Report No. 34474-VN Vietnam Business Vietnam Development Report 2006 November 30, 2005 Poverty Reduction and Economic Management Unit East Asia and Pacific Region Document of the World Bank IMF International Monetary Fund JBIC Japan Bank for InternationalCooperation JSB Joint Stock Bank JSC Joint Stock Company LDIF LocalDevelopment InvestmentFund LEFASO Vietnam Leather and Footwear Association LUC Land-Use Right Certificate MARD MinistryofAgriculture andRural Development MDG MillenniumDevelopmentGoal MOC Ministryof Construction MOET Ministryof Educationand Training MOF MinistryofFinance M O H MinistryofHealth MOHA MinistryofHome Affairs M O I MinistryofIndustry MOLISA MinistryofLabor, Invalids and SocialAffairs MONRE MinistryofNatural Resourcesandthe Environment MOT MinistryofTransport MPDF MekongPrivate Sector Development Facility MPI MinistryofPlanningand Investment NBIC National Business Information Center NGO Non-Governmental Organization NOIP National Office for IntellectualProperty NPL Non-Performing Loan NPV Net PresentValue ODA Official Development Assistance OOG Office of Government OSS One-Stop Shop PCF People's Credit Fund PCI Provincial Competitiveness Index PER-IFA Public ExpenditureReview-IntegratedFiduciary Assessment PFMB Protection Forest Management Board PMRC Prime Minister Research Commission SAV State Audit of Vietnam SBV State Bank o fVietnam SCIC State Capital InvestmentCorporation SEDP Socio-Economic Development Plan SFE State Forest Enterprise SIDA Swedish International Development Agency SME Small and MediumEnterprise SOCB State-Owned Commercial Bank SOE State-Owned Enterprises SPS Sanitary and Phyto-Sanitary SSC State Securities Commission UAIC Union of Associations of Industryand Commerce UNCTAD UnitedNations Conference for Trade andDevelopment UNDP UnitedNations Development Program UNICEF UnitedNations Children's Fund U.S. The United States USAID U S Agency for International Development USBTA US-Vietnam Bilateral Trade Agreement VAS VietnameseAccountingStandard VASS VietnameseAcademy of Social Sciences VAT Value Added Tax VBARD VietnamBankfor Agriculture andRuralDevelopment VBF VietnamBusiness Forum VBP VietnamBank for the Poor VBSP VietnamBankfor SocialPolicies VCA Vietnam CooperativeAlliance VCCI Vietnam Chamber of CommerceandIndustry VDG VietnamDevelopment Goal VGCL Vietnam GeneralConfederationofLabor VHLSS VietnamHouseholdLiving Standards Survey VITAS Vietnam TextilesAssociation VLSS VietnamLiving StandardsSurvey VNPT VietnamPost andTelecommunicationsCorporation VSI Vietnam Social Insurance WHO WorldHealth Organization WTO World Trade Organization W V S WorldValues Survey This report was prepared in partnership by the Asian Development Bank (ADB), the Department for International Development (DFID) o f the United Kingdom, the Japan Bank for International Cooperation (JBIC), the Mekong Private Sector Development Facility (MPDF) of the InternationalFinance Corporation, the UnitedNations DevelopmentProgramme (UNDP),the United States Agency for International Development (USAID) and the World Bank. Some o f these donors conducted major analytical efforts which provided very useful inputs in the preparation o f this report. The Making Markets Workfor the Poor project by ADB, the Vietnam CompetitivenessInitiative byUSAID, and the Investment Climate Assessment by the World Bank are outstanding in this respect. The donors involved in the preparation of this report also contributed key inputsthrough their engagement in a range of specific pieces of analytical work. Examples include the regular evaluations o f the Joint Vietnam-Japan Initiative by JBIC, the Private Sector Discussion Series by MPDF, the work on the impacts of accession to the World Trade Organization, studies produced by specialized United Nations agencies such as the International Labour Organisation (ILO), and the support to data development at the General Statistics Office (GSO) by the World Bank and other donors. The donors involved in this partnership effort also provided inputs and guidance to the overall effort through a Steering Committee comprising Kanopkan Lao-Araya (ADB), Alan Johnson (DFID), Yuho Hayakawa, (JBIC), Quynh Trang Phuong Nguyen (MPDF), Jonathan Pincus (UNDP) and DennisZvinakis (USAID). The preparation of the report involved consultations with Vietnamese researchers and practitioners who participated on a personal capacity. Their inputs and feedback were coordinated through a Reviewing Committee comprising Dr.An Van Dinh (Central Institute for Economic Management, CIEM), Cuong Manh Nguyen (Ministry of Labor, Invalids and Social Affairs, MOLISA), Doi Duc Do (Ministry of Natural Resources and the Environment, MONRE), Lawyer HuynhHuu Tran (Vietnam Chamber of Commerce and Industry, VCCI), Lan Chi Pham (Prime Minister Research Commission, PMRC), Son Kim Dang (Ministry o f Agriculture and Rural Development, MARD), Dr. Thang Nguyen (Vietnamese Academy of Social Sciences, VASS) andThuy DinhPham (GSO). Several individual researchers and teams were incharge of producing or coordinating key pieces o f analytical work that were used as inputs for the report. Among then are Loren Brandt (University o f Toronto) on land, Amanda S. Carlier (World Bank) on the investment climate assessment, Paulette Caste1 (consultant) on social insurance, Van Nhu Dang (VASS) on social impacts o f global integration, Emilio Fukase (World Bank) on capital markets, Kamran Khan (World Bank) on provincial finance, Tien Van Nguyen (VASS), Thuy Dinh Pham (GSO) and Rob Swinkels (World Bank) on the rural investment climate assessment, Hang Thi Thu Pham (VCCI) on women entrepreneurs, Martin Ravallion (World Bank) and Dominique van de Walle (World Bank) on the impacts of land market development, Thomas A. Rose (World Bank) on the financial sector assessment, William Smith (ADB) on land markets, Cuong Tien Tran (CIEM) on the post-equitization study, Wim Vijverberg (University o f Texas at Dallas) on household enterprises and Michael Warlters (World Bank) on infrastructure. The team in charge of formally writing the report was led by Martin Rama and included Noritaka Akamatsu, Viet Tuan Dinh, Quy Toan Do, Quang Hong Doan, Daniel Riley Musson, Dzung The Nguyen, Minh Van Nguyen, Duc Minh Pham, Hoa Thi Mong Pham, James Seward, Vivek Suri, Rob Swinkels, Son Thanh Tran and Carolyn Turk from the World Bank. The writing team also reliedon a vast range of documents and studies, producedby local and foreign experts. Findings and recommendations from those studies are reproduced throughout the report. It is difficult to give each of the experts involved, individually, the credit they deserve. Their studies are listedinthe bibliographic section ofthe report. Huong Thanh Hoang (National Economics University) helped with the processing of household survey data, Phuong Thu Nguyen (VASS) with data from the investment climate survey, Tuyet Anh Pham (VASS) with data from enterprise censuses, Sa Kim L e (VASS) with the comparative analysis of studies on impacts of global integration, and An The Ngo (Hanoi Agricultural University) with maps. Nha Thi Vu (Vietnam Development Information Center) was in charge of bibliographicalresearch and referencing. Support in the processing of the document was providedby a World Bank team includingDung Thi Ngoc on editing, Ha Thanh Hoangon publishing, andHangThuNguyenandHedwigAbbey on administration. Overall guidance was provided by Homi Kharas and Klaus Rohland from the World Bank. Steve Price-Thomas (Oxfam GB) and Mary Hallward-Driemeier(World Bank) were the peer reviewers. The report was also reviewed by the International Monetary Fund (IMF). Commentsand suggestionsby numerous colleaguesare gratefullyacknowledged. TABLE OF CONTENTS Acknowledgements ExecutiveSummary.. ............................................................................... i I ANEMERGINGECONOMY . ................................................................. 1 1. Businessesof all Sorts.................................................................... -3 2. Business and Development............................................................... 18 3. Efficiency and Competitiveness ......................................................... 29 4. The Investment Climate.................................................................. 42 I1. KEY MARKETSAND INPUTS ............................................................. 57 5. Banking and Finance.................................................................... -59 6. The Land Market.......................................................................... 72 7. The Labor Market......................................................................... 84 8. Infrastructure Services.................................................................... 98 I11 . POLICIES FORBUSINESS ............................................................... 109 9. Global Integration., ...................................................................... 111 10. Domestic Reforms ....................................................................... 123 11. The Local Dimension.................................................................... 137 12. Social Impacts............................................................................ 148 Bibliography...................................................................................... -159 Statistical Appendix Boxes Box 1.1: How Many Household Businesses?.......................................................................... 4 24 Box 2.2: Private Sector Champions: Women Entrepreneurs.................................... Box 2.1: Public Sector Champions: Electricity of Vietnam..................................... 28 Box 3.1: LinkingFarmers to World Markets....................................................... 35 Box 3.2: Learning from FDIEnterprises............................................................ 37 Box 4.1: A Market without Property Rights:Real Estate inHCMC............................ 45 Box 5.1: Who Gets Access to Credit?.................................................................................... 61 70 Box 6.1: LandRelations inBlack Thai Villages................................................... Box 5.2: The Stock Exchange versus the Informal Market....................................... 80 Box 7.1: Migrants inthe Textile and Garment Sectors............................................ 90 Box 7.2: Taiwanese Firms in China and Vietnam., ................................................ 92 Box 7.3: Corporate Social Responsibility inVietnam............................................. 93 Box 8.1: Provincial Development Investment Funds............................................. 102 Box 8.2: Common Weaknessesof Vietnamese Feasibility Studies............................. 106 Box 9.1: Incumbents versus Newcomers inthe WTO........................................... 112 Box 9.3: Government Assistance and Enterprise Survival ..................................... Box 9.2: Customs Reform: Trade, Governance or Both?.................................................... 117 120 Box 9.4: Will Customers Abandon Domestic Banks?......................................................... 121 Box 10.1: Business Associations inVietnam .................................................... 124 Box 10.2: A Roadmap for Banking Reform...................................................... 127 Box 10.3: FosteringCompetition inAir Transport?............................................................ 129 132 Box 11.1: The Provincial Competitiveness Index.,............................................... Box 10.4: Social Insurance: Getting the Incentives Right....................................... 144 Box 12.1:Becoming a Pollution Heaven?............................................................................ 151 Tables 11 Table 4.1: BindingConstraints inVietnam and Elsewhere....................................... Table 1.1:Not too Profitable. but Paying Taxes.................................................... 43 Table 4.2: Corruption inthe East Asia Region...................................................... 49 Table 4.3: Which Are The Most Corrupt Government Agencies?......................................... 51 63 Table 5.2: A Small Stock Market.,.................................................................. Table 5.1: Relative Size of Financial Institutions.................................................. -69 Table 6.1: Progress inthe Titling of Agricultural Land............................................ 73 85 Table 8.1: Catching Up with the Neighbors......................................................... Table 7.1: The Structure of the Labor Force......................................................... 99 Table 8.2: Investment in Infrastructure as a Share of GDP...................................... 100 101 Table 9.1: Trade Barriers and the Effective Protection Rate.................................... Table 8.3: Where Do the Resources Come From?................................................................ 114 Table 9.2: Key Indicators of Trading Companies................................................. 118 138 Table 12.1: Economic and Social Impacts in Selected Sectors................................. Table 11.1:Striking Contrasts: from Northto South............................................. 154 Table 12.2: The Safety Net for Redundant SOE workers ........................................ 157 Figures Figure 1.1: An Entrepreneurial Country............................................................... 5 Figure 1.3: ...but Unevenly Spread across the Country............................................ Figure 1.2:Enterprise Registration i s on the Rise................................................... 7 9 Figure 1.4: A Declining State Share of the Economy.,............................................ 10 Figure 1.6: ... but Concentrated inFew Areas................................................... Figure 1.5: FDIRegaining Momentum., ............................................................. ;..14 13 Figure 1.7: Employment by FirmSize.,............................................................ -15 Figure 1.8: Growth, Decline and Exit: the Implicit Trends....................................... 17 Figure 2.1: The Private Sector as an Employment Engine.,...................................... 19 20 Figure 2.3: Investment by Institutional Sector., .................................................... Figure 2.2: The "Cost" of Creating a Job............................................................ 21 Figure 2.4: Productivity Gains at the Aggregate Level............................................ 22 Figure 2.5: Productivity Gains at the Enterprise Level............................................. 23 Figure 2.6: Business Activity and Poverty........................................................... 25 Figure 3.2: ...or One of the Most Attractive Destinations?.................................................... Figure 3.1: A Bad Place to do Business., .? ......................................................... 30 31 Figure 3.3: More like China, less like ASEAN...................................................... 33 Figure 3.4: Key Performance Indicators after Equitization....................................... 38 Figure 3.5: Who GainedPower with Equitization?................................................................ 39 Figure 3.6: Vietnam's Strengths and Weaknesses.................................................. 40 Figure 3.7: Vietnam Ratings inInternationalPerspective......................................... 41 Figure4.1: Government Agencies: Friends or Foes?............................................................. 47 Figure 4.2: Spendingon UnofficialPayments and "Gifts" ...................................... -50 Figure 4.3: BindingConstraints by Enterprise Ownership.,...................................... 52 Figure 4.3: BindingConstraints inRural Areas..................................................... 53 Figure 5.1: RapidFinancialDeepening.............................................................. 60 Figure5.2: BadLoans as a Share of Total Credit................................................... 66 Figure 5.3: Public Debt over Time.................................................................... 68 Figure 6.1: LandRegistrationacross Provinces..................................................... 74 Figure 6.2: Property Prices for Urban and Rural Land............................................. 76 Figure 6.3: Rental Prices for Upscale Housing and Offices.,..................................... 77 Figure 7.2: Determinants of Labor Earnings........................................................ Figure 7.1: Where Are the Jobs? And how muchDo They Pay?.......................................... 86 88 Figure 7.3: Number of Strikes by Enterprise Ownership.,........................................ 91 94 Figure 7.5: A gradual Formalizationof Employment............................................. Figure7.4: Key Characteristics of the Unemployed................................................ -96 Figure 8.1: InfrastructurePriorities: the Views of Entrepreneurs................................. 99 Figure 10.1: Tax Revenue and the Emergence of Small and MediumEnterprises...........134 Figure 11.1:BindingConstraints across Regions................................................. Figure 10.2: Corruptionby Agency: the Good, the Bad and the Ugly......................... 136 143 Figure 11.2: Provincial Governance and Enterprise Registration............................... 145 Figure 12.1: Changes in Sectoral Output due to Global Integration............................ 153 Figure 12.2: Changes inReturns to Capital and Wages.......................................... 155 Figure 12.4: Budget Transfers to Provinces and Poverty Reduction........................... Figure 12.3: Reform Impacts versus External Shocks............................................ 156 158 EXECUTIVE SUMMARY Business development has been one o f the main forces behind rapid poverty reduction in Vietnam. Together with the redistribution o f agricultural land, and the broad coverage o f social services, it allowed a large fraction o f the population to engage in more productive occupations and raise their living standards. The reform process launched almost two decades ago unleashed an enormous entrepreneurial energy. Rural households started commercializing their agricultural output, and running small business activities. Foreign investors were attracted in large numbers, and state-owned enterprises (SOEs) initiated a long (albeit gradual) restructuring process, paving the way to the development o f a multi-sector economy. Private sector enterprises started formalizing and expanding, in a process that caught speed after 2000; they account now for 33 percent o f the value o f production in manufacturing. Together with thousands o f foreign companies and several million household businesses, they provide wage employment to 21 percent o f Vietnam's labor force. Massive job creation has allowed absorbing 1.4 to 1.5 million entrants to the labor market every year, and offered a way out o f farming for the rural population, and especially for young women. Over the last decade, the average wage has grown at around 10 percent per year in nominal terms, or roughly 7 percent in real terms. The enormous opportunities created through this process have led to an inclusive growth pattern. While real GDP per capita has increased by 5.9 percent per year since 1993, the Gini index (measuring inequality) has increased only slightly, reaching 0.37 in 2004. During that period, the poverty rate fell from 57 to less than 20 percent. This virtually unmatched performance should not hide the fact that businesses are strugglingwith important constraints. Insufficient availability o f finance, difficulties inaccessing land and continuous gaps in infrastructure services (in spite o f enormous investment efforts) are among the most important obstacles identified by entrepreneurs. In a booming labor market, retaining qualified personnel and finding the skills required to move up the ladder are also perceived as barriers to business development. As a result o f these constraints, the domestic private sector remains dominated by small enterprises. In between a myriad household businesses and a few thousand large SOEs and foreign companies, there are not so many small and medium enterprises, and only a handful o f domestic private firms have made it to the top. Sustaining business development in Vietnam requires the completion o f the structural reform agenda. Fully developing the land market, restructuring the financial sector, managing state assets in a more efficient and transparent manner, mobilizing resources for infrastructure development, are the key priorities in this respect. Further integration with the world economy, especially through the accession to the World Trade Organization (WTO), i s bound to lock-in some of these changes, and level the playing field between domestic and foreign enterprises. But there i s also a complementary reform agenda, aimed at leveling the playing field between the domestic private sector and SOEs and mobilizing capital (both public and private) in an efficient way. Global integrationand domestic reforms are neededto sustain rapid economic growth while avoiding the accumulation o f large contingent liabilities for the government. As Vietnam fully emerges from the grim poverty that prevailed barely one decade ago, business development also holds the key to building a prosperous and inclusive society. Targets for the future are still phrased, much too often, in terms o f growth rates o f GDP, thus evoking a mechanical accumulation process. But Vietnam can by now hold a higher ambition: that of becoming a middle-income country. This will entail going beyond structural reforms and laying the foundations of a modern market economy. Introducing competition and proper regulation in infrastructure services, modernizing tax administration, reforming the legal andjudiciary systems, reducing corruption, improving governance at local levels, are all part of a second generation of reforms that need to be put on track for Vietnam to move up to the nextphase. These reforms are bound to further boost business development and living standards more generally. But they should also be accompanied by reforms aimed at keepinggrowth inclusive. Effective safety nets and budget allocation norms can help households and regions cope with adverse shocks, or increasing disparities, as the economy restructures and Vietnam becomes a global player. Inthe longer term, modernizingthe social sectors holds the keyto preservingsocial inclusion. A solid performance In less than two decades, since the beginning of economic reforms, Vietnam managed to nurture an unusually diverse business sector. Building on a thriving, entrepreneurial culture, roughly every other household runs by now a small business of one kind or another. Owing to a state-led development strategy after independence, the government still owns thousands o f SOEs, some very big. Vietnam has also been one of the largest recipients o f Foreign Direct Investment (FDI) inthe entire world, inrelative terms, with the value of projects registered getting at times close to a tenth of the country's Gross Domestic Product (GDP). And over the last five years, the registration o f private sector enterprises has boomed, reflecting both the formalization o f existing businesses but also, to a large extent, the creation of new ones. This remarkable diversity i s acknowledged as one of the key features, if not one of the strengths, of Vietnam's economic transition. But the very diversity of Vietnam's enterprise sector is a challenge for economic analysis and policy making. Different types of businesses contribute differently to economic growth and poverty reduction. For government policies to be effective, it i s necessary to make sense of the main buildingblocks o f Vietnam's business sector, and how they fit together. When the various data sources available are analyzed and brought together, it appears that there is indeeda "missing middle" in the distribution of firms by size. But it is gradually being filled in, whereas enterprises at the low end are becoming increasingly formal. Whether the performance o f Vietnam's business sector compares favorably or unfavorably to that of other countries, especially in the East Asia region, i s more controversial. Competitiveness assessments, relyingon the characteristics of the regulatory framework inwhich firms operate, put Vietnam in the bottom half o f the world distribution or close to the middle at best. On the other hand, direct assessments by foreign companies, under the form o f opinion surveys or actual investmentsflow, rank it in the top quarter, or even higher. Deeper insightscan be gained by analyzing the way inwhich Vietnam's businesses actually operate and participate in the global economy, either directly or indirectly. Those analyses show that Vietnam's pattern of integration in world trade is closer to that of China than to that of other countries in the Association of South East Asian Nations (ASEAN). But they also reveal a still insufficient integration of domestic companies in global value chains. Total factor productivity i s growing very rapidly across the board; but growth is faster in foreign companies than in domestic ones, regardless of their ownership. Productivity spillovers occur between foreign and domestic companies, especially private ones, but labor turnover and outright imitation seem to be more important channels of transmission than business-to-business transactions. The equitization of SOEs, in spite of its limitations, appears to be contributing to productivity gains, as it establishes a more arms-length relationship with government authorities. .. 