: j = 1, 2 – capital and labour inputs;

of which – elasticity of output with respect to input j; Elasticity of output with respect to capital and labour can be calculated as follows: – technological progress itK = βK + βKKlnKit + βKLlnLit + βtKt

– technical efficiency change αitL = βL + KLlnKit + βLLlnLit + βtLt

Returns too scale of sector i, at time t: RTSit = αitK + αitL

Therefore: (A.1.3.3) Capital and labour shares: ρitK = αitK/RTSit, ρitL = αitL/RTSit

Analysis of determinants of technical efficiency – a component of TFP, using the stochastic frontier production method, is available in Stata. TFP is defined as (A1.3.4) A.1.4. Estimation of sector-level FDI’s forward and backward linkages of which , witj price of factor j in sector i at time t. sitj – share of factor j in total cost39

From (A.1.3.3) and (A.1.3.4): Horozontaljt is to measure the presence of foreign firms in sector j at time t, defined as follows: y (A.1.3.5) ∑ ∀j∈i j,t Horizontal = (A.1.4.1) i,t Y i,t of which: returns to scale in sector i, therefore share of factor j or elasticity of output with respect to factor j in the case of constant returns to scale (CRS). where: Y Under the CRS assumption, the third term in equation (A.3.5) expresses the degree of productivity i,t gross output/labour of foreign-invested firm j of the sector i at time t growth induced by raising scale of sector i, or alternatively, scale efficiency (SEC). Yi,t total gross output/labour, of the sector i at time t. Therefore, from equation (A.1.3.5), TFP growth can be decomposed into 4 components: Usually, the conventional measure of horizontal presence of FDI in a sector will be calculated using �� the output measure of FDI firms within this sector at a point of time. However, one can calculate two �� measures of horizontal effects, namely (i) the horizontal output measure of FDI presence; and (ii) the Estimation horizontal employment measure of FDI presence. By including the horizontal employment measure of FDI presence together with the horizontal output measure of FDI presence, one can disentangle the Stochastic frontier production function in translog form can be expressed as follows: effect of labour mobility, i.e. labour turnover from other spillover effects such as the competition effect lnYit = β0 + tt + 0.5βttt2 + βKlnKit + βLlnLit + 0.5βKK(lnKit)2 + 0.5βLL(lnLit)2 + βKLlnKitlnLit + βtKtlnKit+ or the demonstration effect. βtLtlnLit – uit + εit (A.1.3.6) Following Javorcik (2004) and others, we define Backwardjt as Assume: uit = uie-ηt ~ N+(µ, σu2), εit ~ N(0, σε2), cov(uit, εit) = 0 Backward = a Horizo n tal (A.1.4.2) i,t ∑ j if j ≠i ij j,t Equation (A.1.3.6) cannot be estimated by OLS because of the endogeneity problem (output depends on where aij is taken directly from input-output table. capital and labour in the production function, but capital and labour are determined by output in factor (y − e ) demand functions. Therefore, IV estimation will be applied, with lagged values of capital and labour ∑∀j∈i j,t j,t Forwardjt is defined as Forward = a being used as instruments. These instruments are available thanks to the availability of panel data of j,t ∑i if i≠ j ij Yi,t − Ei,t Enterprise Census. (A.1.4.3)

Technological progress (TC) and technical efficiency change (TEC) can be calculated as follows : where aij is the direct IO coefficient. Since IO table does not allow one to calculate the value of ej,t , one should assume that the proportion of foreign export within sector is linear correlation with the equity TCit = βt + βttt + βtKlnKit+ βtLlnLit share of foreign firms. Hence it can be approximated as follows:

39 Total cost of capital and labour is the sum of total wage bill and depreciation

94 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 95 k j,t (A.1.4.4) where k j,t is capital stock of foreign firm of sector i at time t and Ki,t is total sectoral capital stock of sector i at time t. A.1.5. The Wage Growth and Firm’s Competitiveness Nexus: A Simple Model

Assume that the production function has the Cobb-Douglas form with time varying technology as follows: (A.1.5 1).

Under the assumption of constant returns to scale (CRS), we have t+βt=1.

(A.1.5.2) with - labour productivity, per hour; - capital intensity; A – TFP

Labour demand is derived from firm profit maximizing behavior:

(A.1.5.3) where Pt – nominal output price, Wt – nominal wage, Rt – nominal rental price

First order condition of profit maximization with respect to labour yields:

(A.1.5.4) where - real wage

Therefore, real wage growth in the period from time t to t+1 is calculated as follows:

(A.1.5.5)

From (A.1.5.5), if worker’s wage growth in real terms is proportional, by factor , to labour productivity growth, wage growth can be said to be competitiveness-neutral, as it is in line with technological change in the economy or sector. If the rate of wage growth is greater (smaller) than , then wage growth is competitiveness-hurting (competitiveness-enhancing), both may not be sustainable in the long term, because they either hurt owner of capital (the former case), or worker (the latter case).

If βt+1=βt = β (for example, in the short run, technology does not change), one can expect a one-to-one relationship between wage growth and productivity growth. In the longer run, this may no longer be valid. In other words, if technology changes in favour of capital (i.e. capital-biased technology), wage growth is expected to fall behind productivity growth and vice verse. Importantly, such a relationship may not be uniform, because the rate of technological change may vary across sectors. One can expect that this rate is higher in sectors with higher degree of global integration in general and GVC integration in particular, as technological spillovers along GVCs tend to be more rapid.