11 For Vietnam's business sector to continue its expansion, it is necessary to remove several important constraints currently faced by enterprises. A recently completed investment climate survey (ICs), with a representative sample made of domestic private enterprises, SOEs and foreign companies, goes a long way towards identifying those constraints. For more than one third o f the respondents, access to credit is reported as a major or severe obstacle to business growth. Apart from foreign companies, which do not appear to face credit rationing, the importance attached to this constraint i s similar for the other types o f enterprises. Access to land i s the second most important constraint. It is seen as particularly severe by foreign companies, but not surprisingly it does not affect SOEs much. Insufficient skills and education o f the workforce, and poor transportation infrastructure, rank third and fourth. In all four cases (except maybe for skills) the severity o f the constraints i s significantly higher in Vietnam than in the rest o f the East Asia region or inthe rest o f the world. On the other hand, some o f the issues emphasized by competitiveness rankings, such as the legal system, bureaucratic procedures and corruption, are downplayed by respondents. The fraction rating any o f these issues as a severe or major constraint to business growth inVietnam i s significantly lower than in either the rest o f East Asia or the rest o f the world. The ability to conduct business-to-business transactions on a trust basis, or using rudimentary but reliable enforcement mechanisms, may explain the low importance attached to the legal system. The considerable simplification of procedures introduced in2000, and reinforced through mechanisms like one-stop shops (OSSs), could account for the downplaying of bureaucracy and red tape. The lack o f importance attached to corruption is more unexpected. This result i s further probed using a variety o f sources. They all yield a consistent picture, one where corruption directly affecting businesses i s quite prevalent, but petty. More common in agencies like the traffic police, customs, tax administration or land registration offices, it translates into many small bribes, accounting for about 0.7 percent o f total sales, which i s less than in other countries at a similar development level. This i s not to ignore other modalities o f corruption or collusion, as in SOEs and in public investment projects; because these other modalities may not affect businesses directly, they cannot be appropriately captured through ICs. But overall, these findings should give pause to reassess commonly held, largely anecdotal views on corruption in Vietnam and stimulate hrther empirical research. Binding constraints The prominence o f access to finance as a constraint to business development seems in contradiction with the substantial financial deepening achieved by Vietnam over a relatively short period o f time. The number of active savings and loan accounts i s high for a country at its development level, whereas banking credit has been growing steadily; even too fast, probably, given current deficiencies in the assessment o f credit risk. But perceptions and facts can be reconciled by analyzing the allocation o f credit inVietnam. Small-scale loans are easily available to farmers and small businesses, probably reflecting government policies towards poverty reduction. But collateral i s needed for more sizeable enterprises to get access to credit, and it i s valued in a very conservative way. Still limited progress in the issuance o f land-use right certificates (LUCs) thus compounds credit rationing. Heavy reliance on collateral i s the flip side o f limited risk assessment by lenders, and especially by the four major state-owned commercial banks (SOCBs) which account for three quarters of total credit. Their insufficient profit orientation and the interference by government authorities, especially at local levels, have led to the accumulation o f a sizeable volume o f non-performing loans (NPLs) in the banking system. The poor quality o f many loans creates a substantial fiscal burden for the government, aggravated ... 111 by the absence o f effective mechanisms to seize the assets o f bad debtors. As for formal capital markets, they are still too under-developed to channel a sizeable volume o f resources to new projects. But thriving informal transactions on unlisted stocks raise the prospect o f faster capital market development, provided that proper regulations are in place. A land market has already emerged in Vietnam, despite the still partial development o f formal titling. Many transactions take place inthe absence o f proper legal documentation, simply because local officials know which piece o f land or property "belongs" to whom, and arbitrate in case o f conflict. From this perspective, the absence o f formal property rights has not been incompatible with market development. But land titling i s bound to increase efficiency, and it should also provide a more sound foundation for the development o f a real estate market. However, the dynamics o f the property market in recent years also show that titling alone is not enough to ensure efficiency. Price "bubbles" in real estate need to be kept under control by making speculative transactions more burdensome, as the Vietnamese government did recently. And some o f the most difficult issues, such as facilitating the consolidation o f agricultural land, restructuring state forest land, supporting community-based land management among ethnic minorities, and recovering idle land from urban SOEs, will still require specific policies, going beyond the issuance o f LUCs. As Vietnam develops and urbanizes, one o f the most difficult challenges will be the conversion o f agricultural land into residential and industrial land, on a massive scale. Success will critically depend on fighting corruption in land conversion and properly compensating the affected populations. Enormous progress has been accomplished in recent years in infrastructure development Vietnam i s by now investing about one tenth o f its GDP in power, transport, telecommunications and water and sanitation projects; as a result, it i s rapidly catching up with its neighbors in terms o f availability and cost of services. Electrification and telephone penetration have seen the most remarkable improvements, and the road network has expanded considerably as well. Still, enterprises in Vietnam complain about insufficient transport infrastructure, and excessively expensive electricity and telephone services. Making further progress in infrastructure development requires a diversification o f funding sources and improved transparency in resource mobilization, especially at local levels. The adoption o f sound regulatory frameworks, facilitating cost recovery and promoting competition, could also help attract private participation in infrastructure and further contribute to business development inVietnam. There have been a few promising initiatives in this respect, across most infrastructure sub-sectors, especially in terms o f increasing competition among suppliers. However, most business participants are still SOEs. Looking forward, private sector participation cannot be relied upon to address, alone, the infrastructure needs o f Vietnam in the coming years. N o r even to cover a large share o f those needs. In this area, business development crucially depends on the quality o f public investment, and on the appropriate pricing o f the services it provides. Unfinishedreforms Integration inthe world economy has been one o f the main drivers o f economic reform in Vietnam. From participation in the ASEAN Free Trade Area (AFTA) to the implementation o f the USBTA, it has made markets more competitive and has forced domestic enterprises to improve their productivity. But the impending accession to the WTO takes the agenda even farther, because o f the encompassing nature o f the commitments it implies, both at the border and behind the border. Much the same as participation in AFTA, it will lead to a reduction in trade barriers. In spite o f the considerable progress accomplished so far inremoving import quotas and reducing tariffs and subsidies, protection rates remain high in Vietnam. Further liberalization iv should lead to increased efficiency. The process needs to be supplemented by substantive improvements in trade logistics. Increasinglycongested ports, with relatively expensive services, and a customs department focused on control (legally or otherwise), rather than on trade facilitation, are obstacles to business development. Much the same as adherence to the USBTA, accession to the WTO i s about institutional changes, from a more level playing field among enterprises, to competition in key services such as banking, to improved sanitary standards to strengthened intellectual property rights. These changes will boost productivity inthe medium to long term, but they will represent an important challenge for Vietnamese enterprises in the short term. If global integration is one of the main drivers of economic reforms in Vietnam, their engine i s clearly homegrown. Since the beginning of the renovation process the government has been steadily pushingfor change across all policy areas, with the objective o f building a market economy with a socialist orientation. The five-year plans articulating this vision are also changing along the way. They rely less on quantitative targets for material production, and focus more on policies and development outcomes. Consultation with stakeholders at large is also becoming part of the process, and business associations are having increasingly more clout. Improving the mobilization of capital, both private and public, is one o f the main components of the unfinishedreform agenda. On the private front, the banking system needs to be thoroughly reformed. This involves transforming SOCBs into autonomous profit-making institutions and creating a modern central bank, in charge of banking supervision and monetary policy. Banking reform should lead to a more efficient allocation of credit and mitigate the potential burden of bad loans on the government budget. On the public front, the equitization process needs to be completed, and the SOEs remaining in state hands need to be managed at arms length from the ministries regulating the sectors they operate in, so as to promote competition throughout the economy. Strengthening governance i s another key component o f the unfinished reform agenda. This includes completing the reform of public financial management, starting that of tax administration, settingup mechanisms to manage public debt at all levels, and increasing transparency in public procurement. Closely related to this, there i s also a need to fight corruption, especially at atime when rapid economic growth will make the opportunities for graft grow faster than government systems improve. And the legal andjudiciary reform agenda, which is much influenced by the implementation of the USBTA and the accession to the WTO, needs to be pursued vigorously. Transactions based on trust and relatively simple enforcement mechanisms will become increasingly ill-suited to the needs of businesses as the economy develops and domestic firms grow. In a country as decentralized as Vietnam, some of the most important changes will have to take place at the local level. Increased delegation of decision-making power to local authorities has led to vast gaps in business buoyancy, and more generally in economic and social development, across Vietnam. Contiguous provinces, with similar endowments but different commitment of the local authorities to economic reform, are often worlds apart in terms o f enterprise registration, investment by foreign companies and the availability o f wage employment. Rigorously measuring the quality o f governance (either locally or at the national level) i s not easy, but benchmarking provinces on key aspects potentially affecting business activity i s a step inthe right direction. The modernizationof provincial planning processes, along the lines observed at the national level, relying on consultation with the local business community, should encourage further improvements. Combined with appropriate budget allocations for provinces with stronger needs, the promotion of local leaders who succeed in fostering growth and reducing poverty would provide a powerful incentive for change. V Keeping growth inclusive The next five-year cycle offers the prospect for Vietnam to complete its transition, across several dimensions: towards increasing reliance on market mechanisms, towards full membership inthe global economy, and towards middle-incomecountry status. But success inthese multiple transitions will also require that business development benefitthe population at large, as it has so far. The labor market and the social sectors are fundamental inthis respect. The surge in business activity prompted by economic reforms has led to a en enormous increase in the demand for labor. This has translated into a double mobility: occupational, from agriculturaljobs to non-farm employment; and geographical, from rural to urban areas. But wage employment i s unevenly distributedacross the country. And the same i s true of labor earnings, which vary considerably from province to province. Disparities in earnings have been reduced with economic growth, as also has the earnings gap between men and women. But the labor market reward to education has increased, opening the prospect of a different kind of inequality as Vietnam develops. Importantly, in spite o f the enormous structural transformation, unemployment has not been a major issue. Most of the jobless are young, relatively educated entrants to the labor market, although over-staffing i s still common in the state sector. It i s rather the excessive turnover of qualified workers that creates a problem for business. But the main weakness o f Vietnam's labor market is the social protection system built on it. Traditionally conceived for public sector workers only, this system still needs to complete the transition to a market economy, so as provide affordable insurance against major risks to workers who are increasinglymobile, occupationally and geographically. Further integratinginto the world economy, completingthe unfinishedreform agendaand strengthening governance at local levels, are all bound to foster business activity and increase economy efficiency. Ina way, these developments holdthe key for Vietnam to become a middle- income country. But an important issue is whether there will be a social or an environmental price to pay for increased efficiency. Increasingly relyingon market mechanisms will unleash the potential of the most productive members of society, but it could also harm some of the weakest ones. And even a perfect market could be unable to address environmental externalities, so that rapid growth would lead to increasing pollution and degradation. Based on a systematic review of the available evidence, the foreseeable adverse effects do not appear to be large; except maybe for the environment. This calls for a more systematic use o f enforceable environmental standards. But potentially larger social impacts might simply be unforeseeable with the relatively rudimentary analytical tools at hand. It is therefore important to put in place effective mechanisms to rapidly cope with the unforeseen. Assistance to workers who lose their jobs and budget transfers, from the provinces benefiting the most from economic reforms to those adversely hit, could ensure that inthe quest for prosperity inclusion i s not sacrificed. Looking forward, social services need to be upgraded, as part o f what could be seen as a second generation o f economic reforms. Vietnam managed to build inclusive basic health and education systems inthe period under central planning, reaching a vast majority o f the population and delivering social indicators comparable to those o f middle-income countries. But these systems did not cope well with the transition to a market economy, and they were anyway ill- suited to address the more complex and expensive needs o f an increasingly more prosperous population. The challenge now i s to combine the efficiency o f market mechanisms with the broad coverage that characterized education and health services in the period under central planning. Success inthis combination could lay the foundations o f modern universal systems. In a paradoxical way, the relatively good quality of targeting mechanisms makes it possible to use them as a tool to attain universal coverage. Not by providing transfers and subsidized credit, as vi they still do to some extent, but rather by giving the poor access to services which are otherwise deliveredfor a fee. Targeting and "socialization" policies can then go hand inhand. But proper attention to incentives is neededfor this combinationto work. The links between this social inclusion agenda and business development are stronger than it may seem at first. Businesses should become important suppliers o f social services over time, as public funding should not be confused with public provision. Tertiary education, health care services and, maybe in a not-so-distant future, pension programs, can all be supplied by non- state organizations, including private enterprises, cooperatives and non-government organizations (NGOs). Meanwhile, social insurance, ifproperly reformed, could accumulate a massive amount o f resources before the active workers of today start retiring. Those resources will need to be invested; ifmanaged in a sound and transparent manner, they could make a major contribution to capital accumulation. Third, and most obviously, good human resources cannot be dissociated from a healthy and educated population. The development of the education system, inparticular, i s the key to tap all the talent available in each generation of Vietnamese children, and to build up the skills the business community needs. Last but not least, inclusive development is at the root o f social stability. And this has been one of the main advantages o f Vietnam over its competitors. vii ... V l l l PARTI: ANEMERGING ECONOMY 1. BUSINESSESOF ALL SORTS Few countries can match the staggering diversity o f Vietnam's business sector. Building on a thriving, entrepreneurial culture, roughly every other household runs a small business o f one kind or another. Owing to a state-led development strategy after independence, the government still owns thousands o f SOEs, some o f them very big. Vietnam has also been one o f the largest recipients o f FDI in the entire world, in relative terms, with the value o f projects registered getting at times close to a tenth o f the country's GDP. And over the last five years, the registration o f private sector enterprises has boomed, reflecting both the formalization o f existing businesses but also, to a large extent, the creation o f new ones. This remarkable diversity o f the business sector is acknowledged as one o f the key features, if not one o f the strengths, o f Vietnam's economic transition. In 1986, the Sixth Communist Party Congress adopted a program o f major reforms, to abolish the system o f "bureaucratic centralized management based on state subsidies" and move to "a multi-stakeholder, market oriented" economy, including a role for the private sector. In 2001, the NinthParty Congress explicitly recognized FDI as one o f the sectors inthe economy. In2006, the Tenth Party Congress could well admit private sector entrepreneurs as Party members, thus stressing the equal status and legitimacy acquired by businesses o f all sorts. But the very diversity o f Vietnam's enterprise sector i s a challenge for economic analysis and policy making. Different types o f businesses contribute differently to economic growth and poverty reduction. For government policies to be effective, it is necessary to make sense o f the main building blocks o f Vietnam's business sector, and how they fit together. When the various data sources available are analyzed and brought together, it appears that there i s indeed a "missing middle" inthe distribution o f firms by size. But it is gradually being filled in, whereas enterprises at the low end are becoming increasingly formal. Household businesses It may be surprising to learn that the Vietnamese population is even more business- oriented than the Chinese. This was one o f the findings o f the World Values Survey (WVS) conducted in more than 65 countries across all continents. The W V S was implemented for the first time in Vietnam by the Institute o f Human Studies, in 2001. I t s questionnaire aims, among other objectives, at assessing the respondents' attitudes toward a market economy. One question uses a ten-point scale to elicit support for private business ownership (10 points in the scale) versus government ownership (one point). The mean score among Vietnamese respondents was 5.6, compared to 4.2 among Chinese. Overall, respondents in these two countries are more attached to government participation than citizens in more established market economies. However, the magnitude o f the differences i s small, with the attitude o f Vietnamese respondents beingquite close to that o fthe Japanese. Expectedly, younger respondents are more favorable to private ownership and competition than their elders. This can be seen by classifying W V S respondents according to their birth age, with 1975 as the cutoff point. In this respect, it i s worth keeping in mind that roughly two thirds o f the Vietnamese population was born after reunification. Educational attainment is also associated with a stronger support for the market economy. Results are somewhat more unexpected when classifying respondents by region of residence. In many ways, the North and the South of Vietnam are culturally different. Southerners originally came from the North but when conquering the new land they relied more on individual initiative than on collective, village-level decision making. Southerners also used to be immersed ina market economy beforereunification. Yet respondents from the North are more positive towards the market than people from the South. This may be due to economic reforms having had a bigger impact on everyday life inthe North than inthe South. Box 1.1: How Many HouseholdBusinesses? According to GSO's Vietnam Household Living Standards Survey (VHLSS) of 2004, there are around 9.3 million non-farm household enterprises. But the Annual Household Business Survey (AHBS) of 2004, also conductedby GSO, puts the figure at about 2.9 million. At 6.4 million, the gap betweenboth sources i s very large. Such understanding should also help assess the consequences for the measurementof GDP, whichrelies onthe AHBS. Within the VHLSS 2004, a non-farm household enterprise covers all kinds o f income-earning activities that are on own account and are unrelatedto the production or sales of agricultural products (inits various forms). The VHLSS gathers information referringto the 12 months prior to the visit of the enumerator, thus including low-seasonactivities by farming households. The AHBS, on the other hand, refers to its unit of observationas an "individual business establishment." A practical way to examine the gap betweenthe AHBS and the VHLSS is to assume that the latter offers an exhaustive account of all non-farm self-employment activities, regardless of how insignificant they may be, and to impose the AHBS inclusion conditions on those activities. The objective of the VHLSS is to measure income, poverty and living standards inhouseholds, thus making it plausible that non-farmself-employment activities are indeedcoveredwell. Six inclusion conditions are applied to the VHLSS set of household activities: (i)the revenue source i s valid, (ii)responses are complete, (iii)enterprises must be in operation for at least three months during the year, unless they are new, (iv) enterprises must have a fixed location, unless they are inthe transport sector, (v) enterprisesoperate a significant part ofthe month, and (vi) enterprises are in operation on the first day of October. These inclusion conditions reduce the total number of household enterprises to 6.1 million. Therefore, there are 3.2 million household enterprises not accounted for by the AHBS. The under-estimation of the number of household businesses by the AHBS has implications for the measurementof Vietnam's GDP. The average value added generatedover a year by a household enterprise, according to the VHLSS, is about 15.5 million dong. Multiplying this figure by 6.1 million yields the total value added of the household business sector in Vietnam, equivalent to roughly 13 percent of total GDP. Given that 3.2 out of the 6.1 million household enterprises are unaccounted for by the AHBS, national accounts could underestimate the value added of the sector by the same proportion. This would imply that Vietnam's GDP is roughly 7 percent higher than officially reported. But the under-reporting is probably less, as the 3.2 million household enterprises not covered by the AHBS are presumablythe smallestones. Source: Basedon Wim Vijverberg (2005). 