Deviations from the competitiveness-neutral wage growth rate (i.e from ) should be explained by non-technological factors including labour market regulations and the country’s political economy landscape (interaction between players such as FDI, domestic firms, trade unions, non-governmental organizations, which to a large extent depends on their respective market and bargaining powers). The former binds behavior of the latter and vice verse, the latter interact with one another to shape up the former.

Source: Adapted from Tran Ngo Minh Tam et al. (2018)

96 Productivity and Competitiveness of Viet Nam’s Enterprises k j,t A.1.6. Miscellaneous • Market power (A.1.4.4) An indicator that measures monopoly level of an industry at 3-digit level is as follows where k j,t is capital stock of foreign firm of sector i at time t and Ki,t is total sectoral capital stock of sector i at time t. CRi = S21i + S22i + … + S2ni (A.1.6.1) A.1.5. The Wage Growth and Firm’s Competitiveness Nexus: A Simple Model where: Ski – ratio between number of workers in firm k and that in industry i (k=1,2…n) When Ski is market share of firm k in industry i (k=1,2…n), this is the Herfindahl index (also known as Assume that the production function has the Cobb-Douglas form with time varying technology as Herfindahl–Hirschman Index, HHI, or sometimes HHI-score), which is a measure of the size of firms in follows: (A.1.5 1). relation to the industry and an indicator of the amount of competition among them.

Under the assumption of constant returns to scale (CRS), we have t+βt=1. As such, it can range from 0 to 1.0, moving from a huge number of very small firms to a single monopolistic producer. (A.1.5.2) • Firm specialization index with - labour productivity, per hour; - capital intensity; A – TFP A similar index can be calculated to measure the degree of firm’s specialization

Labour demand is derived from firm profit maximizing behavior: CRi = S21i + S22i + … + S2ni (A.1.6.2)

(A.1.5.3) where: Ski – revenue share of activity k out of total revenue of firm i (k=1,2…n) where Pt – nominal output price, Wt – nominal wage, Rt – nominal rental price Clustering

First order condition of profit maximization with respect to labour yields: One can estimate the clustering effect by calculating the industry concentration ratio at 3- or 2-digit (A.1.5.4) level by district: where - real wage LQij = Lij/Li./(L.j/L..) (A.1.6.3) where: Lij – Number of workers of industry i in district j Therefore, real wage growth in the period from time t to t+1 is calculated as follows: Li. – Number of workers of industry in the country (A.1.5.5) L.j – Number of workers in district j L.. – Number of workers in the country From (A.1.5.5), if worker’s wage growth in real terms is proportional, by factor , to labour productivity growth, wage growth can be said to be competitiveness-neutral, as it is in line with technological change • Revealed Comparative Advantage Index (RCA) in the economy or sector. If the rate of wage growth is greater (smaller) than , then wage growth is The RCA indicates whether a country is in the process of extending the products in which it has a trade competitiveness-hurting (competitiveness-enhancing), both may not be sustainable in the long term, potential, as opposed to situations in which the number of products that can be competitively exported because they either hurt owner of capital (the former case), or worker (the latter case). is static. It can also provide useful information about potential trade prospects with new partners. Countries with similar RCA profiles are unlikely to have high bilateral trade intensities unless intra- If βt+1=βt = β (for example, in the short run, technology does not change), one can expect a one-to-one industry trade is involved. RCA measures, if estimated at high levels of product disaggregation, can relationship between wage growth and productivity growth. In the longer run, this may no longer be focus attention on other nontraditional products that might be successfully exported. valid. In other words, if technology changes in favour of capital (i.e. capital-biased technology), wage growth is expected to fall behind productivity growth and vice verse. Importantly, such a relationship The RCA index of country I for product j is often measured by the product’s share in the country’s exports may not be uniform, because the rate of technological change may vary across sectors. One can expect in relation to its share in world trade: that this rate is higher in sectors with higher degree of global integration in general and GVC integration RCAij = (xij/Xit) / (xwj/Xwt) (A.1.6.4) in particular, as technological spillovers along GVCs tend to be more rapid. where xij and xwj are the values of country i’s exports of product j and world exports of product j and Deviations from the competitiveness-neutral wage growth rate (i.e from ) should be explained by where Xit and Xwt refer to the country’s total exports and world total exports. A value of less than unity non-technological factors including labour market regulations and the country’s political economy landscape implies that the country has a revealed comparative disadvantage in the product. Similarly, if the index (interaction between players such as FDI, domestic firms, trade unions, non-governmental organizations, exceeds unity, the country is said to have a revealed comparative advantage in the product. which to a large extent depends on their respective market and bargaining powers). The former binds behavior of the latter and vice verse, the latter interact with one another to shape up the former.

Source: Adapted from Tran Ngo Minh Tam et al. (2018)

96 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 97 • Screening for best performing (outstanding) sub-sectors within manufacturing, services and agriculture in the period 2011-2016, subject to the availability of GDP deflators) in terms of:

- productivity performance: breaking down manufacturing into four sub-groups (quadrants): (i) low initial level, low growth (laggard); (ii) low level, high growth (emerging); (iii) high level, low growth (mature); and (iv) high level, high growth (outstanding), with the average productivity level and average productivity growth rate are used to divide the XY space into the four quadrants.

The third dimension – the sub-sector share of employment is integrated into the analysis and depicted by a circle.