4 This nationwide support for private initiative is fully reflected in the extent to which Vietnamese households own and run small businesses. Assessing exactly how many household businesses there are in operation i s not straightforward, with different survey instruments providing different answers (Box 1.1). But a plausible estimate, including the business activities o f farming households in the low season, puts their number at 7.4 million, or roughly one every other household. An international comparison might help put Vietnam's dynamism in perspective. Statistics on the number of household businesses are not readily available for many countries; understandably so, given the inherent measurement difficulties. But more comparable statistics exist on the fraction of rural self-employment accounted for by non-farm activities. The analysis o f those statistics shows that the fraction increases with a country's level of development (Figure 1.1). What i s remarkable about Vietnam is the much faster speed at which this has been happening since the beginningof economic reforms. Figure 1.1:An EntrepreneurialCountry 80 70 - A Costa Rica 60 50 40 30 20 10 0 0 2,000 4,000 6,000 8,000 10,000 GDP per capita in PPP US$ Source: Own estimates, based on data from GSO and Jean 0. Lanjouw and Peter Lanjouw (2001). Employmentfigures are basedon mainoccupation. Household businesses are more prevalent inthe lowlands. In2004, they accounted for 28 percent of employment in the Red River delta region, and for 28 percent in the Mekong delta region. Employment here means having a household business as either a main or second source o f income. Over time, however, the contribution of non-farm self-employment to household income declined across all regions, except the South Central Coast and South East. In 1993, almost 30 percent o f the employed population o f Vietnam worked in a household business. By 2004, the fraction had declined to roughly 25 percent. The decline can be explained by the substantial expansion of wage employment, from 26 percent of the total in 1993 to 31 percent in 2004. However, at any point in time, better-off households have a higher share of income originated in non-farm business activities. This pattern i s consistent across household surveys. Runningahousehold business is associated, other things equal, with higher living standards. 5 The sectoral distribution o f household businesses has not changed much over time. The two most popular industriesare "commerce" and "processing and manufacturing". The former account for more than 40 percent of all household businesses; the latter represent about one quarter. At the other end, "construction" and "mining" account for roughly 1 percent of total household businesseseach. The characteristics of household business entrepreneurs also vary across sectors. The average business owner i s 38 years old and has nearly eight years o f education. Entrepreneurs in construction, "other services" and transportation have comparatively higher levels of education. Household businesses in commerce, food processing and manufacturing are mainly operated or managed by women. Those in construction and transportation industries are largely run by men. Overall, women entrepreneursheaded 55.7 percent of all household businesses in2002, compared to 52.5 percent in 1998 and only 49 percent in 1993. In 2004, businesses in women-headed households are more profitable than those inmen-headedhouseholds. A systematic comparison of data from 1993, 1998, 2002 and 2004 also suggests that household businesses are becoming gradually more "professional". The number of operating days per month, and months per year, has increased. The percentage of enterprises with a fixed location has increased as well. Also, the share of loss-making household businesses was 0.3 percent in2004 compared to 4.8 percent in2002, compared to 7.3 percent in 1993 and 8.2 percent in 1998. The emerging private sector A series of policy reforms gradually laid the foundations for sustained private sector development. The passage of the Foreign Direct InvestmentLaw, in 1986, benefited domestic entrepreneurs, albeit indirectly. A legal framework was first established in 1990, with the approval of the Private Enterprise Law and the Company Law. The Constitution officially recognized the role of the private sector in 1992. The Domestic Investment Promotion Law, promulgated in 1994 and revised in 1998, gave Vietnamese investors access to some o f the incentives available to foreigners. But the most important milestone for private sector development was, undoubtedly, the EnterpriseLaw o f January 2000. On the surface, this piece o f legislation combined the previous Company Law and Private Enterprise Law. In practice, it represented a radical change in approach. Until then private enterprises had been allowed to operate provided that they complied with a series of government approvals and controls. The Enterprise Law, by contrast, protected the right o f citizens to establish and operate private businesses without unnecessary intervention from government officials. The most important innovation introduced by the Enterprise Law was the simplification of registration procedures. The business community refers to this innovation as the permission "to register first, then to check". The Enterprise Law also led to the elimination o f over one hundredbusiness licenses. As a result, it reducedthe time and cost needed to register businesses. It also improved the confidence of business community in local government apparatus, by reducingthe opportunities for corruption. The number o f private enterprises registered every year has been increasingly steadily, ever since (Figure 1.2). A controversial issue, hotly debated at times, is how many of the enterprises registered under the Enterprise Law are actually new. Or even, how many of them do actually exist. This uncertainty has been on occasion usedto cast doubts on the vibrancy of Vietnam's private sector. 6 Data on enterprise registrationare collected by the National Business Information Center (NBIC), under the Ministry o f Planning and Investment (MPI). The information feeding the N B I C database i s provided by the various offices o f the Department for Planning and Investment (DPI) across the country. The enterprise census uses a business register that is updated through a link with the tax database o f the Ministry of Finance (MOF). It is to DPI offices that would-be entrepreneurs submit their applications for company registration and from which they receive a business registration license. On the other hand, data on firms in operation comes mainly from GSO's enterprise censuses, which are conducted annually since 2000. Taken literally, data from enterprise censuses suggests that only half o f the registered firms are actually in existence. Figure 1.2: Enterprise Registrationi s on the Rise.. . 50,000 ' 40,000 30,000 20,000 10,000 0 1999 2000 2001 2002 2003 2004 2005e Source: Own calculations using data from NBIC. Figures reflect new registrations per year. The figures from these two sources are not as inconsistent as it appears, however. N o t all businesses that register actually start operations. Some fail to obtain funding, the market opportunity that others were trying to capture may pass, and the potential partners of yet others may lose interest. In a small survey o f enterprises conducted by the World Bank, about eight percent of the firms that registered never made it to the tax code stage, suggesting that they never actually started operations. More importantly, not all businesses that start operations survive. A three-wave survey o f small and medium enterprises (SMEs) conducted by the Institute for Labor Studies and Social Affairs (ILSSA) found an exit rate in excess o f 15 percent per year inthe early 1990s; but it had declined to less than 10 percent inrecent years. These figures are not unusually high by international standards. But most closures go unregistered in Vietnam, as there are no incentives to report them. Over the years, this turnover accumulates into a sizeable share o f all registered enterprises, accountingfor most o f the gap betweenN B I C and GSO figures. The other controversial issue is how many o f these enterprises are new. A variety o f sources suggest that the answer is: a majority, but by no means all. About 45 percent o f the enterprises in the small survey conducted by the World Bank, mentioned above, were already in existence in 2000, mainly under the form o f household businesses. One o f the most important reasons for them to register was to obtain invoice books for the Value Added Tax (VAT), without which goods and services cannot be sold to government and SOEs. Another survey conducted in 7 2001 by VCCI found that roughly 70 percent of the registered enterprises were truly new. In the ILSSA survey, inturn, the mean establishment year was 1990, with a standard deviation o f 8.3 years. But an in-depth analysis revealed how difficult it was to determine the actual origins o f participating firms, as their backgrounds varied greatly and were in many cases complicated and circuitous. However, the analysis suggested that many of the enterprises had been "upgraded" from household businesses, while others had been registered by people who have been working illegally or unofficially inthe same industryfor years. One reason to believe that many of the registered enterprises are new i s the sustained flow of remittances from overseas Vietnamese. It is estimated about 3 million people of Vietnamese origin, or close to 4 percent of the country's population, live abroad on a permanent basis. Their transfer of resources to Vietnam has grown steadily in recent years, from a mere 35 million dollars in 1991 to over four billion by 2005. While these estimates are subject to some uncertainty, many observers believe that most o f the remittances are already being channeled through formal mechanisms. According to the IMF, a substantial part o f the transfers corresponds to investments in real estate, the capital market and small businesses. Such investments are sometimes reported as unrequited transfers, rather than FDI or portfolio investment. Registeringthe investment in the name of a family member, who is a citizen and legal resident of Vietnam, is a way to avoid the more burdensome administrative requirements or procedures requiredfor nonresidents. Registrationdata can be usedto infer the characteristics o f private enterprises inVietnam: Simple corporate forms, such as sole proprietorships and limited liability, account for about 90 percent of registrations. But there is a distinct trend towards more complicated forms such as joint stock companies (JSC). Registered enterprises are also small in size, both in terms o f their number of employees and their registered capital. At the end of 2003, the average sole proprietor firm had 15 employees; the corresponding figures for limited liability companies and JSC were 38 and 53, respectively. According to the ILSSA survey, the main customers were individuals for two thirds o f the firms, and other private firms for about a fifth. SOEs were the main outlet for 9 percent o f private enterprises, followed by local authorities (two percent) and state trading companies (1 percent). Export markets and FDIfirms were the main market for only 2 percent o f private firms. Enterprise registration i s also very unevenly distributed across provinces, both in terms o f numbers of firms and total capital (Figure 1.3). A few companies, such as Biti's, Kinh D o and Trung Nguyen Coffee, have succeeded in developing a recognized brand name. However, examples of this kindare far and between. The ILSSA survey also provides interesting information on the characteristics of private sector entrepreneurs in Vietnam. The overall picture that emerges is that o f a middle-aged male with 10 or more years of education and previous employment in a position o f responsibility, most probably in the state sector. About a quarter o f the owner-managers are actually older than 50, and very few o fthem are less than 29 years old. Only one-fifth of private sector entrepreneurs are women, although there are important regional differences. For instance, in Ho Chi Minh City (HCMC) 63 percent of the enterprises are male-owned, compared to 89 percent in Ha Tay and QuangNam. State-owned enterprises The state sector of Vietnam was established shortly after independence from France, in 1954, by both nationalizing existing privately owned enterprises and building new SOEs. Given 8 BUSINESSESALL SORTS OF the backward state of its agrarian economy, Vietnam's leaders argued that the SOE sector should be constructed usingthe Soviet Union economic model, which at the time was perceivedto bethe quickest way to develop the economy. Significant investment was devoted to industrial SOEs after reunification, through the second five-year plan. The business establishments of the former administration inthe South were forcefully transformed into northern-style SOEs. Meanwhile, by early 1978, 1,500 private enterprises from the South, employing 130,000 workers, had been nationalizedand converted into 650 SOEs. Figure 1.3: ... butUnevenlySpreadacross the Country Total Employment in Enterprises(2003) Total Equity of Enterprises(20 per 1 million inhabitants per I millioninhabitants (`000people) @//ion VNDJ Less than 15 Less than750 15to 20 750 to 1100 20 to 25 I100 to 1500 25 to 35 I500 to 2500 m50 35 to 50 2500 to 10000 mMore to 200 10000to 25000 than200 Source: Own calculations, basedon data from GSO andNBIC. The reversal of this trend started in 1986, with the Sixth Party Congress. In 1987, autonomy was given to SOEs to formulate and implementtheir own long-term, medium-term and short-term operating plans. Mandatory production targets were reduced to no more than three and the system whereby government provided the inputs was abolished. Outputs outside the mandatory plan could now be sold to other trading firms or even to consumers directly. Profits were to be calculated based on true costs. And except for compulsory transfers to the budget, they could be retained by the enterprise and used at its own discretion. In 1991, SOEs deemed inefficient or lacking capital and technology or not having sufficient demand for their products were forced to dissolve or merge with other units. As a result, by April 1994 the number o f SOEs had been reduced to 6,264, or roughly half those in operation at the heyday o f state-led development. In 1995, the Law on SOEs was enacted. Apart from conferring equal legal status to all of them, this law gave them the right to do business freely with each other and with non- state enterprises, including foreign partners under the form of joint ventures. SOEs were also allowed to hire and fire employees and set wages, within policy guidelines. However, they could not freely dispose of the capitalthat the government had entrustedthem with. The next stage in the transformation o f SOEs was organized around the so-called ownership transformation process, the most important part of which i s known as equitization. 9 This process amounts to divesting some of the state capital to the private sector. Until quite recently, those acquiring the divested capital were mainly workers and directors of the SOEs, making equitizationresemble an "insider privatization". Many of the SOE managers gaining control through this process are engineers by training, with a specialization in their particular industry. This professional background enables them to master new technology, such as that transferred by foreign investors to a joint venture, and to implement it in the production process. However, they often lack business education and management skills, which limits their capacity to engage in fundamental management change, including the reorganizationo f work processes and the delegation o f decision-making authority. More recently, however, the equitization of larger companies and the auctioning o f their shares have succeeded in attracting outside investors. On average, the state holds roughly 46 percent o f the capital in equitized enterprises, employees own 38 percent, and outsiders 15 percent. The outsiders' share doubled since 2003. But so far there are only 20 equitized enterprises with foreign ownership. The equitization process startedvery slowly, with only about 100 SOEs divested in 1998. It doubled its speed over subsequent years, and accelerated in 2003, when roughly 300 SOEs were divested. At present, ownership transformation affects some 500 SOEs every year. More than 2,500 o f them have been equitized by now, and about 900 more (many of them quite large) should follow inthe nexttwo years or so. Meanwhile, the creation of SOEs effectively came to a halt in 2001. The combination o f these two trends has resulted in a substantial decline in the number of SOEs, and an increase in the number of JSCs. At present, some 3,200 SOEs are in operation. By end 2006, there should be some 3,500 JSCs, o f which 900 with dominant state ownership. All of this is resulting in a sustained (albeit gradual) decline o f the state share o f the economy (Figure 1.4). Figure 1.4: A Declining State Share of the Economy I 75 ~_..._._I.".lll.I .....".."lll"".l"..." ...."..."l".."..."l ...."..."l"l.."" .ll....ll.lI. ..... "l".l1 I" "" , 1997 1998 1999 2000 2001 2002 2003 2004 -- Industrialoutput -Non-oil exports - - - -Bankingcredit Source: Own calculations, usingdata from GSO and SBV. From an administrative perspective, there are three types o f SOEs: those associated with General Corporations (GCs) 91, which report to the Office o f Government (OOG), those reporting to line ministries, and those under the authority o f provincial governments. GCs 91 are 10 BUSINESSESOF ALLSORTS sometimes referredto as "giants", butthey are inreality more modestly sized. Onlythree of them have capital or turnover in excess of one billion dollars. Some of them have elements of a monopoly business. In the case o f power generation, oil production, and railways, there i s more o f a "natural" monopoly situation, determined by technological constraints. Other GCs have a market share in excess o f 30 percent of the market for some aspect of their business. CGs related to cement, shipping lines, coal, air transport, steel and tobacco, among others, fall in this category. Whether they can actually exploit this market power i s unclear, as they do not own their affiliated SOEs, which are linkedto them only from an administrativepoint ofview. Because of political patronage from their respective authority, SOEs tend to be exposed to conflicting interests, including those o f the state in general, their direct authority and their employees. This conflict leads them to pursue not only profits, but also a broader set of objectives, reflectingthe interests o f their various stakeholders. Not surprisingly,they have come under criticism for poor performance. It appears indeed that their rates of return have been low (Table 1.1). But large losses are uncommon. An indication that SOEs inVietnam are not as poor performers as in other countries is given by their contribution to government revenue. Currently, some 54 percent of all corporate income tax (CIT) revenue, and about 42 percent of all VAT revenue from domestic production, are paid for by SOEs. Table 1.1:Not too Profitable, but Paying Taxes Rate of return Number of Revenues Liabilities Profits Corporate Numberof on equity (in SOEs (trillion (trillion (billion tax (billion employees percent) dong) dong) dong) dong) per firm Insolvent 295 14 26 (1,347) 12 345 Negative 709 56 25 (1,208) 1 244 Less than 2 907 63 37 228 58 298 2 to 5 617 74 33 477 123 434 5 to 10 605 80 44 1,173 299 480 10 to 15 353 45 25 1,348 320 532 15 to 20 234 104 20 1,313 307 521 20 or more 566 113 48 13,024 3.794 617 Total 4,286 549 I 228 15,008 4,914 41I Source: Own calculations, basedon data from GSO. Figures are for 2003. However, there are important weaknesses hidden behindthe relatively solid performance o f SOEs in Vietnam. The most obvious one i s the implicit transfer many o f them receive from the rest of society, by operating in sectors sheltered by tariffs and barriers to competition, by getting access to considerable amounts o f land at low cost, and also by getting their bad debts rolled over or even written off, The profitability of SOEs would be lower if these costs (which are very real to society) were factored in. A more subtle reason has to do with the mechanisms underlying this relatively solid performance. Unlike other transition economies, Vietnam has barred SOE managers from appropriating state capital or land, thus avoiding the worst forms of asset stripping observed elsewhere. But it has not been able to prevent SOE directors from appropriating some o f the profits o f the companies they manage. This gives them an incentive to 11 make the companies efficient, which i s one of the reasons why their performance is relatively solid, but it has been an important source of corruption. Foreigndirect investment One o f the first concrete steps towards economic renovation, in 1987, was to promulgate a Law on Foreign Investment. Together with the establishment and gradual improvement of the legal framework for FDI, Vietnam also signed international bilateral and multilateral agreements on investment encouragement and protection. Such agreements, which concern 45 countries and territories so far, have a wider scope of application than the regulations stipulated inthe Law on Foreign Investment. The trend was consolidated by the Ninth Party Congress, in 2001, which emphasized the role of FDI in "export orientation, construction of socio-economic infrastructure facilities, as well as transfer o f advance technology and creation of additional employment." Because o f this important role, the Party put forward the task o f "generating fundamental changes inattracting FDI". The main goal was to raise the quality of FDIinflows to Vietnam, by further attracting investments from multinationals, particularly in industries with high technological content. Although foreign investors are more limited in the range of sectors in which they can operate than domestic firms, the gap i s narrowing quickly. In 2002, foreign ownership in non- state enterprises was allowed in 35 designated industries, including agriculture, forestry and fisheries, science and technology, education and medicine. In 2003, both domestic and foreign investors were allowed to build power plants of up to 100 MW. In 2004, one-hundred-percent foreign-owned banks were authorized. These policies resulted in a rapid increase of FDI commitments, albeit from a low level. Admittedly, measuring FDI inflows i s difficult. GSO provides data on commitments as well as on implemented investments. However, the latter include domestic borrowing, so that they are higher than inflow data. IMF data focuses on actual inflows, because of their relevance from a balance-of-payments perspective. This includes not only the equity inflows, but also the foreign borrowing of FDI enterprises, and it could be argued that such borrowing i s not really FDI, because it is debt creating. But the distinction is not as clear when borrowing is from the parent company. Data from the United Nations Conference for Trade and Development (UNCTADj are the closest to equity inflows, as they do not include borrowing. Based on GSO data, cumulative FDI rose from 28 projects for a total of 140 million dollars in 1988, to over 700 projects and 5.5 billion dollars in 1993, to nearly 2,400 projects for more than 30 billion dollars in 2000. Two phases can be distinguished in this process. FDI inflows were unusually large in the mid-1990s. With commitments almost 10 percent of GDP between 1994 and 1997, Vietnam became then the top recipient o f FDI among all developing countries and transition economies. But the East Asia crisis broke this trend, and it is only since the beginningof the decade that FDIto Vietnam started growing again, after experiencing a sharp decline (Figure 1.5j. Only now, almost a decade after the onset o f the East Asian crisis, i s the volume o f FDIcommitments getting again close to 10percent o f GDP. The characteristics o f FDI projects have also changed over time. In 1996, the average project size was about 23 million dollars. But it had declined to 5 millions in 2000, and to 2.5 millions in2003. This said, even at their highest point, prior to the East Asian crisis, FDIprojects inVietnam were relatively small by international standards. It is worth notingthat only about 80 of the 500 biggest multinational corporations in the world have established a presence in Vietnam, compared to roughly 400 in China. 12 BUSINESSESOF ALLSORTS This relatively small project size is associatedwith the country origins of FDIto Vietnam. Asia is, .byfar, the most important source of capital. The main home economies are Singapore, Taiwan (China), Korea, Hong Kong (China) and Japan. Taken together, these countries account for almost two thirds of total FDI to Vietnam. Such predominance of regional investors partly explains why FDI inflows fell so sharply following the East Asian. Outside of Asia, France, Netherlands and the United Kingdom are among the most important investors. Figure 1.5: FDIRegainingMomentum... 12 b lo g i I 0 , I 1999 2000 2001 2002 2003 2004 2005e 1 1-Approvals - - - , Inflows I Source: Own calculations based on data from GSO, IMF and MPI. The United States does not feature prominently, but this is to some extent misleading. After the signing of the US-Vietnam Bilateral Trade Agreement (USBTA), investment from the U.S. companies in Vietnam has increased steadily. Between 2002 and 2004, United States- related FDI grew by 27 percent per year, compared to just around 3 percent from 1996 to 2001, However, much o f this growth has been managed from the regional subsidiaries of the U.S. companies, and it thus appears as if it originated in Asia. While on the surface the accumulated FDIfrom the UnitedStates only accounts for 730 million dollars, taking regional subsidiaries into account bringsthis figure up to about 2.6 billion. The industry distribution of FDI has changed over the years as well. Construction was one of the most important industries at the beginningof the period, with a peak in 1996. The average project size in this industry has been traditionally large, with projects focusing on hotels, office construction and infrastructure. Manufacturing industries(such as chemicals, construction materials and electric equipment) and services (mainly transportation, communication and finance) have become more important in recent years. FDI in agriculture and textiles and clothing has been low but stable. On the other hand, there i s a substantial amount of FDI in oil and gas exploration, most often under the form o f business contracts. Foreign-invested enterprises now account for a large share of the value of production in some sectors, including oil and gas (almost 100 percent), automobile assembling (84 percent), electronics (45 percent), textile and garments (41 percent), chemicals (38 percent), steel (32 percent), cement (30 percent), rubber and plastics (26 percent), and food and beverages (25 percent). The distribution of FDI across industries has been similar for Asian and non-Asian investors, except for a somewhat higher presence o f Asian firms in textile industries. On the 13 other hand, FDI investments display a considerable geographical concentration, although an increasingly large number o f provinces have become important recipients inthe period before the East Asian crisis (Figure 1.6). Figure 1.6:... but ConcentratedinFewAreas Total RegisteredForeignDire Investment(FDI) 1988 - 1997 Total RegisteredForeign Dire Investment (FDI) 1998 - 2004 Million USDper million inhabitants Mi//ionUSDper million inhabitants Lessthan 5 Less than 5 5-20 5 20 20.50 20 50 50 100 - = 50 100 100 -300 100-300 300 -700 300 700 - -- - Morethan 700 Morethan 700 t' Source: Own calculations based on data from GSO and MPI. Figures are based on FDIcommitments. During the first FDI wave, until the East Asian crisis, joint ventures were the most common form of investment, often with an SOE as the Vietnamese partner. Up to 1998, about two thirds of total FDI commitments were under this form. There are several reasons for this bias. In the early years of economic reform, SOEs were the only possible legal partners for foreign investors. And domestic private sector firms were still very weak anyway. Moreover, the privileged position of SOEs also made the joint venture modality more attractive, compared to green-field investments. SOEs have better access to commercial land, as well as good political contacts, which are essential inareas where the rule ofthe law is not fully established. The composition of FDI has substantially changed in recent years. In 1997, legal restrictions to partner up with private investors were removed. Since then, the share o f joint ventures in total registered capital has fallen to 42.5 percent, and the number of joint ventures with private sector firms has increased dramatically. Fully-owned foreign enterprises now account for 45.5 percent of committed capital, with build, operate and transfer (BOT) contracts and business cooperation contracts accounting for the rest. In 1991, wholly owned foreign affiliates accounted for about 20 percent of total investedcapital and 10 percent of the number o f projects. By 2000 these proportions had increased to almost 90 percent and 83 percent, respectively. Vietnam i s also becoming an exporter of capital. It recently licensed 13 overseas investment projects, for nearly 34 million dollars. The projects include exploration and 14 BUSINESSES OF ALLSORTS exploitation of crude oil in Malaysia, exploiting salt mines in Laos, and processing wood and establishing a supermarket in South Africa. Between 2000 and 2004, licenses for investment abroad were granted to 122 projects, totaling 230 million dollars. In2005, Vietnam i s expected to invest more than 300 million dollars abroad. Firmsizes: a missingmiddle? One of the main concerns that can be raised, almost twenty years after the beginning o f economic reforms, refers to the size distribution o f enterprises inthe resulting "multi-stakeholder" economy. Vibrant economic development and massive job creation are often associated with a dynamic SME sector. And it is true that Vietnam has seen a steady decline in the number of SOEs, combined with a dramatic increase in the number of private firms, both domestic and foreign-owned. But this change in the distribution o f enterprises by institutional type may not be associated with a change intheir distribution by size. The concerns in this respect is that thriving enterprises are either very small (inthe case of the domestic private sector) or quite large (in the case o f the FDIsector), whereas the "middle" o f the distribution remains more or less empty. Figure 1.7: Employment by Firm Size 65% I formal c0 E c S household a businesses I Firm size (2001) 15 15% 64% c c P) -0Q 10% E, - 5 * m c L -E 0 0 5% formal c enterprises ~ ~ Q household 0Yo businesses 1 <5 10 49 - 200 - >=I000 1 1 499 Firm size (2003) I Source: Own calculations using data from GSO. Figures refer to businesses in activity only. The top panel combines household survey data from 1998 and enterprise census data from 2001. The bottom panel uses data from 2004 and2003 respectively. A careful analysis o f data on enterprises which are actually in operation can be used to assess whether there i s a missing middle in the distribution o f enterprises by size. Two data sources need to be combined: household surveys such as the VHLSS, to account for household enterprises, and GSO' enterprise censuses, to capture registered enterprises. One important difference between the businesses covered by these two data sources refer to their formality, or lack thereof. Businesses in the GSO enterprise censuses can be SOEs, or privately owned, or foreign-owned companies; and they can operate under different corporate models, such as sole proprietorship, limited-liability or joint-stock company. But they are all formal. On the other hand, businesses in the VHLSS are informal; or, more precisely, they are a hybrid between the formal and informal sector. They are not formally registered under the Enterprise Law, but they are most often "listed" by local authorities. Inprinciple, businesses with 10 employees or more should register under the Enterprise Law. But remaining informal reflects to some extent a choice, involving a trade-off between barriers to enter into certain kinds o f contracts with suppliers, customers and creditors, on the one hand, and less stringent regulations on taxation, accounting requirements and transparency on the other. When combining these two sources o f data, it appears indeed that both very small enterprises, with up to five workers, and very large ones, with more than 1,000, account for the bulk o f employment inVietnam (Figure 1.S). However, the comparison between 2001 and 2003 also shows that this pattern i s changing. The share o f enterprises employing between 50 and 1000 workers increases during this period, whereas the share of those employing between five and 50 workers declines. Importantly, the share o f informal businesses in total employment falls quite rapidly. Formalization is particularly strong among businesses with five to 50 workers. Information on the growth, decline or disappearance o f enterprises over time can also be used to forecast changes how the distribution o f enterprises by size will change in the coming years. Enterprise censuses allow constructing so-called transition matrices, reporting the change insize (including possible closure) o f enterprises betweentwo points intime. They also provide 16 BUSINESSESALLSORTS OF information on the size distribution o f new enterprises between two points in time. A hypothetical distribution can thus be generated, assuming the same growth rate in the number o f enterprises, year after year, the same distribution by size o f the newly created ones, and the same probabilities o f growing, declining or closing as implied by transition matrices. Figure 1.8: Growth, Decline and Exit: the Implicit Trends. 0.25 I Source: Own calculations based on GSO data. Enterprise sizes are measuredinnumberof persons employed. Usingdata from the enterprise censuses o f 2001 and 2003, this exercise suggests that the share o f businesses with 10 to 50 workers will increase steadily in the coming years, until accounting for close to 40 percent o f total employment by formal businesses (Figure 1.9). This will be at the expense o fthe employment shares o f both very large and very small businesses. 17 2. BUSINESSAND DEVELOPMENT The growth o f business is seldom viewed as a development objective on its own. Much the same as good governance, it i s rather seen as an instrumental objective, aimed at building a better society, one with higher living standards and lower poverty levels. A thriving business environment is, above all, a way to elicit the energy and the creativity of the entire population, offering a reward to those who accumulate and innovate simply because they make more and better goods and services available to the rest. InVietnam, business development has been one o f the main forces behindrapid poverty reduction. Together with the redistribution of agricultural land, and the broad coverage of social services, it allowed a large fraction of the population to engage in more productive occupations and raise their living standards. The reform process launched almost two decades ago unleashed an enormous entrepreneurial energy, and led to massivejob creation. Business development has allowed absorbing 1.4 to 1.5 million entrants to the labor market every year, and offered a way out of farming for the rural population, especially for young women. Over the last decade, the average wage has grown at around 10 percent per year in nominal terms, or roughly 7 percent in real terms. The enormous opportunities created through this process have led to an inclusive growth pattern. While real GDP per capita has increased by 5.9 percent per year since 1993, the Gini index (measuring inequality) has increased only slightly, reaching 0.37 in 2004. Duringthat period, the poverty rate fell from 57 to less than 20 percent. Job creation Everyyear, some 1.4 to 1.5 million young Vietnamese enter the labor market. Sustained growth in living standards, and even overall social stability, very much depend on their ability to findjobs. A recent study on Vietnamese youth conducted by the UnitedNations Children's Fund (UNICEF) jointly with the Ministry of Health (MOH) and GSO is quite telling regarding the insertion of these new entrants into the labor market. Its findings are based on a nationally representative sample o f men and women who are 14 to 25 years old. More than half o f this group has a simple job, such as non-skilled worker, but there are important differences between urban and rural areas. Urban youth tend to report occupations that require a higher level of professional skills, with only one third of them having simplejobs. Also, more than a third o f the working youth are self-employed, whereas about a fifth works in a household business, and close to one tenth have a salaried job with the private sector. SOEs account for less than 7 percent o f total employment among the Vietnamese youth. The UNICEF survey uncovers a high level o f satisfactionwith current jobs. On average, 78 percent of respondents are satisfied, but the proportion climbs to almost 82 percent among urban groups and women. Its lowest level corresponds to unskilled agricultural jobs; but even there, at 73 percent, it i s remarkably high. While there are no major gaps in job satisfaction across regions, it i s clear that a booming economy helps. Indeed, the highest job satisfaction levels (close to 81 percent) are inthe RedRiver delta and the South East. The lowest ones (below 70 percent) are inthe North Central coast. BUSINESS DEVELOPMENT AND In questions relating to future aspirations, half o f the respondents identify work as their highest priority. The answer i s quite similar across all age groups, between male and female respondents, and in urban and rural areas. When invited to make recommendations to the government, on what could improve young people's lives, more than 40 percent state that increased opportunities for work should be the number one priority. The recent experience o f Vietnam suggests that business development i s fundamental to address this concern. This is revealed by the enormous diversity o f experiences across provinces. Growth in wage employment has been strongly associated with business registration under the Enterprise Law (Figure 2.1). Figure 2.1:The Private Sector as an Employment Engine 70% - HCMC Danang 0. 60% * A A 9 0 Hanoi 50% i Binh Duong 40% 30% 20% Mau 10% Son La Lang Son Lao Cai 0% 1 ~ 5 5 5 6 6 5 7 7 5 8 8 5 Registered enterprises per million inhabitants (in log) I Source: Own calculations based on data from MPI and GSO. Figures on enterprise registration correspond to the period 2000-2004; figures on employment creation are for 2002- 2004. The recent experience o f Vietnam also suggests that the "cost" o f creating a job varies considerably across businesses o f different sorts. Admittedly, referring to the "cost" o fjobs is a shortcut that could be questioned on analytical grounds. The idea o f a given cost implicitly assumes a lack o f flexibility in the way the factors o f production can be combined, as if every job had a given capital requirement. In fact, more expensive jobs, associated with more capital per work, also tend to be more productive. So, from an economic perspective, high-cost jobs could (at least in principle) be preferred to low-cost ones. However, a country facing 1.4 to 1.5 million entrants to the labor market, and having limited resources for capital accumulation, cannot afford to offer jobs with high capital intensity to everybody. In recent years, the domestic private sector has been the most effective in terms o f generating employment at a low cost (Figure 2.2). A very crude way to see this i s to compute the average capital per worker across SOEs, FDI firms and the domestic private sector. It then appears that capital intensity is highest in FDI firms and lowest in the domestic private sector. But this comparison is potentially misleading, as some SOEs are still suffering from the over- staffing they inherited from the period under central planning. If their workforce were adjusted downwards, so as to maximize profits, their capital endowment per worker would increase. 19 A slightly more rigorous way to assess the cost of a job in each sector is to look at changes in capital and labor endowments over time (as opposed to their levels at any point in time), Itthenappearsthat the higher capital cost perjob is in SOEs, not inFDIfirms. Again, this does not necessarily implythat SOEs are inefficient. For instance, the power sector requires very heavy capital investments to cope with the booming demand for energy, but stable employment in this sector would actually indicate a increase in its productivity. What is clear, when looking at this somewhat more rigorous assessment, is that only the domestic private sector is in a position to massively createjobs inan affordable way. Figure2.2: The "Cost" of Creating a Job Capital per employeein 2000 and 2003 400 350 n z 300 > 250 200 150 100 50 0 SOE FDI PrivateDomestic Enterpriseownership type Additional capital per additional employee in 2001-2003 I 700 1 I 1 600 0 5c 500 400 - 300 ,O i5 200 100 0 SOE FDI Private Domestic Enterprise ownership type Source: Own calculations based on data from GSO. Outputgrowth At the risk of simplifying,there are two main ways to increase output per worker in the short to medium term. These are to raise the endowment of capital available to each worker and 20 BUSINESS DEVELOPMENT AND to use whatever capital is available more efficiently. These two main drivers are often referred to as capital accumulation and total factor productivity growth. (Note that the in the longer term, total factor productivity growth cannot be dissociated from the upgrading o f skills). A rigorous analysis o f the contribution of these two drivers to labor productivity growth can only be conducted at a disaggregated level. Ideally, the analysis should focus on individual enterprises; if not, at least on relatively homogeneous sectors. Economy-wide indictors of capital accumulation and total factor productivity growth need to be interpretedwith much caution. Still, they can be used to illustrate broad economic trends. The speed at which capital is being accumulated in the Vietnamese economy can be measured through the fraction o f total income that is invested every year. This investment rate has increased steadily inrecent years. By now, about 37 percent o f the country's GDP i s devoted to increasingthe stock of capital, compared to 30 percent duringthe East Asian crisis. This i s one o f the highest investment rates in the world, and it may not be advisable to raise it even further. Mobilizing such a huge amount of resources in a transparent and efficient way is indeed a challenge. Banks may not be able to rigorously assess the risk of all the projects they are financing, whereas the capacity o f the public sector to conduct rigorous appraisals on its own investmentsi s still very limited. However, it is encouraging to see that the composition of capital accumulation i s changing as well. Over the last few years, the investment rate o f the domestic private sector has increased by half; it now accounts for a third o f total capital accumulation (Figure 2.3). Because o f the private sector's strong profitability focus, it is likely that its investments are carefully screened out. A similar, if not more demanding type of screening applies to FDI investments, which were relatively stable until 2004, but are gaining speed o f late. It i s reassuring to see that public investments funded directly by the budget have increased steadily. Most of these investments are relatedto the development of large-scale infrastructure, which tends to have high payoffs ina low-income country. Moreover, the transparency o f budget allocations has improved considerably in recent years, with appropriations now subject to closer scrutiny by the National Assembly and People's Councils at all levels. Figure2.3: Investmentby Institutional Sector 10.0 ' -_.-- --- / c - - .fie.. .L I I 1998 1999 2000 2001 2002 2003 2004 , ....-..Statebudget --SOE -Non-state domestic ---FDI Source: Own calculations based on data fiom GSO. 21 BUSINESS On the other hand, it i s not clear that investment projects by SOE are always subject to rigorous appraisal procedures. The pressure to invest profitably could be undermined in their case by the involvement of multiple stakeholders, with potentially conflicting interests. SOEs also tend to have better access to financial resources, from SOCBs and policy lending institutions such as the Development Assistance Fund (DAF). The "softer" budget constraint they face, compared to their private sector counterparts, could also result in less discipline when preparing and appraising investment projects. Measuring total factor productivity gains i s considerably more difficult than measuring capital accumulation. The standard technique involves several questionable assumptions. It basically amounts to estimating the productivity of an additional unit of capital and an additional unit of labor, then multiplyingthe actual change inthe amounts of capital and labor used in the economy by the corresponding productivities, and finally comparing the result to the actual change in output. The gap between the actual change in output, and its predicted increase based on estimated productivities, gives a measure of the gain in efficiency experienced by the economy. Estimates o f total factor productivity gains based on the standard technique have been conducted on many countries. They usually range from 1to 4 percent per year. Note, however, that the standard technique requires a reliable measure o f the change in the capital stock from one year to the next. While more-or-less accurate figures can be produced for aggregate investment, reliable indicators of the overall stock of capital tend to be unavailable in developing countries. The problem is actually worse in a transition economy, where old investments were registered based on administrative costs often bearing little resemblance with reality. Under-estimating the stock of old capital leads to over-estimating the rate at which the capital stock is growing, hence to under-estimating total factor productivity gains. A related problem concerns capacity utilization. In an economic downturn, like the one experienced by Vietnam in the aftermath of the East Asian crisis, some existing capital i s under-utilized and output falls short of its potential. In measuring total factor productivity gains it i s the potential output that counts, more than the observed output. Figure2.4: Productivity Gains at the Aggregate Level 1 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 -Raw - - - #Adjusted 1 Source: Khoi Quang Chu(2004). With these caveats in mind, it i s still interesting to see that, according to available estimates, total factor productivity gains have been quite high in Vietnam (Figure 2.4). "Raw" 22 BUSINESS DEVELOPMENT AND figures, not taking capacity utilization into account, put the annual gains at two to three percent per year after the East Asian crisis. This is comparable to estimates available for good economic performers such as the United States. "Adjusted" estimates, taking into account the decline in capacity utilization after the East Asian crisis, are much higher. They increase from roughly 4 percent per year in the early 1990s to more than 6 percent at present. On the surface, this i s too high to be credible. However, estimates at the enterprise level, using a totally different data source, yield comparable figures. Figure 2.5: Productivity Gains at the Enterprise Level Solow method I m 0 > 7 C 6 I = 5 I -m a 4 3 I z 2 a 1 I 0 SOE Equitized Private FDI 1 Type of enterprise Panel data estimates 12 2 SOE Equitized Private FDI Type of enterprise Source: Own estimates based on GSO data. Productivity gains Total factor productivity gains are somewhat easier to evaluate at the enterprise level, as this is where innovations actually occur. The availability of disaggregated data from enterprise 23 BUSINESS censuses allows conducting a more refined estimate than macroeconomic data permit. Also, the availability o f tens of thousands o f observations, and the possibility to observe the same enterprise at two points in time, make it possible to use more sophisticated (and potentially more reliable) estimation techniques. The focus here is o n firms in manufacturing; agricultural enterprises and most firms in services are set aside. Interestingly, different techniques yield relatively similar results. Reassuringly, the results are also similar to those obtained with the standard, crude technique at the aggregate level. What is striking, in all cases, is the sheer magnitude of the estimated total factor productivity gains (Figure 2.5). Equally interesting is the variation intotal factor productivity gains across businesses of different sorts. Regardless o f the estimation technique, gains appear to be substantially higher in the FDI sector. This may be due to the fact that many FDI investments are new, with many establishments still inthe process o f reaching their full capacity. Their growth in output could b e unusually high during this start-up phase, without this implying that they are becoming much more efficient. At the other end, total factor productivity gains are lowest, but still high by international standards, for both SOEs and domestic private enterprises. Box 2.1: Public Sector Champions: Electricity of Vietnam Unlike many other Vietnamese SOEs, EVN has been successful in establishing a corporate culture and a commercial orientation, particularly in recent years. EVN's financial accounts are strictly separate from the government budget, and EVN receives no government budget subsidy support for investment or its operations, with the exception o f certain resettlement financing. The company is now facing commercial terms for borrowings, except in areas with a social objective, such as rural electrification. It introduced international accounting standards (IAS) early on, and relies quite systematically on competitive biddingfor its procurement. EVNhas beenprofitable untilnow, which is unusual for a public utility ofthis sort. Itcovers all its costs, including depreciation and financial expenditures related to its borrowing, from revenues. This has been accomplished while keeping costs to consumers at levels which are in line with international standards. Management o f consumer accounts receivable is exceptionally effective for a developing country. Steady reductions in transmission and distribution losses have also helped. Non-technical losses, including theft of electricity, are low compared to many countries. However, benchmarking EVN's efficiency would require more disclosure about its cost structure. Primarily managed by EVN, Vietnam's power sector has kept pace with extraordinary increases in demand, and succeeded at maintaining basic service for its customers most of the time. Its rural electrification program has been one o fthe most successful inthe world. The number o f rural households with access to electricity increased from 51 percent in 1996 to 88 percent in 2004. Widespread anecdotal evidence also points to substantial increases in the quality o f electricity service over the last ten years, with supply to most customers in urban and peri-urban areas becoming noticeably more reliable. However, service i s not yet seen as reliable enough by f m s , which report an average of 12 power outages or surges per year and, in the case o f small and medium-sized firms, losses amounting to two to three percent of sales due to outages. Looking forward, EVN's own success raises new challenges. As a market for electricity is gradually being set up, EVN will need to be broken up into truly separate corporations. The existing model o f EVN as the holding company for all o f the state's assets inthe power sector cannot be retained iftrue competition is to beachieved among existing andnew generators, andlater, inwholesale andretail supply o f electricity. The market requires independence among players, to avoid perceived conflicts o f interests or potential collusion. Source: World Bank (forthcoming). 