Quadrant 2: Low level, high growth - emerging Quadrant 4: High level, high growth - OUTSTANDING

Quadrant 1: Low level, low growth - laggard Quadrant 3: High level, low growth - mature

- pro-worker characteristics (employment and income generation): breaking down manufacturing into four sub-groups (quadrants): (i) low productivity (in 2016, captured by the Enterprise Census 2017), low average income40; (ii) low productivity, high average income; (iii) high productivity, and (iv) high productivity, high average income - best performing

- Position in the value chain (proxy): breaking down manufacturing into four sub-groups (quadrants) by the value added over revenue ratio: (i) low level, low growth; (ii) low level, high growth; (iii) high level, low growth; (iv) high level, high growth

Services: similar to manufacturing as discussed above

Agriculture: similar to manufacturing as discussed above, but built on the World Bank’s ISPARD work on variation of labour productivity across agricultural sub-sectors, highlighting high

It should be noted that the average labour productivity of a sub-sector is calculated as the sum of value added by all firms in the sub-sector divided by the sum of workers of all firms in the same sub-sector. The growth rate of labour productivity of the sub-sector is calculated by dividing the average labour productivity of the sub-sector in the final year over the same figure in the initial year.

The screening process will be conducted in two steps. In step 1, it will be done for two-digit sub-sectors. In step 2, sub-sectors in quadrant 4 (high level, high growth) will be selected for further screening, but at four-digit level. Sub-sectors in quadrant 4 as identified in the second step will be selected for international comparison, and for the construction of detailed sector profiles (average size, age, spatial distribution, FDI share, human capital, R&D spending etc.).

40 The average income can be calculated directly from the dataset of Enterprise Census (the first best solution) or from data of the Labour Force Survey 2017

98 Productivity and Competitiveness of Viet Nam’s Enterprises • Screening for best performing (outstanding) sub-sectors within manufacturing, services and Appendix 2. NACE Industrial Classification agriculture in the period 2011-2016, subject to the availability of GDP deflators) in terms of:

- productivity performance: breaking down manufacturing into four sub-groups (quadrants): (i) low Industries are classified according to the Statistical Classification of Economic Activities in the European initial level, low growth (laggard); (ii) low level, high growth (emerging); (iii) high level, low growth Community (NACE) as follows: (mature); and (iv) high level, high growth (outstanding), with the average productivity level and average productivity growth rate are used to divide the XY space into the four quadrants. Sector Name of sector Agriculture and related service activities The third dimension – the sub-sector share of employment is integrated into the analysis and depicted Agriculture, forestry, Forestry and related service activities by a circle. and fishing Fishing and aquaculture Quadrant 2: Low level, high growth - emerging Quadrant 4: High level, high growth - Mining of coal and lignite OUTSTANDING Mining of metal ores Quadrant 1: Low level, low growth - laggard Quadrant 3: High level, low growth - mature Other mining and quarrying Mining, electricity, Mining support service activities - pro-worker characteristics (employment and income generation): breaking down manufacturing and water Electricity, gas, steam and air conditioning supply into four sub-groups (quadrants): (i) low productivity (in 2016, captured by the Enterprise Census 2017), low average income40; (ii) low productivity, high average income; (iii) high productivity, and (iv) Water collection, treatment and supply high productivity, high average income - best performing Sewerage and sewer treatment activities Waste collection, treatment and disposal activities; materials recovery - Position in the value chain (proxy): breaking down manufacturing into four sub-groups (quadrants) Construction of buildings by the value added over revenue ratio: (i) low level, low growth; (ii) low level, high growth; (iii) high level, low growth; (iv) high level, high growth Construction Civil engineering Specialized construction activities Services: similar to manufacturing as discussed above Manufacture of pharmaceuticals, medicinal chemical and botanical products Agriculture: similar to manufacturing as discussed above, but built on the World Bank’s ISPARD work on Manufacture of computer, electronic and optical products variation of labour productivity across agricultural sub-sectors, highlighting high Manufacture of chemicals and chemical products High-tech Manufacture of electrical equipment It should be noted that the average labour productivity of a sub-sector is calculated as the sum of value manufacturing added by all firms in the sub-sector divided by the sum of workers of all firms in the same sub-sector. Manufacture of machinery and equipment n.e.c The growth rate of labour productivity of the sub-sector is calculated by dividing the average labour Manufacture of motor vehicles; trailers and semi- trailers productivity of the sub-sector in the final year over the same figure in the initial year. Manufacture of other transport equipment Manufacture of coke and refined petroleum products The screening process will be conducted in two steps. In step 1, it will be done for two-digit sub-sectors. In step 2, sub-sectors in quadrant 4 (high level, high growth) will be selected for further screening, Manufacture of rubber and plastics products but at four-digit level. Sub-sectors in quadrant 4 as identified in the second step will be selected for Medium-tech Manufacture of other non-metallic mineral products international comparison, and for the construction of detailed sector profiles (average size, age, spatial manufacturing Manufacture of basic metals distribution, FDI share, human capital, R&D spending etc.). Manufacture of fabricated metal products, except machinery and equipment Repair and installation of machinery and equipment

40 The average income can be calculated directly from the dataset of Enterprise Census (the first best solution) or from data of the Labour Force Survey 2017

98 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 99 Manufacture of food products Manufacture of beverages Manufacture of tobacco products Manufacture of textiles Manufacture of wearing apparel Low-tech Manufacture of leather and related products manufacturing Manufacture of wood and of products of wood and cork, except furniture; manufacture of articles of straw and plaiting materials Manufacture of paper and paper products Printing and reproduction of recorded media Manufacture of furniture Other manufacturing Air transport Legal and accounting activities Activities of head office; management consultancy activities Architectural and engineering activities; Technical testing and analysis Advertising and market research Other professional, scientific and technical activities Employment activities Security and investigation activities Motion picture, video and television programme activities; Sound recording and music publishing activities High-tech Broadcasting and programming activities knowledge-intensive Telecommunication services Computer programming, consultancy and related activities Information service activities Scientific research and development Financial service activities (except insurance and pension funding) Knowledge-intensive Financial service activities (except insurance and pension funding) financial services Other financial activities Publishing activities Travel agency, tour operator and other reservation service activities Education Human health activities Other knowledge- Residential care activities intensive services Creative, art and entertainment activities Libraries, archives, museums and other cultural activities Lottery activities, Gambling and betting activities Sports activities and amusement and recreation activities Activities of other membership organizations