24 The similarity in the performance o f these two types o f businesses could be misleading, however. SOEs still enjoy considerable advantages compared to domestic private enterprises, especially in terms o f their access to land and finance. If the playing field were more level, private sector firms could be expected to outperform SOEs in terms of their total factor productivity gains. However, it should not be concluded that all SOEs are inefficient. The example o f Electricity o f Vietnam (EVN) suggests that public sector entities are capable to deliver remarkable performances as well (Box 2.1). It is also worth noting that equitized SOEs do better than both unrestructured SOEs and domestic private firms. To some extent, as in the case o f FDI enterprises, such remarkable performance could be to a large extent reflect a "catch-up" effect. Equitization reduces the administrative burden on enterprises and makes them more independent from the often conflicting interests o f their main stakeholders. It therefore allows them to move closer to the efficiency frontier. Standard estimation techniques are likely to confuse this movement with a shift o f the efficiency frontier. After the catch-up effect is over, equitized SOEs should not be expected to do much better than private sector firms. But the estimated total factor productivity gains, no matter how temporary, suggest that the equitizationprocess has been an effective tool to raise efficiency inthe SOE sector. Poverty reduction There i s a clear correlation between the number o f enterprises in a province, relative to its population, and the speed at which poverty i s being reduced (Figure 2.6). Thus, the dynamic business environments o f BinhDuong, Danang, Hanoi and HCMC have provided the context for improved welfare for the local populations. Of course, correlation i s not causation. But there i s evidence that thejobs created bythese enterprises are helping people to move out o f poverty. Figure 2.6: Business Activity and Poverty Enterprises in operation per province in 2003 (in log) Source: Own calculations basedon data from GSO. 25 BUSINESS At a national level, the proportion ofthose at work derivingtheir main income from wage employment rose from 16 percent in 1993 to 27 percent in 2004. The drop in household dependence on farm incomes almost mirrors this pattern. Meanwhile, the average wage grew by about 10 percent in nominal terms over this period, or roughly 7 percent in real terms. Regional patterns confirm that reductions in poverty are more rapid where non-farm wage employment is more dominant. About 35 percent of the population inthe RedRiverDelta and 45 percent o f the population inthe Southeast report that their main source of income i s wage employment. But the proportion i s only 10 percent inthe North West. However, the relationship between wage employment and poverty reduction is not entirely straightforward. Those deriving their income from wage employment are also representingan increasingly large share of the poor. Inthe Southeast, 19 percent of those inthe poorest fifth o f the population depend on non-farm wage employment as their main income source. The story i s similar in the Red River delta. This suggests that not all jobs created by businesses are associatedwith poverty reduction. Many o f those employed inthe areas where enterprises are developing rapidly come from across the country. Those migrating to the Hanoi and the industrial areas in the north are traveling primarily from the RedRiverDelta. Those migratingto HCMC and the industrial zones inthe Southeast come from a broader range of regions. A recent survey of 5000 migrants found that more than two-thirds of migrants had remittedat least one million dong during the previous 12 months. Migrants reported that the funds they were remitting home were spent largely on meetingeveryday expenses, includinghealth and education costs. Gender equality A thriving business environment can also foster economic and social development objectives in less obvious ways. Empowering women i s one of them. Vietnam has a solid track record interms of gender equality, compared to other countries. Educational attainment i s similar for men and women, health indicators are roughly comparable too, and even labor force participation is unusually even across sexes, with women accounting for 49 percent of the work force between ages 16 and 55. However, at 40 percent, the share of Vietnamese women in total wage work is substantially lower. And not so many women are in managerial positions. Bringing a sizeable share of cash income to the household, and being champions in the business arena, should raise the voice of women at home and in society at large, while giving them more independence. Business development could be helpinginthis respect. Garment and footwear production have been among the most dynamic sectors o f the Vietnamese economy in recent years. More than 80 percent o f the workers in this sector are female. At present, there are some 1,100 textile and garment factories inoperation, including 231 SOEs, 500 private or joint-venture companies and 354 FDI enterprises; together, they employ more than 2 million workers. There are also 240 footwear enterprises, of which 76 are SOEs, 84 are private or joint-ventures and 80 are foreign-invested. Taken together, they employ 400 thousand workers. A recent study by Mekong Economics, on women workers in 21 of these garment and footwear enterprises, provides revealing information regarding their profile, working conditions, overall satisfaction and aspirations. The study found that most female workers were young, single, migrants. Through their employment inthe garment and footwear industriesthey were able to earn money, live away from 26 8USlNESS AND DEVELOPMENT their villages and in some instances study or access more information on social issues. Moving away from the village also means that these young women had to rent accommodation, live independently o f their families and negotiate a new living environment, including on access to basic social and health services. On the positive side, the average income o f these female workers was approximately one million dong per month, which is considerably more than the minimum wage. On the negative side, while women occupied all kinds o f positions, including managerial and technical positions, they were mainly concentrated on the production line, mainly on sewing and over-locking jobs. These positions are characterized by low salaries, long working hours and minimal opportunities for skills development. Also, the research found that female workers were not receiving all the benefits stipulated by labor legislation. Working conditions were especially poor in some private , and SOE factories, characterized by uncomfortable workshops, few breaks and limited access to snacks, clean toilets and rest places. Yet the study found that most workers, male and female, were satisfied with their jobs. More than 40 percent o f them stated that they were motivated to work in a factory, because they thought it was a good opportunity for them. More women than men declared that their wages were sufficient to meet their needs. Migrant workers were especially upbeat. The city or industrial zone their factory was located in was seen by them a place o f opportunity, where it was possible to gain experience and acquire information that was unavailable in rural areas. Some also referred to the possibility o f working and studying at the same time, although very few o f them were actually doing so. In spite o f the shortcomings, when putting all the pieces together the authors o f the study conclude that the garment and footwear industry has provided young Vietnamese women a route out o f farming. Vietnamese women are also making their way into top managerial positions (Box 2.2). While both sexes are equally represented in the day-to-day running o f household businesses, relatively fewer women lead registered enterprises. The case o f those who do helps understand how business development affects their lives. A survey o f Vietnamese women entrepreneurs was recently conducted by MPDF. It found that 69 percent of them were inthe 35-54 age bracket, and 80 percent were married. Their average household size, including spouse, children, parents or other relatives was slightly above five persons. Just 12 percent had only a primary level o f education, while 25 percent had secondary schooling, 27 percent some post-secondary education, and fully 34 percent had a university or graduate degree. Only 6 percent o f them were from ethnic minorities. The women answering the survey were the owners o f established, substantial business enterprises. Most had been in business for five or more years. On average, they employed 68 permanent workers and 25 part-time or seasonal workers. The MPDF survey revealed that these women share many o f the same concerns as male business owners. For instance, the ability to find and retain good staff, the acquisition o f financial management skills, and the existence o f restrictive laws and regulations were judged to be either important or extremely important by between 70 to 80 percent o f respondents. The majority of respondents declared that gender made no difference in dealing with bureaucracy and corruption, or in gaining access to capital. However, respondents also reflected on aspects o f business ownership and management which were more difficult for them because they were women. The main challenge was to balance their professional activity and their family life, given the long hours needed to manage their businesses. Respondents also described personal safety issues and the ability to network and develop business relationships as more problematic for women than for men. On the other hand, dealing with clients and customers was seen as easier for women. 27 On balance, Vietnam's businesswomen claim that their gender does not put them at a disadvantage. When asked whether being a woman was a net negative, net positive, or made no difference in their business, two thirds o f respondents chose the third option. While a 54 percent majority feels that there i s no gender difference in"being taken seriously as a business owner," 30 percent say this is easier as a woman, and only 7 percent say it i s more challenging. Perhaps this i s why Vietnam's businesswomen largely feel that they are, indeed, taken seriously. Box 2.2: PrivateSector Champions: Women Entrepreneurs Ms.M y Le startedher businessfrom zero. In 1987, when she hadbeenmarried for nine years and had five children, she left the SOE factory she was working for. At that time, a new policy had been introduced stipulating that families should not have more than two children. My L e moved then into trading on agricultural products. At first her husbandwas not supportive, because he thought that she had been dragged into that line o f businessby people he did not trust. Moreover, the business was risky, and My L e was losing money. Only in 1990 did the husbandrecognize that she was trying very hard anddeservedto be supported. My L e registeredher private enterprise, later called M y L e Co. Ltd., in 1993. Hers was the fust private company in what was then Song Be province (transformed into Binh Phuoc and Vinh Long today). At that time, farmers lacked amarket for raw cashew nuts. M y L e and her husbandventured to invest in a factory to process the nuts. The company did not have enough capital, management experience or skilled workers. The small amount of money M y L e borrowed through Vietnam's Women Union was fundamental to get her started. By 1996, the company had about 200 workers and a turnover of 2.5 billion dong. Ten years later, employment has increasedtenfold, and turnover has reached 150 billion dong. The factory has a work floor of 25,000 square meters, is equipped with modem technology, and processes more than 30 tons ofraw cashew nuts every day. About 95 percent ofthe output is exportedto the UnitedStates, the EuropeanUnion, China, Singapore and other foreign markets. The companyhas built its customer base on its reputation for high-quality nutsreliability indelivery. Similar stories to M y Le's echo around the community of women entrepreneurs. A recent study that captures the experiences of more than 60 of them relates the common hurdles they encounter: access to capital and land, a lack of business management skills, and the difficulties associated with managinga business and a family simultaneously. Many ofthose finally succeeding inbusiness have a number of false starts, with previous businesses either faltering or failing as they develop the skills to handlethese constraints. I Source:Basedon Froniga Greig and Thi Nguyen (2004) and HangThi Thu Pham (2005). I 28 3. EFFICIENCY AND COMPETITIVENESS Few would question that the business sector of Vietnam is thriving, and has made a substantial contribution to economic and social development in recent years. But whether its performance compares favorably or unfavorably to that of other countries, especially in the East Asia region, i s more controversial. Assessments of Vietnam's strengths and weaknesses, and collective anxiety about its capacity to make a place for itself in the global marketplace, often take the form o f discussions about "competitiveness". This is a somewhat elusive concept, considered fundamental by most management experts and meaningless by many professional economists. However, it captures the idea of relative position, a concept which i s somewhat easier to grasp. Unfortunately, the available competitiveness ratings provide radically different assessments o f Vietnam's relative position, which can only add to the confusion. Ratings relying on the characteristics of the regulatory framework in which firms operate put Vietnam in the bottom half o f the world distribution or close to the middle at best. On the other hand, direct assessments by foreign companies, under the form of opinion surveys or actual investments flow, rank it in the top quarter, or even higher. Deeper insights can be gained by analyzing the way in which Vietnam's businesses actually operate and participate in the global economy, either directly or indirectly. Those analyses show that Vietnam's pattern o f integration inworld trade is closer to that of China than to that of other ASEAN countries. But they also reveal a still insufficient integration o f domestic companies in global value chains. Total factor productivity i s growing very rapidly across the board; but growth i s faster in foreign companies than in domestic ones, regardless of their ownership. Productivity spillovers occur between foreign and domestic companies, especially private ones, but labor turnover and outright imitation seem to be more important channels of transmission than business-to-business transactions. The equitization o f SOEs, in spite of its limitations, appears to be contributing to productivity gains, as it establishes a more arms-length relationship with government authorities. In the end, the key to competitiveness i s the ability o f businesses to transform themselves, with the public sector having a very important role to play in this respect. Based on this broader perspective, a more comprehensive rating systemto assess Vietnam's strengths and weaknesses can be proposed. The conventional wisdom@) There are a variety o f cross-country ratings which can be linked, in one way or another, to the concept of competitiveness. In some cases, the link i s rather tenuous. For instance, several risk rating agencies classify countries into a number of categories, depending on the risk level associated with their sovereign bonds. While sovereign risk is quite far removed from competitiveness, ratings of this sort do have an influence on FDIinflows. Other ratings get much closer to the competitiveness notion, if nothing else because o f their name. Three have received especial attention in Vietnam in recent years. These are: the Growth Competitiveness Index produced by the World Economic Forum, the Doing Business ratings produced by the International Finance Corporation, and the FDI Potential index'produced by UNCTAD. Each o f these ratings covers more than one hundred countries. All of them put Vietnam inthe bottom part of the worldwide competitiveness distribution, or around its middle at best (Figure 3.1). The Growth Competitiveness Index is a complex combination of a vast number of cross- country indicators, such as the corruption perceptions index by Transparency International. These indicators are classified under the headings of "technology", "public institutions" and "macroeconomic environment", and grouped into various categories such as openness, government, finance, infrastructure, management, labor, and institutions. The Growth and Competitiveness Index shows a sustained decline of Vietnam's absolute ranking over time; currently, it is in 81` position among 117 countries. Butthe decline i s mainly due to the constant addition o f (typically poorer) developing countries to the database. If the relative position i s considered instead, Vietnam's rating has improved quite steadily. Inthe year 2000, it was in the 90thpercentile; by 2005, it had moved up to the 70thpercentile. Figure 3.1: A Bad Place to do Business...? Source: Own calculations, usingdata from World EconomicForum(2005), World Investment Report (2005) and World Bank (2005). Figures refer to a country's position among all those considered, from best (at the top) to worst (at the bottom). The Doing Business index considers 10 topics: starting a business, protecting investors, hiring and firing workers, enforcing contracts, registering property, closing a business, getting credit, dealing with licenses, paying taxes, and trading across borders. In its most recent release, the aggregate index is calculated as the average of the rankings (in a scale from zero to one hundred) for each of the 10 topics. The ranking on each topic, in turn, i s the average o f the rankings for its component indicators. The most recent Doing Business report ranks Vietnam as number 99 out of 155 countries. Comparative ratings had not been provided in previous years, when the number of topics covered by the report was smaller. But it is relatively straightforward to apply the same methodology to data from previous years. The result does not uncover any clear trend, with Vietnam fluctuating aroundthe 70thpercentile. The FDI Potential Index captures several factors, apart from market size, which are expected to affect a country's attractiveness to foreign investors. These factors include GDP per capita, the growth rate o f GDP over the last decade, the share o f exports to GDP, indicators o f 30 EFFICIENCY AND COMPETITIVENESS telephone density, a measure o f commercial energy use per capita, the share o f research spending in GDP, the share of tertiary students in the population, a country risk indicator, the share in worldwide exports of natural resources, the share in worldwide imports of parts and components for automobiles and electronic products, and the worldwide share in inward FDI. The FDI Potential Index is an average of the values of all twelve factors (each normalized to a scale going from zero to one hundred). It makes Vietnam appear under a more favorable light, as it ranks 68 out of 140 countries. There i s no fundamental change inthis ranking over time. Beyond the diversity of their underlying indicators, what these three rankings have in common i s their focus on factors that could potentially make a country attractive to investors. Now, which are those factors in reality i s a matter for debate. It could be argued, for instance, that the topics selected by the Doing Business report give more weight to the regulatory environment than a "real" investor would. While a scholarly discourse would emphasize transaction costs and limited protection of property rights as major obstacles to business, investors in a developing country might be used to operate informally, enforcing deals through trust or reputation, while seldom havingrecourseto the legal system. A differentapproach is to focus on what investors actually do. Inthis respect, the flow o f FDI into a country is a good measure of its attractiveness. To make this indicator comparable across economies of different sizes, the FDI inflow can be scaled by the country's GDP. From this perspective, and after excluding economies with a total GDP below 10 billion dollars per year, Vietnam ranks 25 out o f 88 countries. While its ranking was much higher before the East Asian crisis, there is a steady improvement since the year 2000 (Figure 3.2). Figure3.2: ..or One ofthe MostAttractive Destinations? , , 120 ` +FDI inflows as share of GDP rank ---c-- JBIC survey Source: Own calculations, based on data from WIR (2005) and JBIC (2004). Figures refer to a country's position among all those considered, from best (at the top) to worst (at the bottom). Vietnam's ranking i s even higher inthe eyes of Japanese investors. For the last 15 years, a survey has been conducted in Japan to assess trends of overseas business operations and FDI activities. In 2004, the survey covered 932 manufacturing companies that had three or more foreign affiliates, including at least one manufacturing base. When asked about promising countries for business operations in the medium term, Vietnam ranked number four, worldwide, 31 after China, Thailand and India. This represents an improvement compared to previous years, althoughVietnamhas featured highinthis survey inrecent years. The mainreasons for Japanese surveyrespondentsto select Vietnamwere its inexpensive labor force, its potentialfor growthas a market, its excellent humanresources and its potentialas a source of risk diversification. Interestingly, two of these reasons, namely the quality of the workforce and the cost of labor, are not among the indicators considered by the World Competitiveness Index, the DoingBusiness Index, or the FDIPotentialindex. International integration Another way to assess the competitiveness of the Vietnamese economy is to look at its actual performance inworld markets. Increased integrationin the world economy has been a central objective of the economic reform process in Vietnam. There have been several important milestones in this respect. In 1993, all enterprises with a business license were authorized to engage in foreign trade in the goods specified in their license, without requiringany additional authorization. In 1991, a duty rebate scheme was introduced, enabling export-orientedfirms not to pay trade taxes on their imported inputs. In 1995, Vietnam became a member of AFTA. Under this agreement, it committed to reducing tariffs on imports from AFTA members, on all but a few sensitive products, to less than five percent by the year 2006. In 2001, long and protractednegotiations culminated in the signing of the USBTA, which is considered the most comprehensive of all bilateraltrading agreements signedbetweenthe UnitedStates and a developingcountry. In2004, the requirementfor exporters to surrender foreignexchange to the State Bank of Vietnam (SBV) was abolished, after havingconsiderably weakened in 1999 and 2001. Export duties and import quotas have been gradually dismantled; by now the former only apply to crude oil and scrap metal, and the latter to sugar and refinedpetroleumproducts. And it is expected that Vietnam will be able tojoin the WTO as early as 2006. The result has been a remarkablyopen economy, with the sum of exports and imports representing by now morethan 130percent oftotal GDP. Exportshave grown at an average rate of 18 percent per year since 2000. The pessimism that prevailed in some circles before the signing of the USBTA, regardingthe prospects for Vietnamese companies to expand exports to the UnitedStates, is bynow forgotten. Interestingly,the trade patterns ofVietnamhave some resemblancewith those of China, a country whose competitiveness is seldom put in doubt (Figure 3.3). China is now seen as the factory of the world, processing raw materials and semi-finished inputs from developing countries, especially in the East Asia region, and selling the output to industrial countries. China's share inASEAN exports morethan doubledbetween2000 and 2004; it tripledinthe case of Thailand andthe Philippines. But not so inVietnam, where it remained stable until2003. On the other hand, the share of the United States and the European Union in ASEAN exports declined during this period. Not so in Vietnam, where it tripled between 2000 and 2003, and most likely would have continued growing, had it not been for the textile quotas and anti- dumpingproceduresimposedbythe UnitedStates in2004. At present, China is a less important market for Vietnam than for ASEAN, while the opposite was true in 2000. Conversely, the United States and the European Union absorb a higher share of exports in the case of Vietnam than in the case of ASEAN; it was the other way aroundin2000. 