100 Productivity and Competitiveness of Viet Nam’s Enterprises Manufacture of food products Wholesale and retail trade and repair of motor vehicles and motorcycles Manufacture of beverages Wholesale trade (except of motor vehicles and motorcycles) Manufacture of tobacco products Retail trade (except of motor vehicles and motorcycles) Manufacture of textiles Land transport, transport via railways, via pipeline Manufacture of wearing apparel Warehousing and support activities for transportation Low-tech Manufacture of leather and related products Postal and courier activities manufacturing Manufacture of wood and of products of wood and cork, except furniture; Accommodation Less-knowledge manufacture of articles of straw and plaiting materials Food and beverage service activities intensive market Manufacture of paper and paper products services Real estate activities Printing and reproduction of recorded media Renting and leasing of machinery and equipment (without operator); of Manufacture of furniture personal and household goods; of no financial intangible assets Other manufacturing Services to buildings and landscape activities Air transport Office administrative and support activities; other business support service Legal and accounting activities activities Activities of head office; management consultancy activities Repair of computers and personal and households goods Architectural and engineering activities; Technical testing and analysis Other personal service activities Advertising and market research Other professional, scientific and technical activities Employment activities Security and investigation activities Motion picture, video and television programme activities; Sound recording and music publishing activities High-tech Broadcasting and programming activities knowledge-intensive Telecommunication services Computer programming, consultancy and related activities Information service activities Scientific research and development Financial service activities (except insurance and pension funding) Knowledge-intensive Financial service activities (except insurance and pension funding) financial services Other financial activities Publishing activities Travel agency, tour operator and other reservation service activities Education Human health activities Other knowledge- Residential care activities intensive services Creative, art and entertainment activities Libraries, archives, museums and other cultural activities Lottery activities, Gambling and betting activities Sports activities and amusement and recreation activities Activities of other membership organizations

100 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 101 Appendix 3. Results of Econometric Analysis of Enterprise Census 2012 and 2017

A.3.1. Brief Description of Enterprise Census

Since 2001, GSO has conducted census on enterprises of the economy annually. The census covers all enterprises of different types of ownerships as well as industries with more than 10 labourers. Enterprises with less than 10 labourers have been sampled. However, for every 5 years, firms with less than 10 labourers are all surveyed. The censuses ask for information of the entire year before the surveyed years. For example, the census in 2012 asked for information of the entire 2011.

The contents of questionnaires include two parts. The core part covers basic information of operation of firms such as products, labourers, capital and their cost as well as revenues. From this set of information, one can calculate value added of firms and the production function can be estimated as a consequence. The specialized part, or also called rotating modules vary from year to year, collects information on special topics, for example, technology, environment etc.

With annual censuses, one can construct panel data of enterprises over time, for specific periods as well as for the whole period of 2001-2017. Panel data can in particular help researchers to take into account fixed effects, or also called time-invariant unobserved characteristics of firms.

In this study, we focus on the period of 2011-2016 (censuses conducted in 2012 and 2017 respectively) for two reasons: (1) The censuses in the two years cover all enterprises including ones with less than 10 labourers; (2) they have detailed information on labour composition in terms of qualifications and ages as well as the use of computer and the Internet. The very large number of observations, over 200,000 panel firms in the two years of 2011 and 2015 allows us to do various regressions at the firm level.

102 Productivity and Competitiveness of Viet Nam’s Enterprises Appendix 3. Results of Econometric Analysis of A.3.2. Results of Econometric Analysis of Determinants/Associates of Firm-Level Labour Productivity

Enterprise Census 2012 and 2017 Appendix 3 Table 1. OLS regressions - Determinants of productivity