32 EFFICIENCY COMPETITIVENESS AND Figure3.3: More like China, less likeASEAN Exportsto China 2000 2001 2002 2003 2004 I - Malaysia - - -*Indonesia Thailand -Vietnam -ASEAN' 1 I 45 ,-- Exports to USand EU I I I I ~ -- 2000 2001 2002 2003 2004 1-Malaysia - - - aIndonesia -.-. -Thailand -Vietnam -ASEAd Source: Own calculations based on data from WDI (2005). More generally, Vietnam's main sources o f imports are developing countries, while its main export outlets are industrial countries. In2004, one quarter of Vietnam's imports originated inASEAN countries, and another 14percent inChina. Japan, the European Unionandthe United Statesjointly accounted for only one quarter o f Vietnam's imports. But they absorbed more than halfo f its exports. In away, Vietnam has also become one of the factories o fthe world. Domestic integration One relevant question i s whether global integration touches in any way the vast majority of Vietnamese businesses which are not directly involved in international trade. If it does not, thenitwould be difficult to claimthat the Vietnamese economy as awhole is competitive. 33 The emerging private sector is still, to a large extent, inward-oriented. Based on the three-wave survey of SMEs by ILSSA, only 9 percent of the sales correspond to direct exports. Most of this fraction i s accounted for by firms in Hanoi, Haiphong and HCMC, with the figure beingmuch lower elsewhere inthe country. Moreover, exports are not being processed through indirect channels either. For instance, state trading agencies only absorb 1 percent o f all sales by SMEs. This inward orientation is worrisome, given that the labor productivity of the average exporting enterprise i s nearly twice as high as the average productivity of SMEs. However, there are mechanisms linkingnon-exportingbusinesses with world markets and ensuring the transmission of international price signals from ports and airports all the way to farms. Those mechanisms are often referredto as "value chains", and can traced from producers, collectors, processors, wholesalers, retailers and exporters. Value chains in rice and tea have been the subject of careful studies in Vietnam (Box 3.1). One pattern that emerges from these studies i s the reliance on trading through SOEs, which results in an emphasis on volume, as opposed to quality. A stronger involvement of the private sector in the supply chains could help identify new overseas partners and establish higher quality standards. Inthis way, Vietnam could move up the ladder inworld markets, and farmers (which tend to be poorer than the population at large) could appropriate more of the export value. Business-to-business transactions contribute to the integration o f domestic markets in other sectors as well. A small survey of enterprises conducted by the World Bank shed some light on the relationship between contractors and suppliers in Vietnam. Detailed interviewswere conducted in nine large SOEs and 11 FDI companies on the contractors' side, and in 10 small SOEs and 11private domestic enterprises on the suppliers' side. It appearedthat for a third of the contractors, the value of the inputs purchased was equivalent to more than half of their total revenue, The process typically involved a large number of sub-contracts (more than 300 a year, on average) with many different parties. The commercial relationship was even more important when seen from the suppliers' end. On average, suppliers generated two thirds of their total turnover out of these business-to-business transactions. But the entire annual revenue was generatedthrough the contracting processfor one thirdo f the suppliers. Patterns are similar in the three-wave survey o f SMEs conducted by ILSSA. The time dimension of this survey reveals that non-state enterprises have become much more important customers, especially in Hanoi. Their share of sales doubled between 1991 and 2002. By now they account for about half o f the suppliers' turnover. SOEs diminishedsomewhat in importance but the most remarkable change was the almost complete disappearance of that state trading companies from the picture, not only in Hanoi but nationwide. Production on advance order i s the most common mode o f business-to-business contracting. But the percentage o f enterprises always selling on the security of an order has decreased. By now, in HCMC a third of the enterprises never sell on order. Technology spillovers Foreign firms in Vietnam are not yet considered to be an important source o f contracting opportunity, and this raises questions as to their contribution (or lack thereof) to technological development among domestic private enterprises. For instance, more than half of the large FDI and SOE firms interviewedas part of the small survey conducted by the World Bank considered firms located overseas, rather than domestically based companies, to be their most important group of suppliers. Small domestic firms tended to supply other private domestic firms or low- end SOEs, but rarely to FDI companies or to high-end SOEs. A clear illustration i s provided by 54 EFFICIENCY AND COMPETITIVENESS Fujitsu, which was the largest FDI exporter in 2004. On that year, it recorded sales abroad for over 400 million dollars, but spent 94 percent o f this amount on imported inputs. Unlike other East Asian countries at an early stage of their industrialization, it would appear that Vietnam has not integrated well its exporters with its local producers. Box 3.1: LinkingFarmersto World Markets The production and commercialization o f rice and tea affects, in one way or another, a substantial share o f the Vietnamese population. Some 10.3 million households cultivate rice. About 400,000 households are involved ingrowing tea; many o fthem are from ethnic minorities. Inthe case ofrice, a high share ofthe export price (about 80 percent) is appropriatedby farmers, but the export price is not particularly high. Vietnam is indeedthe second largest exporter o f rice in terms o f volume, but only the fourth interms o f revenue. The main reason for the low price is the large proportion o f "broken rice" in total production. This is partly the result o f a very fragmented rice- processing system, leading to the mixing o f paddy varieties by a large number o f small de-huskers. Mixing reduces the incentives for farmers to supply high-quality7less-breakable paddy. Other reasons contributing to the low quality o f rice are the outdated technology o f small domestic millers and processors, the preferential access o f state-owned mills to government contracts and credit, a government focus on export volumes rather than quality, and the lack o f appropriate seed standards. Several measures can be identified to add hrther value to the rice chain. Improving infrastructure would reduce transport costs, and bring farm-gate prices even closer to world prices. The promotion o f farmer organizations that can negotiate on quality and price with millers and traders, would be another important measure. Proper commercial incentives could help improve the quality o f rice seed. Finally, the development of niche products such as aromatic rice, organic rice, and varieties well adapted to the uplands could also lead to higher export prices. In the case of tea, on the other hand, barely more than 40 percent of the export price is appropriated by farmers. The analysis shows that yields are low, product quality is poor, and transaction costs indistribution are high. Two main supply channels operate inparallel. Some farmers produce directly for export markets, through contracts with traders and processors. Many o f these farmers lease their land from a processing company, under the condition that they only grow tea on it, and have to sell it later to the company. However, about 90 o f the farmers are "unlinked" and operate through a second channel, which consists o f selling their tea on a free market, without any coordination or linkage to buyers. The final step in the chain, the export o f tea, traditionally took place through SOEs. However, since 2003 a growing share of exports i s conducted through the private sector. Contract farmers tend to have lower production costs than unlinked farmers. In addition, they have somewhat higher yields and profits. Contract farmers also enjoy other important benefits such as stable sales and prices, and technical training, inputs and credit. Unlinked farmers, by contrast, are constrained by limited access to high-value markets. As unlinked farmers form the bulk of tea growers in Vietnam, and many o f them are poor, an important way forward for them appears to be closer integration inthe value chain. Collective action, channeled through producer associations, could help develop better links with processors and traders, from a stronger bargaining position, so as to shift towards higher quality levels and to mobilize capital accordingly. The private sector can also play a major role by establishing clear quality standards across the value chain. Improvements in the marketing o f tea products are needed as well. But this will requirethe identification o fnew overseaspartners. Tapping the growing domestic demand for specialty teas would also contribute to diversify output and reduce risk. Source: Based on Agrifood Consulting International (2002), Arulpragasam and others (2004), and ADB (2005a and 2005b). 35 To some extent, the bypassing of domestic suppliers is not totally surprising. Increasingly, companies that operate in internationalmarkets tend to have a production network transcending national borders. The geography of Vietnam can only support this trend. A Hanoi- based enterprise is 1,400 kilometers away from HCMC, but only 1,000 kilometers away from Bangkok and 800 kilometers away from Guangzhou. Admittedly, a few FDI-dominated industries do have a high rate of local procurement. Motorbike production is a case inpoint, as some 40 to 80 percent of the procurement of FDI companies is done locally. But even there, the "virtuous circle of assemblers and parts suppliers" has not reachedthe same level o f development as incompetitor countries such as Indonesia and Thailand. However, it i s important to keep in mind that business-to-business transactions are only one vehicle for the transmission of best practices inproduction, management or marketing. Other important channels include quality certifications and outright imitation. Certifications provide an independent quality "stamp" which can go a long way towards reassuring perspective buyers or contractors, regarding the quality standards of the concerned enterprise. Imitation operates through turnover inpersonnel, as skilled workers move from FDIfirms to their local counterparts, and when middlemanagers of those firms start their own businesses. Inboth cases, knowledge is transferred through people. But imitation can also be basedon outright copying. Whatever the mechanism, there is some evidence that the higher technological standards o f FDI companies are gradually permeating domestic firms, in one way or another (Box 3.2). It also appears that the impact i s larger on small and medium private domestic firms, while it seems to be negligible on SOEs. This result needs to be interpreted with caution, as it could be related to the age of the enterprises, rather than to their size or ownership. However, a smaller impact on SOEs is plausible, and this inturn stresses the importance of their restructuring ifthe country as a whole i s to be competitive. Enterpriserestructuring Ownership transformation has been one of the main mechanisms used so far to restructure the state sector. As more than 2,500 SOEs have gone through this process already, and another 900 should follow over the next couple of years, it i s important to assess whether ownership transformation actually leads to a better performance. Assessments o f total factor productivity gains across businesses o f different sorts, reported inthe previous chapter, suggest an affirmative answer. But such assessments should not be taken too literally, because the data are not reliable enough and the methodology can be questioned. A rigorous answer can only be based on an in-depth analysis o f the performance o f SOEs that were subject to ownership transformation. Fortunately, this analysis exists. In2005, CIEM conducted its secondpost-equitizationstudy, focusing on a sample o f 400 SOEs, randomly selected among those which had completed the ownership transformation process at least one year earlier. The sample size was similar to that of the first post-equitization study, which had been conducted in 2002. But the response rate was much higher. In 2002, information could be collected on only 60 percent of the SOEs approached by CIEM; in 2005 the response rate climbedto 94 percent, dispellingconcerns that the results could be tainted by a bias inthe responses(ifrespondentswere systematically different from non-respondents). The assessment o f the consequences o f equitization by survey respondents was overwhelmingly positive. Some 88 o f the enterprises declared that their performance was better or much better than before equitization, whereas 9 percent considered that it had not changed 36 EFFICIENCY COMPETlTlVENESS AND much. Enterprises where the state retained only a minority share were slightly more upbeat in their assessment, but the difference was not statistically significant. Box 3.2: Learning from FDI Enterprises A recent econometric study tried to assesswhether Vietnamese f m s operating ina sector with a large FDIpresence were more productive than otherwise similar f m s . The study was conductedusing detailed data from more than 12,000 firms from GSO's enterprise census of 2001. The sample included 5,673 domestic private f m s , 4,895 SOEs and 1,456 FDI companies (both joint ventures and wholly foreign-owned enterprises). The focus of the analysis was on the determinants of value added per worker across firms. The approach allowed disentangling two different effects of FDI on productivity. One, direct, concerned the gap in productivity between FDI companies and other f m s . Not surprisingly, after controlling for other enterprise characteristics, it appeared that FDI companies were indeed more productive than the rest, regardlessof whether a majority or a minority of capital was foreign. The gap with domestic f m s was biggest inthe mechanicsand electronics industry. The other effect, indirect, concerned the gap in productivity between domestic enterprises, depending on the output share of FDI companies in the sectors they were operating in. A fine disaggregationwas used for this analysis, with sectors defined at the fourth-digit level of the industrial classification. After discarding observations with missing information, the sample for this analysis consisted of 9,590 domestic enterprises, of which 2,865 in food processing, 1,121 in textiles and garments, and 381 inthe mechanics and electronics industry. The results showed a positive impact of the FDI share of sectoral output on the labor productivity of domestic enterprises. The impact was statistically significant inthe case ofthe food processingindustry and oftextiles and garments. Interesting patterns emerged as well across businesses of different sorts. Within the private sector, the positive impact of FDI presence was larger among SMEs than among large-scalefirms. On the other hand, FDI presence had no effect on the productivity of SOEs. These patterns need to be interpreted with caution, as the study did not control for firm age. It could simply be that younger firms are more flexible andresponsivethan older ones. And younger firms tendto be both private and small. Assuming that the findings are right, it is interesting to speculate why FDI presence has no impact on SOE productivity. The authors of the study attribute this result to SOEs' lack of management flexibility, in terms of investment decisions and labor management. Such lack o f managerial flexibility would prevent them from translating into practice what they learn from the FDI f m s operating intheir sector. Source: Basedon Anh Thi Tue Nguyenand others (2005). Objective indicators o f performance, measuring the change between the year preceding equitization and 2004 confirm the respondents' views (Figure 3.4). The vast majority o f indicators considered improved after equitization; some o f them substantially. Only the number of permanent workers declined, but this mainly reflects the initial over-staffing o f the equitized SOEs, estimated at around 14 percent o f their workforce. By 2004, less than 4 percent o f the equitized SOEs were making losses. For the rest, the average rate o f return was 17 percent, three times higherthan inthe year precedingequitization. 37 BUSINESS Figure 3.4: Key Performance Indicatorsafter Equitization Average salary Investment infixed assets Productivity Number of permanentlabor Rateof returns on turnover After-tax profits Pretaxprofits Turnover -0.05 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 1 Average annual growth ~ Source: CIEM and World Bank (2005). Figures are median values across respondents, measuredfrom the year before equitization until2004. These results appear to be reliable. Inthe case of the 2002 survey, it could be argued that the positive assessment resulted from the fact that the "best" SOEs were equitized first. Indeed, inthe pilot equitization period, priority was given to the selection o f efficient enterprises rather than loss-making enterprises, in order to attract more investors. Equitization could then be seen the result o f good performance, rather than the other way around. By now, however, the fraction o f loss-making enterprises among those selected for equitization is much higher. It could also be argued that some of the indicators are over-stated precisely as a result o f the policies underlyingthe equitization process. For instance, the value o f investment could be exaggerated due to the necessary asset re-evaluation. And after-tax profits could have grown mainly because o f the tax break equitized enterprises get during a period o f two to four years. While all this i s possibly true, the increase in labor productivity, turnover and pre-tax profits suggests that the improvement in performance i s real. This improvement is not due to a change inthe management team o f equitized SOEs. In more than 80 percent o f the cases, the chairman o f the Managing Board, the director, the deputy directors and the chief accountant remained intheir jobs. Workers are the main shareholders, and outsiders only hold, on average, 10 percent o f the capital o f equitized SOEs. Moreover, there i s a clear concentration of power, as the chairman o f the Managing Board is also, in roughly 70 percent o fthe cases, the director o f the enterprise. The real change i s in the motivation and autonomy of the management team. More than 96 percent o f the respondents to the 2005 survey stated that their managers paid more attention to performance, and enjoyed a higher autonomy to pursue profit objectives. This higher autonomy came mainly at the expense o f the old supervisory agencies (Figure 3.5). There are also some indications, though not exceedingly strong, that equitization curtails access to credit from SOCBs. Four fifths of the respondents see no change in the relationship between their enterprise and state agencies such as the tax office, customs or land administration agencies. But the fraction drops to a half when asked about the relationship with banks and 38 financial organizations. Some 28 percent of the respondentsdeclare that the relationship became more difficult, especially inthe case o f SOCBs and the DAF. Figure 3.5: Who GainedPower with Equitization? 1.5 1 0.5 0 -0.5 -1 -1.55 Source: CIEM and World Bank (2005). The bars indicate the average score to a question on the changing influence of concerned stakeholders, based on the following options: much more (+2), more (+l), no change (0), less (-1) and much less (-2). Itdoes not follow that equitization solves all the corporate governance problems faced by these enterprises. Considerable confusion remains regarding asset ownership, especially on the transfer o f LUCs, the distinction between land allocation and land lease, and the documentation certifying ownership o f the main technological lines. As a result, 38 percent o f respondents say that ownership rights are similar to those of non-equitized SOEs, 20 declare that they are similar to those of private enterprises, and a staggering 42 percent answer "both". In practice, land continues to belong to the state, and equitized enterprises cannot use the corresponding LUC as collateral when trying to obtain credit from banks. Which shows that equitization i s a model with interwoven features from both SOEs and the private sector. A comprehensiverating The list of potential determinants of a country's ability to prosper in the global market place is bound to be long. Trying to nail it down to a few key features, such as the regulatory framework firms operate in, is bound to give a partial picture o f a country's relative position in the global market place. Productive efficiency and international competitiveness depend on a range of factors, from macroeconomic stability to sound policies for business to high-quality human resources to limited corruption. While it i s perfectly legitimate for different cross-country ratings to focus on specific determinants of business performance, a comprehensive assessmentof the strengthsandweaknesses o f a country needsto take them all into account. 