(1) (2) (3) (4) All firms, All matched All All Matched A.3.1. Brief Description of Enterprise Census 2016 firms between Manufacturing Manufacturing 2015-2016 firms, 2016 matched firms, 2015- Since 2001, GSO has conducted census on enterprises of the economy annually. The census covers 2016 all enterprises of different types of ownerships as well as industries with more than 10 labourers. 0.368*** 0.382*** 0.377*** 0.388*** Enterprises with less than 10 labourers have been sampled. However, for every 5 years, firms with Ln(K/L) less than 10 labourers are all surveyed. The censuses ask for information of the entire year before the (237.03) (219.38) (95.14) (87.18) surveyed years. For example, the census in 2012 asked for information of the entire 2011. 1-4 workers Base 0.465*** 0.423*** 0.419*** 0.382*** The contents of questionnaires include two parts. The core part covers basic information of operation of 5-9 workers (86.31) (70.63) (30.00) (24.19) firms such as products, labourers, capital and their cost as well as revenues. From this set of information, 0.649*** 0.599*** 0.582*** 0.533*** one can calculate value added of firms and the production function can be estimated as a consequence. 10-24 workers The specialized part, or also called rotating modules vary from year to year, collects information on (105.68) (88.89) (39.66) (32.35) special topics, for example, technology, environment etc. 0.730*** 0.673*** 0.686*** 0.627*** 25-49 workers (82.87) (71.64) (37.76) (31.22) With annual censuses, one can construct panel data of enterprises over time, for specific periods as well 0.785*** 0.724*** 0.758*** 0.699*** as for the whole period of 2001-2017. Panel data can in particular help researchers to take into account 50-99 workers fixed effects, or also called time-invariant unobserved characteristics of firms. (68.67) (60.00) (36.35) (30.20) 0.794*** 0.715*** 0.838*** 0.771*** In this study, we focus on the period of 2011-2016 (censuses conducted in 2012 and 2017 respectively) 100-299 workers (59.74) (51.03) (37.50) (30.52) for two reasons: (1) The censuses in the two years cover all enterprises including ones with less than 10 0.809*** 0.708*** 0.904*** 0.825*** labourers; (2) they have detailed information on labour composition in terms of qualifications and ages 300-999 workers as well as the use of computer and the Internet. The very large number of observations, over 200,000 (39.80) (33.51) (31.72) (26.15) panel firms in the two years of 2011 and 2015 allows us to do various regressions at the firm level. 0.826*** 0.706*** 0.975*** 0.893*** From 1000 workers (24.85) (20.88) (24.51) (21.18) -0.00469 0.00636 -0.0332* -0.0429* Female worker ratio (-0.54) (0.66) (-1.65) (-1.95) 0.779*** 0.934*** 0.651*** 0.416** Foreign worker ratio (13.29) (13.70) (3.96) (2.14) Ratio of workers Base without training Ratio of workers with 0.00568 0.00107 0.0204 0.0178 training of less than 3 (0.52) (0.09) (1.14) (0.93) months Ratio of workers with -0.0304** -0.0318** 0.0924*** 0.0802*** primary Vocational (-2.52) (-2.43) (3.99) (3.26) certificate Ratio of workers -0.0193** -0.00315 0.0442** 0.0512** with Secondary (Vocational) Degree or (-1.98) (-0.30) (2.04) (2.19) (Vocational) College Degree Ratio of workers with -0.0217** 0.00663 -0.00993 0.0450 Bachelor degree or (-2.00) (0.56) (-0.36) (1.47) higher

102 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 103 Ratio of workers with 0.146*** 0.102*** 0.146*** 0.122*** Other certificates (11.52) (7.39) (6.64) (5.18) Ratio of workers aged Base 16-30 Ratio of workers aged -0.108*** -0.127*** -0.135*** -0.157*** 31 to 45 (-13.76) (-14.64) (-7.54) (-8.02) Ratio of workers aged -0.151*** -0.167*** -0.248*** -0.252*** 46 to 55 (-11.62) (-11.93) (-7.93) (-7.49) Ratio of workers aged -0.119*** -0.148*** -0.268*** -0.334*** 56 to 60 (-4.90) (-5.73) (-3.91) (-4.54) Ratio of workers aged 0.468*** 0.274*** 0.509*** 0.249*** over 60 (18.39) (10.13) (7.20) (3.34) Education level of 0.0212*** 0.00939 0.0484*** 0.0401*** manager (3.99) (1.63) (4.45) (3.43) 0.0293*** 0.0190*** 0.0143*** 0.00277 Manager's age (18.65) (10.95) (4.26) (0.75) Manager's age -0.000337*** -0.000233*** -0.000193*** -0.0000807** squared (-19.81) (-12.60) (-5.44) (-2.10) Ln(Agglomeration 0.00541*** 0.00134 -0.000504 -0.00177 index) (3.39) (0.76) (-0.17) (-0.56) Industrial ratio of -0.0107 -0.0219 0.112*** 0.120*** workers from foreign (-0.71) (-1.34) (5.83) (5.72) firms in a district Share of workers 0.000505*** 0.000570*** 0.000650** 0.000821** often use computer (3.72) (3.87) (2.08) (2.46) Share of workers 0.000247* 0.000190 -0.000566* -0.000800** often use internet (1.91) (1.37) (-1.92) (-2.53) 0.0874*** 0.0861*** -0.0345 -0.0241 Firm has computers (5.30) (4.79) (-0.95) (-0.61) Firm has its own 0.0556*** 0.0465*** 0.0512*** 0.0410*** website (11.48) (8.92) (4.88) (3.67) Firm use the internet 0.0278*** 0.0274*** 0.0260** 0.0222* for operation (5.32) (4.84) (2.23) (1.78) management Firm use the internet -0.0199*** -0.0190*** -0.0209** -0.0285** for transaction (-4.34) (-3.79) (-2.04) (-2.57) Firm use the 0.0924*** 0.0884*** 0.0594*** 0.0521*** internet for financial (16.82) (14.97) (5.00) (4.12) transaction Agriculture, Forestry -0.271*** -0.245*** 0 0 and Fishery (-13.57) (-10.98) (.) (.) -0.0444** -0.0153 0 0 Mining and Electricity (-2.33) (-0.73) (.) (.) 0.0455*** 0.0651*** 0 0 Construction (4.84) (6.37) (.) (.)

104 Productivity and Competitiveness of Viet Nam’s Enterprises Ratio of workers with 0.146*** 0.102*** 0.146*** 0.122*** High-tech 0.135*** 0.101*** 0.157*** 0.135*** Other certificates (11.52) (7.39) (6.64) (5.18) Manufacturing (8.71) (6.16) (10.98) (8.89) Ratio of workers aged Medium-tech 0.110*** 0.123*** 0.106*** 0.116*** Base 16-30 Manufacturing (10.11) (10.51) (10.24) (10.42) Ratio of workers aged -0.108*** -0.127*** -0.135*** -0.157*** Low-tech Base 31 to 45 (-13.76) (-14.64) (-7.54) (-8.02) manufacturing Ratio of workers aged -0.151*** -0.167*** -0.248*** -0.252*** Knowledge-intensive 0.0670*** 0.112*** 0 0 46 to 55 (-11.62) (-11.93) (-7.93) (-7.49) service (6.99) (10.72) (.) (.) Ratio of workers aged -0.119*** -0.148*** -0.268*** -0.334*** Less knowledge- 0.131*** 0.146*** 0 0 56 to 60 (-4.90) (-5.73) (-3.91) (-4.54) intensive service (16.20) (16.83) (.) (.) Ratio of workers aged 0.468*** 0.274*** 0.509*** 0.249*** Private firms Base over 60 (18.39) (10.13) (7.20) (3.34) 0.280*** 0.273*** 0.106** 0.0940* State Comp. Education level of 0.0212*** 0.00939 0.0484*** 0.0401*** (10.68) (10.21) (2.01) (1.76) manager (3.99) (1.63) (4.45) (3.43) -0.199*** -0.188*** -0.131*** -0.149*** Collective 0.0293*** 0.0190*** 0.0143*** 0.00277 (-12.85) (-11.10) (-3.24) (-3.34) Manager's age (18.65) (10.95) (4.26) (0.75) 0.170*** 0.149*** 0.00636 -0.000397 Mixed Comp. Manager's age -0.000337*** -0.000233*** -0.000193*** -0.0000807** (4.71) (4.09) (0.11) (-0.01) squared (-19.81) (-12.60) (-5.44) (-2.10) 0.445*** 0.352*** 0.0949*** 0.0723*** Foreign Comp. Ln(Agglomeration 0.00541*** 0.00134 -0.000504 -0.00177 (29.10) (21.35) (4.47) (3.11) index) (3.39) (0.76) (-0.17) (-0.56) region==Central -0.342*** -0.240*** -0.425*** -0.360*** Industrial ratio of -0.0107 -0.0219 0.112*** 0.120*** Highlands (-26.39) (-16.46) (-11.55) (-8.79) workers from foreign -0.0971*** 0.00177 -0.105*** -0.0456** (-0.71) (-1.34) (5.83) (5.72) region==Mekong River firms in a district Delta (-11.61) (0.19) (-5.73) (-2.31) Share of workers 0.000505*** 0.000570*** 0.000650** 0.000821** region==Northern -0.351*** -0.273*** -0.330*** -0.288*** often use computer (3.72) (3.87) (2.08) (2.46) Central and Central (-50.97) (-36.19) (-20.17) (-16.29) Share of workers 0.000247* 0.000190 -0.000566* -0.000800** Coast often use internet (1.91) (1.37) (-1.92) (-2.53) region==Northern -0.299*** -0.219*** -0.322*** -0.273*** 0.0874*** 0.0861*** -0.0345 -0.0241 Mountain (-28.96) (-18.94) (-13.77) (-10.42) Firm has computers (5.30) (4.79) (-0.95) (-0.61) -0.172*** -0.111*** -0.182*** -0.135*** region==Ha Noi Firm has its own 0.0556*** 0.0465*** 0.0512*** 0.0410*** (-28.63) (-17.48) (-12.80) (-9.08) website (11.48) (8.92) (4.88) (3.67) region==Red River -0.298*** -0.218*** -0.221*** -0.192*** Firm use the internet 0.0278*** 0.0274*** 0.0260** 0.0222* Delta (except Ha Noi) (-39.53) (-25.76) (-14.45) (-11.49) for operation Ho Chi Minh City Base (5.32) (4.84) (2.23) (1.78) management region==South East -0.0620*** -0.0116 -0.126*** -0.109*** Firm use the internet -0.0199*** -0.0190*** -0.0209** -0.0285** (except Ho Chi Minh) (-7.66) (-1.27) (-7.93) (-6.25) for transaction (-4.34) (-3.79) (-2.04) (-2.57) Involvement in export 0.259*** 0.116*** Firm use the 0.0924*** 0.0884*** 0.0594*** 0.0521*** and import, lag (30.64) (7.57) internet for financial (16.82) (14.97) (5.00) (4.12) 0.984*** 1.117*** 1.482*** 1.722*** transaction Constant (25.06) (25.45) (17.19) (17.89) Agriculture, Forestry -0.271*** -0.245*** 0 0 No. of Obs. 331591 265375 54128 44637 and Fishery (-13.57) (-10.98) (.) (.) Adjusted R2 0.233 0.245 0.256 0.263 -0.0444** -0.0153 0 0 Mining and Electricity F-Statistics 2051 1728 425 356 (-2.33) (-0.73) (.) (.) Prob > F 0.000 0.000 0.000 0.000 0.0455*** 0.0651*** 0 0 Construction (4.84) (6.37) (.) (.) t statistics in parentheses * p<0.10, ** p<0.05, *** p<0.010

104 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 105 Appendix 3 Table 2. OLS regressions - Determinants of productivity