39 Such i s the idea behindthe so-called Country Policy and Institutional Assessment (CPIA) o f the World Bank, and also of similar indices increasingly used by other multilateral and bilateral organizations. The CPIA i s a tool to decide how much support a country should be given. CPIA ratings are reviewed on an annual basis. A high CPIA is in principle associated with potentially better economic performance, which in turn could make foreign aid more effective. Regardless of its aid implications, a rating that i s considered a reasonable predictor o f economic performance could arguably be used to identify a country's strengths and weaknesses, and to produce an overall rating for its position among other developing countries. The CPIA covers four main areas. These are: macroeconomic management, structural reform, social inclusion and equity, and public sector management and institutions. Each o f these four areas is disaggregated into a series of more specific indicators. Thus, macroeconomic management includes macroeconomic stability, fiscal policy and debt policy. Structural reform refers to trade integration, financial sector policies and the business regulatory environment. Policies for social inclusion and equity consider gender equality, equity in the use o f public resources, human resource development, social protection and labor and environmental sustainability. Finally, public sector management and institutions covers property rights and governance, budgetary and financial management, the efficiency of revenue mobilization, the quality of public administration, and transparency, accountability and corruption. Therefore, the CPIA includes 16 indicators in total, each of them further disaggregated into a series o f specific questions, with examples given for possible answers. Figure 3.6: Vietnam's Strengths and Weaknesses ,a I Macroeconomic management I Transparency, accountability & corruption policy Quality of public administration,, bt policy Efficiency of revenue mobiliz Trade 1 Public finance __ft___ Financial sector ! ! Property rights & governance Business regulatory environ-t Pol & institut for environ sustain s e y e n d e r equality Social protection & labor Equity of public resource use Human resources -Vietnam -East Asia 1 Source: Basedon data from the World Bank. A rating from one to five is thus generatedfor each of the 16 indicators inthe four policy areas, with five meaning highly satisfactory and one meaning unsatisfactory. The specific 40 EFFICIENCY AND COMPETITIVENESS questions behind each of the 16 indicators are answered by the World Bank teams working on each (and often in each) of the countries, but subject to a vetting process aimed at strengthening international comparability. As a first step, the process involves benchmarking with respect to two countries in each region. While not being perfect, the CPIA is by now one o f the most reliable cross-country rating systems available. When considered on an indicator-by-indicator basis, the resulting ratings help identify a country's strengths and weakness. When taken together, under the form of an aggregate rating, the provide information on the country's relative position inthe world. Disaggregated CPIA ratings suggest Vietnam's main strengths are in macroeconomic management and in fiscal policies. Public financial management gets relatively highratings too. Vietnam also does well in the social area, including gender equality, the equity in the use o f public resources and the quality of its human resources. But Vietnam still has important weaknesses. One of them is structural reform: ratings are low for trade policies, the financial sector and the regulatory environment for businesses. Vietnam's other weakness concerns the institutional area, with the ratings on the efficiency of revenue mobilization, on transparency, accountability and corruption, and on social protection, being below the average for other East Asian countries (Figure 3.6). Figure3.7: Vietnam Ratings inInternationalPerspective I number of countries ($35) - 35 29 3 . 0 a n d 3 . 1 3.2 3.3 3.4 3.5 3.6 3.7 3.8 3 . 9 4 . 0 a n d below UP Score I Source: Based on data from the World Bank As for the overall rating, in2004 Vietnam scored 3.7 ina scale from one to five. Among the 135 developing countries and transition economies for which the CPIA was computed, there were another 13 with the same rating as Vietnam, and 38 with higher ratings. In relative terms, this means that Vietnam is somewhere between the 36* and the 5lSt ofthe worldwide percentiles distribution (Figure 3.7). But its position improves substantially when only the 81 low-income o f countries, Vietnam falls between the loth and the 22"d percentile, or roughly in the second countries inthe group are considered, as only 13 o f them score higher than 3.7. Withinthis group decile of the distribution. 41 4. THE INVESTMENT CLIMATE I o r Vietnam's business sector to continire its expansion, and support the attainment o f developnient objectives such as economic growth and poverty reduction, it is necessary to remove the constraints faced by enterprises. Some o f those constraints can be identified through quantitative analysis: for instance, on the determinants of productik ity gains over time, or on competitiveness mtings across countries. However, i m c h the same as a poverty assessment lacks credibility if tlic poor have not been heard, an evaluation o f the investment climate requires listening to entrepreneurs. A recently completed investment climate survey for Vietnam. with a representative sample made o f domestic private enterprises, SOEs and foreign companies. allows benchmarking the constraints they face. For more than one third o f the respondents, access to credit i s reported as a major or severe obstacle to business growth; access to land i s the secoiid most important constraint; insufficient ckills and education o f the workforce. and poor transportation infrastructure, rank third and fourth. In all four cases (exccpt maybe for skills) tlic severity o f the constraints i s significantly higher in Vietnam than in the rest o f the East Asia region or in the rest o f the world. On the other hand, some of the issues emphasi7ed by competitiveness rankings. such as the legal system, bureaucratic procedures and corruption. are downplayed by respondents. The ability to conduct business-to-business transactions on a trust basis, or using rudimentary but reliable enforcement r~~eclianisms,may explain the low importance attached to the legal system. The coiisiderable simplification of procedures introduced in 2000. and reinforced through mechanisms like OSSs, could account for the downplaying o f bureaucracy and red tape. The lack o f importance attached to corruption i s more unexpected. This result i s further probed using a variety o f sources. `l`hey all yield a consistent picture, one where corruption directly affecting businesses i s quite prevalent. but petty. But this is not to ignore other modalities of corruption in Vietnam, as in SOEs and in public investmciit pro-jects; these other modalities may not affect businesses directly but they do undermine governance and society more broadly. The views of entrepreneurs Investment climate assessments draw on the experience o f businesses in operation to pinpoint the areas where reform i s most needed. I'heir cornerstone i s the ICs, typical14 produced by the World Rank and ADR in partnership with a local research institution, The ICs asks eiitrcpreneurs in developing countries about the difficulties they encounter in starting and running a business; and, ifthe business fails, in exiting. It also captures their experience in a range o f areas: financing, governance. regulation, tax policy, labor relations, conflict resolution, infrastructure services, supplies and markcting, technology and training. All these are areas where difficulties can add substantially to the costs o f doing business. The ICs attempts to quantify these costs using0b.jecti.e measures, such as monetary costs or time. The ICS for Vietnam, conducted by the International Development Center of' Japan and Concctti in the summer o f 2005, has one o f the biggest samples in the norld. It comprises 1,150 firms, located in 35 provinces across five of the eight regions o f the country, which gives it an unusual regional coverage. Among 58 surveys of this type completed so far, only those for China (3,948 enterprises), India(2,722), Brazil(1,642) and Thailand (1,385) had bigger samples. Inpreparingthe ICs, the overall enterprise population ofVietnam was first identified, by combining GSO's enterprise censuses with data on business registrations from NBIC and data from provincial tax authorities. This ledto a listing of more than 60,000 enterprises in operation by 2004. Within this population, the target of the ICs were manufacturing firms with at least 10 employees in the Red River Delta, the North Central Coast, the South Central Coast, the Southeast and the Mekong River Delta. This represents a total of 9,632 enterprises, including foreign direct investedcompanies, private firms, un-restructured SOEs and equitized SOEs. Table 4.1: BindingConstraints inVietnam and Elsewhere Constraint East Asia Vietnam The World Access to finance 17.4 ** 37.4 30.1 ** Access to land 9.9 ** 26.4 14.5 ** Labor skills and education 23.8 22.3 20.4 * TransDortation 15.2 ** 21.6 12.4 ** Cost of finance 20.2 21.3 36.1 ** Macroeconomic instabilitv 34.1 ** 16.8 40.2 ** Electricity 24.4 ** 15.7 24.4 ** Policy uncertainty 32.5 ** 14.7 40.2 ** Tax rates 28.2 ** 13.8 40.5 ** Corruption 28.6 ** 12.8 36.8 ** Customs and trade regulation 20.1 ** 12.5 21.6 ** Anti-competitive behavior 21.6 ** 12.3 29.7 ** Labor relations 17.4 ** 10.9 17.3 ** Tax administration 22.4 ** 8.7 32.4 ** Telecommunications 12.9 ** 6.5 10.3 ** Legal system 1 27.3 ** I 5.5 1 2 1 . 6 * * 1 Crimeand theft 19.3 ** 4.0 25.7 ** Licenses and permits 14.4 ** 1.4 15.9 ** Source: ICs database of the World Bank, usingun-weightedaverages. Figuresindicate the percentage o f firms which consider that the constraint is either major or severe. Figures for East Asia are based on observations fiom Cambodia, China, Indonesia, Malaysia, the Philippines and Thailand. Figures for the world include 57 developing countries and transition economies, but exclude Vietnam. An asterisk indicates that the figure is statistically different from that for Vietnam at the 90 percentconfidencelevel; two asterisks indicate99 percent confidence. The wealth o f the responses i s such that an entire, separate volume is devoted to their interpretation. Here, it i s usefulto focus on what i s perceived as "severe" or "major" constraints for the growth of business, out o f a list including 18 possibilities (Table 4.1). The top constraint inVietnam is access to finance; itwas identifiedas a severe or major obstacle to business growth 43 by more than 37 percent of respondents. At 26 percent, the second most important constraint is access to land. Next come labor skills and transportation. These are in fact the only four cases in which constraints appear to be more severe than elsewhere inthe region, or across all developing counties. To be precise, the skills constraint is only marginally more important inVietnam than it i s worldwide; and it is not more severe (in statistical terms) than in other East Asian countries. As for the other three top constraints, their high ratings are not surprising,as they clearly match anecdotal assessments. It should be noted that the top four constraints are the subject o f the secondpart o f this report, whose chapters focus on land, finance, labor and infrastructure. All other constraints are significantly less severe inVietnam than inthe region or across all developing countries, some of them by a vast margin. Again, in many cases the low ratings could be expected. For instance, most observers would agree that Vietnam i s characterized by macroeconomic stability, low crime, and predictable government policies. It i s also clear that enormous progress has been made in terms of increasing access to electricity and telecommunication services, and reducing their cost. However, the rating for some o f the constraints defies the conventional wisdom. This i s the case, in particular, for the legal system, licenses and permits, and corruption. Because of the importance attached to these constraints in the public discourse, a closer examinationis warranted. Property rights Only 5.5 percent o f ICs respondents see the legal system as a severe or a major constraint inVietnam, compared to about a quarter elsewhere inthe region or inthe world. This is despite the fact, widely recognized, that Vietnam still has enormous legal development needs, and a poorly functioning judiciary. Enforcing a contract through the courts would actually be considered a hopeless endeavor by many. Which casts doubts on the effective protection o f property rights inVietnam: how can commercial obligations be honored, or assets transferred, ifa written contracts has little more legal value than a piece of paper? In light of the obvious weaknesses of the legal system, why do so few entrepreneurs rate it as a severe or major constraint? One possible answer i s that the enforcement of property rights operates through more informal, but not dramatically less effective mechanisms. Two examples stand out inthis respect; they refer to business-to-businesstransactions and to landsales. The small-scale survey of enterprises conducted by the World Bank, mentioned in previous chapters, allowed analyzing in detail how business-to-business deals are conducted in Vietnam. All the firms in the survey do indeed use some form o f written contract to formalize their deals. However, none of them places much faith inthe ultimate legal enforcement power of those contacts, nor do they seriously entertain the possibility of resorting to legal action if the other party violates its obligations. The main purpose of the contracts seems to be to minimize the room for misunderstanding between the parties. And although very few firms adhere strictly to their terms, suppliers generally attempt to fulfill them. They do so for the sake o f building their reputation and boost the goodwill of the contractor, rather than because they are concerned about the possibility of any legal consequencesof a breach. Reputation, through repeated interaction, i s the cornerstone of business-to-business deals inVietnam. Invirtually all cases, contractors and suppliers have known each other for a number of years, usually several times longer than the term o f their business deal. In the small-scale survey, 98 percent of the contracts were with suppliers that the contracting firm had known for three or more years, whereas the longest contractual agreement was for 12 months. 44 INVESTMENT CLIMATE Apart from carefully selecting their counterparts, the contractors in the sample use a range o f mitigation strategies to protect themselves from a contract breach. For instance, they keep their contracts relatively short-term, with an option to renew. This is to limit exposure, but also to provide an incentive for the supplier to perform, in the expectation o f getting a new contract. Contractors also limit the size o f individual contracts, and diversify suppliers wherever possible. They tend to transfer the financial risk to the supplier, by withholding payments until the terms o f the contract are fulfilled. In some cases, contractors even withhold payments on previous contracts, as a device to guarantee the timing and quality o f the delivery. And they often use guaranteed accounts (effectively frozen deposit accounts in commercial banks) to ensure that parties abide by their commitments. Box 4.1: A Market without Property Rights: Real Estate inHCMC InHCMC severalnewspapers were posting real estate advertisementsas early as 1993. These listings offer insight into how the market was valuing property, in spite of a very limited legal development. Some sellers advertisedthat their property had giay to hop le (literally meaning "legal papers"), others that it had chu quyen tu nhan ("house ownership") and others mentioned both. "Legal papers" are documentsthat can be usedto apply for atitle, underthe form of a BuildingOwnership and LUC. The process involves several administrative steps, through various levels of government, and requiresthe payment of official andunofficial fees. An analysis of the determinants of listed prices was conducted based on a compilation of listing from Tuoi Tre, the leading newspaper in HCMC. Tuoi Tre began a real estate section after the Land Law established private transferable land-use rights. While in the beginning there were few listings, their number increased rapidly over the years. Some of them were posted by development companies and landbrokers; but a vast majority was by owners. The information providedby sellers inthe advertisementsincludesthe address of the property, the dimensions of the house and land parcel, the number of floors and bedrooms, whether it is connected to water, electricity, and phone, whether it has a yard, its number of toilets, its proximity to markets and schools, whether the ground floor i s convenient for commercial activities, and whether the property is ina quiet and safe neighborhood. Many also indicate their property rights status. The data analysis was based on listings from the spring and fall of every year between 1998 and 2001. The results are plausible. Prices increase with proximity to the city center, with the size of the house, with its proximity to the street, with amenities such as toilet and telephone, and with the quality of the neighborhood. Prices also increase over time, reflecting the upward trend inHCMC and other urban areas of Vietnam. The mention of any kind of legal documentation i s associated with higher property prices; on average, by sevenpercent. However, properties havingboth a title and "legal papers" have list prices which are significantly higher than those of properties offering having only a title, even after controlling for all other characteristics. This resultwould not make any sense ina western legal environment, where the title is all that matters. The fact that "legal papers" add to the value of the property shows that the title is just one among several documents helping make a case out of courts, mainly with neighbors and local governmentauthorities. Source: Basedon Annette Kim (2004). From a Western perspective, it is equally surprisingto learn that many land transactions in Vietnam involve no contract. In the western legal context, there is an entire institutional 45 apparatus where a set of private professionals (including insurers, lawyers, brokers and notaries) make sure that the property title is clear before buyers and sellers can transact. But in Vietnam the real estate market emerged, and boomed, despite incomplete legal titling. In the absence o f alternative institutions, a majority o f property disputes i s negotiated with neighborhood block committees and ward officials, under the guidance o f accepted social norms. So, in effect, one has property rights over land if neighbors and local government officials agree. Possessingmore documentation on the property helps inthe negotiationwith neighbors and local officials. Having an LUC i s better than havingjust "legal papers", and the latter are still better than nothing at all. But in the end, there is a market price for every piece of real estate. The price is higher when "legal papers" of one sort or another are available. But it does not drop to zero intheir absence (Box 4.1). However, it is by no means clear that larger business transactions can be conducted inthe same way. For instance, investors willing to launch large housing and development projects would be reluctant to operate on "legal papers" only. The lack o f a modern legal and judiciary system may not be a major problem for small transactions, where trust-based mechanisms do well, and for large firms, which can get more direct access to government if needed. But it i s probably an obstacle for small businesses to grow, and fillthe "missing middle". Bureaucracyand red tape A mere 1 percent of the ICs respondents in Vietnam see licenses and permits as a "severe" or a "major" constraint, compared to an average of about 15 percent elsewhere, both in the region and across all developing countries. This positive perception i s a testimony to the dramatic simplifications inbusiness registration introduced by the Enterprise Law o f 2000. This law also led to a considerable reduction in the number o f licenses in force. The positive perception on licenses and permits might be also reflecting the success in introducing the OSS model for administrative procedures at the local level. But it still defies the conventional wisdom. To some extent, the gap between the ICs responses and the conventional wisdom could just be semantic. For instance, a survey of enterprises jointly conducted by CIEM and the German Agency for Technical Cooperation (Gesellschafl fur Technische Zusammenarbeit, GTZ) suggests that before an enterprise can be in operation, it needs to go through 13 administrative procedures, taking 260 days altogether. However, 231 days in this estimate correspond to the time neededto acquire land, under the assumption that the enterprise wants to purchase it from the government and needs to go through land clearance and compensation of resettled populations. Gettingland in an industrialzone would reduce the waiting period, on average, from 231 days to 83. And delays would be much shorter if the land was purchased or rented from other enterprises. As regards the more narrow business registration process, the Enterprise Law requires its completion within 15 days. The CIEM-GTZ study found that this deadline was respected in two thirds of the cases, with only 5 percent o f enterprises having to wait for 30 days or more. In many provinces the registration process was actually completed in less than one week. Similarly, 70 percent of the companies surveyed receivedtheir tax code within seven days, and only 6 percent of them had to wait for more than 15 days. This kind of expeditious administrative response is probably what ICs respondents have in mind, whereas the 23 1 additional days are likely seen as a severe or major constraint in having access to land, not in registrationor licensing. Findings were similar in the three-round survey o f SMEs conducted by ILSSA, already mentioned inprevious chapters. An impressive97 percent o f the manufacturing firms covered by 46 INVESTMENT CLIMATE this survey reported that licensing and operating permits were either no obstacle or only a minor obstacle to their business. On average, only five to six days per year were lost to administrative procedures. Also, few respondents saw cumbersome governmental regulations and official attitudes as a severe burden. On the contrary, about 61 percent of urban enterprises had received some form o f assistancefrom the government, mainly under the form o f helpwith application and registration formalities. The relative cordiality of government agencies is actually one o f the salient findings o f the ICs for Vietnam. Respondents were asked to choose, among a long list o f options, the two agencies which had been the most supportive, and the two which had been the least helpful. At one end o f the spectrum, DistrictPeople's Committees were perceived as the friendliest, followed by the Business Registration and Licensing offices under DPI. The traffic police came at the other end (Figure4.1). The perceptionwas quite negative about the Customs Department as well. In between, the two extremes, the Tax Authority was seen as most supportive by some respondents, and as least helpfulby others. Figure 4.1: Government Agencies: Friends or Foes? Least Helpful Most Helpful -30 -20 -10 0 10 20 30 District People's Committee Business registration Labor and social security Product quality registration Fire and building safety Tax authority Municipal police Customs department Department of construction Environmental agency Land administration agency Market controller Traffic police Source: Basedon ICs by the World Bank. Tax officials are seldom popular, anywhere in the world, which makes the mixed views about them in Vietnam somewhat intriguing. One possibility i s that they are willing to keep tax payments down, maybe in exchange for "gifts". Indeed, it i s generally believed that the amount o f taxes payable by businesses in Vietnam is determined through negotiations with tax officials. However, the ICs does not confirm this presumption, as most respondents declare that the tax was either completely or largely determined by the firm itself. In addition, aggregate data contradicts the idea that tax payments are set at low levels. The "productivity" o f Vietnam's 47 VAT, meaningthe revenueobtained for eachpoint ofthe main VAT rate, is among the highestin the world. The productivity of CIT is about average. The reasonwhy so many ICs respondents findthe tax department helpfulis therefore unclear. The very low percentage of ICs