(1) (2) (3) (4) All matched All matched All service All agriculture service agriculture firms, firms, 2016 firms, 2016 firms,2015-2016 2015-2016 0.368*** 0.381*** 0.279*** 0.266*** Ln(K/L) (194.40) (179.71) (29.39) (25.21) 1-4 workers Base 0.469*** 0.415*** 0.215*** 0.266*** 5-9 workers (72.55) (58.49) (5.30) (5.53) 0.684*** 0.616*** 0.123*** 0.162*** 10-24 workers (88.86) (73.96) (2.75) (3.14) 0.759*** 0.684*** 0.154*** 0.198*** 25-49 workers (61.88) (53.47) (2.74) (3.16) 0.809*** 0.734*** 0.321*** 0.371*** 50-99 workers (46.08) (40.67) (3.76) (4.10) 0.843*** 0.766*** 0.104 0.0861 100-299 workers (37.30) (33.25) (1.01) (0.81) 0.842*** 0.770*** -0.217 -0.142 300-999 workers (20.21) (18.36) (-1.57) (-1.01) 0.786*** 0.681*** -0.159 -0.344* From 1000 workers (9.43) (8.19) (-0.86) (-1.79) -0.0146 0.00217 0.0563 0.0173 Female worker ratio (-1.40) (0.19) (0.85) (0.23) 0.492*** 0.606*** 0.189 2.903* Foreign worker ratio (7.13) (7.56) (0.29) (1.67) Ratio of workers without Base training Ratio of workers with -0.0198 -0.0231 0.169*** 0.200*** training of less than 3 (-1.20) (-1.29) (2.75) (3.01) months Ratio of workers with -0.118*** -0.126*** 0.181*** 0.287*** primary Vocational (-7.33) (-7.22) (2.59) (3.83) certificate Ratio of workers with -0.0885*** -0.0737*** 0.385*** 0.488*** Secondary (Vocational) Degree or (Vocational) (-7.00) (-5.33) (6.04) (7.03) College Degree Ratio of workers with -0.0943*** -0.0769*** -0.0517 0.162 Bachelor degree or (-6.76) (-5.04) (-0.59) (1.62) higher Ratio of workers with 0.140*** 0.0858*** 0.0731 -0.0242 Other certificates (7.88) (4.45) (0.77) (-0.23) Ratio of workers aged Base 16-30

106 Productivity and Competitiveness of Viet Nam’s Enterprises Appendix 3 Table 2. OLS regressions - Determinants of productivity Ratio of workers aged 31 -0.0895*** -0.108*** -0.310*** -0.236*** to 45 (-9.23) (-10.08) (-4.27) (-2.86) (1) (2) (3) (4) Ratio of workers aged 46 -0.114*** -0.137*** -0.455*** -0.378*** All matched All matched All service All agriculture to 55 service agriculture firms, (-7.09) (-7.91) (-5.67) (-4.22) firms, 2016 firms, 2016 firms,2015-2016 2015-2016 Ratio of workers aged 56 -0.0778*** -0.122*** -0.561*** -0.431*** 0.368*** 0.381*** 0.279*** 0.266*** to 60 (-2.66) (-3.94) (-5.59) (-3.98) Ln(K/L) (194.40) (179.71) (29.39) (25.21) Ratio of workers aged 0.448*** 0.248*** -0.396*** -0.342** 1-4 workers Base over 60 (14.57) (7.58) (-2.91) (-2.31) 0.469*** 0.415*** 0.215*** 0.266*** Education level of 0.0256*** 0.0112 0.110*** 0.0863** 5-9 workers (72.55) (58.49) (5.30) (5.53) manager (3.68) (1.48) (3.06) (2.20) 0.684*** 0.616*** 0.123*** 0.162*** 0.0326*** 0.0216*** 0.0344*** 0.0264** 10-24 workers Manager's age (88.86) (73.96) (2.75) (3.14) (16.41) (9.92) (2.96) (2.00) 0.759*** 0.684*** 0.154*** 0.198*** -0.000368*** -0.000256*** -0.000338*** -0.000270** 25-49 workers Manager's age squared (61.88) (53.47) (2.74) (3.16) (-16.99) (-10.89) (-2.85) (-2.02) 0.809*** 0.734*** 0.321*** 0.371*** 0.00659*** 0.00134 0.0226*** 0.0209*** 50-99 workers Ln(Agglomeration index) (46.08) (40.67) (3.76) (4.10) (3.05) (0.55) (3.26) (2.69) 0.843*** 0.766*** 0.104 0.0861 Industrial ratio of -0.0158 0.0255 0.115 -0.0365 100-299 workers workers from foreign (37.30) (33.25) (1.01) (0.81) (-0.62) (0.94) (0.72) (-0.21) firms in a district 0.842*** 0.770*** -0.217 -0.142 300-999 workers Share of workers often 0.000490*** 0.000482*** -0.00101 -0.000521 (20.21) (18.36) (-1.57) (-1.01) use computer (2.91) (2.64) (-1.04) (-0.48) 0.786*** 0.681*** -0.159 -0.344* From 1000 workers Share of workers often 0.000317** 0.000256 0.00170* 0.00122 (9.43) (8.19) (-0.86) (-1.79) use internet (1.98) (1.49) (1.91) (1.22) -0.0146 0.00217 0.0563 0.0173 Female worker ratio 0.150*** 0.147*** 0.0139 0.0652 (-1.40) (0.19) (0.85) (0.23) Firm has computers (7.14) (6.47) (0.21) (0.85) 0.492*** 0.606*** 0.189 2.903* Foreign worker ratio 0.0550*** 0.0442*** 0.0501 0.0461 (7.13) (7.56) (0.29) (1.67) Firm has its own website (9.08) (6.78) (1.22) (1.08) Ratio of workers without Base training Firm use the internet for 0.0178*** 0.0176** -0.0107 0.00409 operation management Ratio of workers with -0.0198 -0.0231 0.169*** 0.200*** (2.73) (2.51) (-0.27) (0.09) training of less than 3 Firm use the internet for -0.0215*** -0.0162** 0.0514* 0.0485 (-1.20) (-1.29) (2.75) (3.01) months transaction (-3.71) (-2.55) (1.86) (1.62) Ratio of workers with -0.118*** -0.126*** 0.181*** 0.287*** 0.0988*** 0.0929*** 0.0505 0.00228 Firm use the internet for primary Vocational (14.43) (12.63) (0.93) (0.04) (-7.33) (-7.22) (2.59) (3.83) financial transaction certificate (.) (.) (.) (.) Ratio of workers with -0.0885*** -0.0737*** 0.385*** 0.488*** Knowledge-intensive -0.0627*** -0.0237*** 0 0 Secondary (Vocational) service (-9.20) (-3.17) (.) (.) Degree or (Vocational) (-7.00) (-5.33) (6.04) (7.03) College Degree Less knowledge- Base 0 0 0 intensive service Ratio of workers with -0.0943*** -0.0769*** -0.0517 0.162 (.) (.) (.) (.) Bachelor degree or Private firms Base (-6.76) (-5.04) (-0.59) (1.62) higher 0.342*** 0.318*** 0.289*** 0.243*** State Comp. Ratio of workers with 0.140*** 0.0858*** 0.0731 -0.0242 (8.41) (7.69) (3.58) (2.98) Other certificates (7.88) (4.45) (0.77) (-0.23) -0.181*** -0.176*** -0.303*** -0.330*** Collective Ratio of workers aged (-8.12) (-7.36) (-7.44) (-7.15) Base 16-30

106 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 107 0.242*** 0.207*** -0.394 -0.524** Mixed Comp. (4.26) (3.62) (-1.61) (-2.03) 0.772*** 0.702*** 0.535*** 0.308 Foreign Comp. (33.38) (28.62) (3.12) (1.51) region==Central -0.373*** -0.242*** -0.193* -0.0178 Highlands (-22.84) (-13.06) (-1.76) (-0.15) region==Mekong River -0.0882*** 0.0378*** 0.304*** 0.454*** Delta (-8.24) (3.16) (2.78) (3.87) region==Northern -0.399*** -0.293*** -0.0612 -0.00781 Central and Central (-45.90) (-30.60) (-0.60) (-0.07) Coast region==Northern -0.352*** -0.247*** -0.0675 0.0371 Mountain (-25.19) (-15.66) (-0.62) (0.32) -0.187*** -0.109*** -0.119 -0.0609 region==Ha Noi (-25.59) (-14.08) (-1.13) (-0.55) region==Red River Delta -0.341*** -0.228*** -0.114 -0.0567 (except Ha Noi) (-34.95) (-20.64) (-1.09) (-0.51) Ho Chi Minh City Base region==South East -0.0220** 0.0556*** 0.121 0.427*** (except Ho Chi Minh) (-2.08) (4.61) (1.07) (3.44) Involvement in export 0.360*** 0.517*** and import, lag (33.38) (5.27) 1.012*** 1.164*** 1.215*** 1.254*** Constant (20.88) (21.56) (4.16) (3.75) No. of Obs. 223785 178352 5638 4250 Adjusted R2 0.229 0.244 0.339 0.381 F-Statistics 1546 1308 70 62 Prob > F 0.000 0.000 0.000 0.000 t statistics in parentheses * p<0.10, ** p<0.05, *** p<0.010

108 Productivity and Competitiveness of Viet Nam’s Enterprises 0.242*** 0.207*** -0.394 -0.524** Mixed Comp. Appendix 4. List of Manufacturing Sub-sectors and VSIC (4.26) (3.62) (-1.61) (-2.03) 0.772*** 0.702*** 0.535*** 0.308 codes Foreign Comp. (33.38) (28.62) (3.12) (1.51) region==Central -0.373*** -0.242*** -0.193* -0.0178 Highlands (-22.84) (-13.06) (-1.76) (-0.15) Mã VSIC Manufacturing sub-sectors region==Mekong River -0.0882*** 0.0378*** 0.304*** 0.454*** 10 Food Delta (-8.24) (3.16) (2.78) (3.87) 11 Beverage region==Northern -0.399*** -0.293*** -0.0612 -0.00781 12 Tobacco Central and Central (-45.90) (-30.60) (-0.60) (-0.07) Coast 13 Textile region==Northern -0.352*** -0.247*** -0.0675 0.0371 14 Apparel Mountain (-25.19) (-15.66) (-0.62) (0.32) 15 Leather -0.187*** -0.109*** -0.119 -0.0609 16 Wood region==Ha Noi (-25.59) (-14.08) (-1.13) (-0.55) 17 Paper region==Red River Delta -0.341*** -0.228*** -0.114 -0.0567 18 Printing (except Ha Noi) (-34.95) (-20.64) (-1.09) (-0.51) 19 Coal, petroleum products Ho Chi Minh City Base 20 Chemical products region==South East -0.0220** 0.0556*** 0.121 0.427*** 21 Pharmaceutical production (except Ho Chi Minh) (-2.08) (4.61) (1.07) (3.44) 22 Rubber, plastic Involvement in export 0.360*** 0.517*** 23 Non-metal products and import, lag (33.38) (5.27) 24 Metal products 1.012*** 1.164*** 1.215*** 1.254*** 25 Prefabricated metal Constant (20.88) (21.56) (4.16) (3.75) 26 Electronics, PC, optical products No. of Obs. 223785 178352 5638 4250 27 Electrical equipment Adjusted R2 0.229 0.244 0.339 0.381 28 Unclassified machines, equipment F-Statistics 1546 1308 70 62 29 Motor vehicles Prob > F 0.000 0.000 0.000 0.000 30 Other means of transport 31 Beds, cabinets, tables, chairs t statistics in parentheses 32 Other processing and manufacturing * p<0.10, ** p<0.05, *** p<0.010 33 Repair, maintenance

108 Productivity and Competitiveness of Viet Nam’s Enterprises Volume 1: Manufacturing 109