respondents who see licenses and permits as a serious constraint also clashes with the perception that their number has beenincreasingsurreptitiously in recent years. The Enterprise Law led to the removal o f 159 licenses o f all sorts; it also limited the power to introduce new ones to a narrow list of national authorities. However, this has created some confusion. For some, "license" refers to any type of document issued by a competent authority that enterprises are required to obtain in order to legally conduct their business activities. Permits, certifications, authorizations, approvals, attestations and the like, all fall under this broad definition. The vagueness of the concept makes it difficult to determine whether "licenses" are indeed making a discrete comeback in recent years. According to an estimate by VCCI, they are. From 194 licenses inforce in2002, their number would have increased to 246 in 2003 and to 298 in 2004. Moreover, 55 percent of these licenses were apparently issued by provincial authorities, and another 22 percent by local authorities at district, commune and village levels, which seems to contradict the Enterprise Law of 2000. It is still possible to reconcile this upward trend with the low percentage of ICs respondents who see licenses and permits as a serious constraint. To begin with, at around 300, their number would be low by internationalstandards. For instance, the radical regulatory reform introduced by Korea in the aftermath of the East Asian crisis slashed the number o f business regulations by half; but there were still 6,308 of them in force once the process was completed. Also, some of the licenses and permits are sensible. Thus, a concrete mixer struck serving construction sites in Hanoi needs a license to circulate inthe city (and only at night). This license i s issued for each construction site and i s valid for one month only. This i s a common practice in other countries. Other licenses are clearly questionable, especially inthe services sector. Inareas such as accounting, professionals with university diplomas are required to get further accreditations, including weeks o f mandatorytraining, before beingauthorized to work. Barriers o f this sort may be downplayed inthe ICs survey becauseof its focus on enterprises inmanufacturing. Corruption Perhapsthe most surprisingresult of the ICs for Vietnam i s the relatively low importance attached by respondents to corruption. In 2005, the Corruption Perceptions Index of Transparency International gave a rating of 2.6 to Vietnam, on a scale o f one to 10 where 10 corresponds to the highest integrity standards. Most other countries in the region got higher grades, including Malaysia (.5.1), Korea (5.0), Thailand (3.8), Laos (3.3) and China (3.2). Only the Philippines (2.5) and Indonesia(2.2) appearedto be more corrupt. Yet, according to the ICs, corruption is less severe in Vietnam than in any other developing country in the region, save Malaysia (Table 4.2). Can this inverse ranking be trusted? One possibility is that entrepreneurs in Vietnam simply have learned how to live with corruption, to the point where they don't see it as a constraint anymore. The ICs uncovers bribes and gifts in dealings with most government agencies. For instance, the average bribe payment is estimated at approximately 3.6 million dong in the case of customs, 3.4 million for tax administration, 1.9 million for the police and for market controllers, 1.4 million in the case of environmental agencies and 1.2 million for business registration and licensing. The average bribe i s less than one million dong in the case of district People's Committees, labor and social 48 security, fire and building safety and construction inspectors. However, many respondents declare that they do not pay bribes. Table 4.2: Corruptioninthe East Asia Region 1II s corruption a constraint to business? (percent o f responses) No 1 Minor I Severeor major 1I Cambodia 4.7 39.4 55.9 China 24.1 48.5 27.3 Indonesia 29.3 29.2 41.5 Malaysia 53.8 31.7 14.5 Philippines 40.6 24.3 35.2 Thailand 49.7 32.1 18.3 Vietnam I 52.3 I 17.8 I 14.2 I Source: ICs database ofthe World Bank, using un-weightedaverages. On average across all agencies, taking into account the fact that no unofficial payments are made in many cases, the average bribe is 1.8 million dong. This i s consistent with the findings o f the pilot corruption diagnostic study by the Internal Affairs Committee o f the Party. This diagnostic study is based on the "triangulation" o f responses from households, enterprises and government officials. Its enterprise sample was randomly drawn from a population which was similar to the one considered by the ICs. Butthe overlap betweenthem i s probably minimal, or non-existent. The pilot corruption diagnostic study estimates that the average additional cost per transaction ranges from 0.1 to 2.1 million dong. Some of these payments take place several times a year. Yet, the overall amount o f bribes paid by enterprises inVietnam is not high when compared to other developing countries; certainly not as high as could be expected, given Vietnam's development level (Figure 4.2). From this perspective, the picture that emerges is one of a country where corruption i s widespread, but petty. Another possibility is that the ICs responses in relation to corruption are biased or inaccurate. Yet, there i s substantial coincidence between those responses and the findings o f several other studies, based on entirely independent data sources. First i s a previous, ICs-like ' survey conducted by MPDF in 2001. Among other possible constraints, the questionnaire asked whether corruption was a severe obstacle, a major one, a minor one or no obstacle at all. The scores associatedto these answers were four, three, two and one respectively. The average score for corruption was 2.02, indicating that it was perceived as a minor constraint. Second i s the three-round survey o f SMEs by ILSSA. Again, "informal fees" did not appear to impose a very heavy burden on business, although about 70 percent of respondents stated that they had paid such fees inthe past year. Third is the small survey of enterprises conducted by the World Bank. This survey paid particular attention to inspections and visits by officials from various government agencies, such as the police, firemen, market controllers, officials from information and culture departments, health staff from the labor department, and environmental inspectors. Based on the responses provided by interviewees, MPDF grouped the responses into three broad 49 categories: demanding, neutral, and undemanding. Only a fifth of the respondents had experienced demanding inspection visits. Finally, there i s the survey o f women entrepreneurs conducted by MPDF in the summer of 2005. Some 60 percent of respondents declared that racketeering or the need to make informal payments were not an important concern. Overall, then, there is a broad coincidence with the ICs results. Figure4.2: SpendingonUnofficialPaymentsand "Gifts" I P 10 m ~ m c 0 Nicaragcy 1 ~ b 8 Guatema& *Algeria Cambodia (enya Lithuania 0 I I 0 5,000 10,000 15,000 20,000 ~ , Per capita GDP (in PPP dollars) Source: ICs database o f the World Bank, usingun-weightedaverages. Furtherevidence that ICs responsesare notbiased or inaccurate comes from the resulting ranking o f government agencies, according to their levels of corruption. This ranking is strikingly similar to the one emerging from the pilot corruption diagnostic by the Internal Affairs Committee o fthe Party (Figure 4.3). This, again, gives credibility to the ICs results. However, these results refer only to corruption modalities directly affecting businesses. They do not cover graft in SOEs or in land conversion, two areas where anecdotal evidence and press reports reveal much more serious problems. They do not refer to public investment projects either, an area where outright corruption might not be as prevalent as it i s often claimed, but where collusion by bidders and distribution of public contracts among SOEs remain common practices. Moreover, an average bribe of 1.8 million dong, while small from the point o f view of an enterprise, represents an average monthly salary inVietnam. The implications o f generalized corruption of this sort on social morality are not petty by any means. Together with unofficial payments in social services, from allegedly voluntary "private lessons" in education to unaffordable health care, petty corruption o f this sort i s deeply resented by the Vietnamese population. This can explain the gap between a relatively relaxed business sector, and the perception of generalized corruption reflected in cross-country comparisons like that conducted by Transparency International. 50 Table 4.3: Which Are The Most Corrupt Government Agencies? InvestmentClimate Survey Corruption Diagnostic Survey Traffic police Land administration agency Customs department Customs department Tax department Traffic police Land administration agency Tax department Market controller Regulators inconstruction Construction permitauthorities Constructionpermitauthorities Import/ export license authorities Health care Planning and investment agencies Regulators intransportation Economic police Source: Based on data from ICs by the World Bank and on Communist Party of Vietnam (2005). Inthe case of the ICs, rankings are based on the share of respondents declaringthat corruption is widespread or gifts are requiredto get a favorable decision. Only agencies with a share in excess of 5 percent are reported. A level playingfield? A comparison o f binding constraints across enterprises with different ownership structures shows that not all of them experience the same problems (Figure 4.3). Access to finance is not an issue for foreign companies, whereas it affects in a similar way all domestic firms, regardless o f whether the state owns some o f their capital. Remarkably, equitized enterprises are the ones suffering the most, which suggests that divestiture has been effective at hardening the budget constraint. On the other hand, SOEs are clearly at an advantage compared to all other firms when it comes to access to land. The fraction o f private enterprises and foreign companies seeing land as a severe or major constraint i s almost twice as high. Private sector firms are the outlier regarding labor skills. This is a much less severe constraint for them than it i s for the other, usually larger businesses. A more limited concern about skills is consistent with a situation of high labor turnover, whereby qualified workers learn on thejob inlarge companies, and especially in foreign ones, and then move into takingjobs with bigger responsibilities in smaller enterprises, or even creating their own. On the other hand, it i s foreign companies which appear as outsiders inrelation to transportation infrastructure. This may reflect their much higher needs, especially to support timely delivery of their products to world markets. Enterprises operating mainly in a less demanding domestic market might be able to cope better with the constraints imposed by poor infrastructure development. One important area in which the playing field is not level yet is public procurement. In the ICs questionnaire, this area falls under the broader heading of anti-competitive behavior. And in this respect, Vietnam fares better than other countries in the East Asia region or in the world. But the relatively upbeat assessment of ICs respondents probably says more about the gradual opening of markets to competition, as the country integrates into the global economy, 51 than about access to public contracts. While public procurement has become increasingly transparent inVietnam, some areas remain dominated by SOEs. This i s the case, in particular, o f infrastructure contracts, given that relatively few private enterprises operate in large-scale construction activities. SOEs are increasingly competing with each other, and bidding for projects away from their own provinces. Butreliance on common cost norms, and in some cases outright collusion, reducethe efficiency of public markets. Figure 4.3 : BindingConstraints by EnterpriseOwnership c I i a, Q 30 c .-.-E C 25 5!! 20 ` I 0 .-b 15 2 `3 b 20 E 10 bg! L 0 10 F 5 a, d% > a 0 -< 0- v) -_ , Private Equitized SOE FDI i 1 Rivate Equitied SOE FDI Access to finance Access to land =3 35 c zz 45 30 a, Q 40 .-c 35 1 .-e C p E 30 20 c ln 25 8b 20 `3 E 10 `3 15 E b2%! b 10 5 g! a, 5 $ 0 a, v) 0 Private Equitized SOE FDI Private Equitized SOE FDI Labor skills and education Transportation Source: ICs database o fthe World Bank, usingweightedaverages. On the other hand, the Common InvestmentLaw (CIL) passed in late 2005 represents a significant step towards leveling the playing field, as it applies to all forms of businesses, regardless of ownership or corporate structure. Untilthe passage o f this law, there were separate investment regimes for foreign companies, for domestic private firms, and for cooperatives. The CIL also encourages private participation in a wider range of sectors, with firms being free to invest in any activity that is not explicitly included in a "negative list." This approach provides greater certainty and clarity to investors than "positive lists." 52 INVESTMENTCLIMATE The rural investmentclimate Investment climate assessments based on surveys in the spirit o f the ICs o f the World Bank focus on urban areas. For instance, the I C s for Vietnam included only enterprises with more than 10 employees and was limited to the manufacturing sector. However, there are compelling reasons to pay attention to the investment climate inrural areas as well. That i s where most o f the poor live, and the creation o f wage jobs i s one o f the main avenues out o f poverty. But in addition, the bindingconstraints to business development could be very different. Rural areas have lower population densities, which make infrastructure and many services costly to maintain. Small enterprises, many o f them owned and operated by households, dominate. Markets are less developed and transaction costs are higher. There are also distinct patterns o f seasonality, affecting production, employment, accessibility and migration. And services are generally less developed. Unfortunately, there i s very limited experience to date in implementing rural investment climate surveys, with just a precedent in Sri Lanka and ongoing work in Pakistan and Tanzania. InVietnam, a pilot exercise was conducted duringthe summer of 2005 inH a Tay andVinh Phuc provinces, as a joint endeavor o f GSO, VASS and the World Bank. This pilot rural investment climate survey aimed at covering the full spectrum o f business activities. It included 100 formally registered enterprises, half o f them in manufacturing and the rest in services, some with less than 10 employees. It also included 200 household businesses not formally registered as enterprises, and 100 enterprises which did not have any business. The latter group served to uncover possible constraints preventing business start-ups. The survey was limited to an area defined by GSO as rural. Given the interest in comparing the findings o f the rural investment climate assessment with those o f the ICs, the questionnaires were kept as similar as possible. On the surface, the rating o f constraints is quite different inrural and urban areas (Figure 4.4). A systematic comparison with the ICs suggests that access to electricity, ambiguous policies, and crime appear to be much more serious constraints for rural businesses than for larger urban ones. For instance, more than two fifths o f all formal businesses have a generator to provide back-up during power interruptions. In contrast, access to finance, customs and trade regulations, corruption, and labor issues are more important for rural businesses. 53 Figure 4.4: Binding Constraints inRural Areas 0% 5% 10% 15% 20% 25% 30% 35% ' Source: Basedon pilot RuralInvestment Climate Survey by GSO, VASS and the World Bank. Figures indicatethe fraction ofrespondentsidentifying eachofthe issues as a severe or major obstaclefor runningtheir businesses. However, this comparison i s to some extent deceiving. Households which do not operate a business indicated that the main obstacle for starting one i s lack o f start-up capital. Limited access to finance was a severe or major constraint for half o f the households. As for household businesses, less than one third o f them currently have a loan from a financial institution, compared to roughly two thirds o f formal enterprises. To obtain the loan, borrowers typically have used land and buildings as a guarantee, but the collateral was worth one to three times the amount of the loan. Credit i s typically short terms and mainly used to purchase inputs. Most household businesses without a loan say that they never applied for one. The heavy collateral requirements andthe short-term nature ofthe loans could be among the reasons for limited access to finance among rural entrepreneurs. Businesses in trade and services, in turn, saw access to land as the biggest obstacle, with 28 percent of them reporting it as a major or severe constraint. About a seventh of respondents said that they had to wait for more than 18 months to get their land requests approved and building permits processed. Respondents were also asked to describe the attitude and behavior of public officials towards their business establishment. The answers were overwhelmingly positive and many, especially household businesses, were reluctant to give a negative score to any government agency, with the exception of the traffic police. Reluctance to be critical can be due to the fact 54 lNVESTMENT CLIMATE that survey was conducted by a government agency (GSO),and not by a private consultancy firm, as the ICs. Still, figures on unofficial payments were consistent with those gathered through the ICs. The median spending on bribes and gifts was 2.0 million dong. Bothhouseholdbusinesses and formal enterprises ranked the traffic police as being most corrupt, followed by the customs office. However, only 8 percent o f respondents indicatedthat corruption was a serious constraint for the growth o ftheir business. 55 PARTII: KEYMARKETS AND INPUTS 5. BANKING AND FINANCE Based on entrepreneurs' own assessment, insufficient access to finance i s the most important obstacle to business growth inVietnam. The share of firms that see it as a "severe" or "major" constraint to their development is also significantly higher than in other countries, either in the region or worldwide. This may come as a surprise, given that Vietnam has achieved a substantial financial deepening over a relatively short period of time. The number o f active savings and loan accounts i s high for a country at its development level, whereas banking credit has beengrowing steadily; even too fast, probably, given current deficiencies inthe assessment of credit risk. Reconciling perception and facts requires a better understanding o f bank lending practices in Vietnam, disentangling who has access to credit and under which terms. Looking forward, it i s also clear that the structural weaknesses of the banking system need to be addressed. The poor quality of many loans creates a substantial fiscal burdenfor the government, aggravated by the absence of effective mechanisms to seize the assets of bad debtors. As for formal capital markets, they are still too under-developed to channel a sizeable volume o f resources to new projects. But thriving informal transactions on unlisted stocks raise the prospect of faster capital market development, provided that proper regulations are inplace. Access to credit The last decade has beencharacterized by remarkable financial deepening inVietnam. In the initial stages of economic transition, banks were mainly serving as government windows to channel resources to SOEs. Among individuals, large transactions were often conducted in gold or in dollars. By contrast, total credit represents now more than 60 percent of GDP, and the figure has been increasing steadily (Figure 5.1). The growth inmobilized deposits has averaged 25 percent per year. It has been twice as large in the case o f the Vietnam Bank for Agriculture and Rural Development (VBARD), thanks to its large branch network. The overall monetization o f the economy, measured by the standard ratio between the M 2 monetary aggregate and GDP, has been increasing steadily as well. Needless to say, the financial deepening of Vietnam does not match that o f its neighbors yet. The ratio o f bank credit to GDP is almost twice as high in Thailand, and three times higher inChina and Malaysia. But ina way, this is reassuring. The quality of credit inVietnam is still poor, due to a legacy of policy lendingto SOEs, but also to the inability (or the unwillingness)o f SOCBs to properly assess the risk associated with loan applications. A relatively modest ratio of credit to GDP means that even if the share of NPLs i s high, the problem remains manageable; at least for now. The depth attained by Vietnam's financial deepening is even more impressive when looking at the fraction o f economic units which use the banking system. Based on a rough estimate, there are about 25 million active savings accounts, which is close to the total number of households. This figure was compiled by the World Bank based on annual reports by banks and other sources. The total number o f microfinance accounts, including savings and credit, exceeds 9 million, which again is highby international standards. Semi-formal institutions have played an important role in extending the reach o f financial instruments among the population at large. Chiefly among them are the People's Credit Funds (PCFs). Established in 1997, these are commune-level savings and credit cooperatives whose objective i s to promote self-help and mutual assistance. Other mechanisms to provide credit at the local level include the National Target Program for Poverty Reduction, and an increasingly large number o f microfinance schemes supported by internationalNGOs. Figure 5.1:Rapid Financial Deepening I 2000 2001 2002 2003 2004 2005p - -- I Credit -M2 ` I Source: Own estimates,basedon data from GSO and SBV. Nevertheless, the share of these informal and semi-formal mechanisms as a source o f credit has gradually declined duringthe last 10 years, in favor of the formal sector. For instance, VBARD is now responsible for a majority of the loans made to rural households. This was not always so: data from the 1993 Vietnam Living Standards Survey (VLSS) shows that 40 percent of loans in rural areas were from private individuals, 33 percent from private money lenders, and 25 percent only percent from banks and other formal sources. Another illustration of the extent of formalization is provided by a two-round survey o f farm households conducted by Hanoi Agricultural University and the University o f Sydney in four provinces: Ha Tay and Yen Bai in the North, and Binh Duong and Can Tho in the South. Roughly 400 households were surveyed in 16 communes, from two districts in each province. Almost two thirds of them had a loan, with formal credit institutions accounting for the bulk o f them, On average, about 30 percent of households had borrowed from VBARD, with the share being much higher in Can Tho. Loans from other commercial banks were also common in the southern provinces, but virtually absent inthe north. Households in northern provinces relied on a more diverse range of semi-informal credit sources, including PCFs and local mass organizations such as the Farmers' Union and the Women's Union. As for firms, a survey of nearly 500 women entrepreneurs conducted by MPDF revealed that 80 percent of them had access to external sources o f finance, and more than halffelt that their access to capital was sufficient to support the continued growth o f their businesses. However, private sources of finance, including retained earnings and resources from family and friends, 60 BANKING ANDFINANCE remained important in their case. Other studies, dealing with household businesses and private enterprises more broadly, uncover some systematic patterns inaccess to credit (Box 5.1). Box 5.1: Who Gets Access to Credit? Combining information on who has had access to loan and who has been denied such access provides useful information on how the credit market actually operates. Comparing the number o f successful andunsuccessful borrowers it i s possible to assess the extent o f credit rationing, which is one o fthe key mechanisms through which loan applications considered too risky are turned down, even ina well-functioning market. More importantly, this comparison sheds light on the criteria usedby lenders to decide whether a loan application is worth supporting. InVietnam, several studies have attempted to assessthe extent o f credit rationing, and its determinants. One of those studies concerned Ba Be district in Bac Kan province, and Yen Chau and Mai Son districts in Son La province. These are poor communities inthe mountainous regions o f northern Vietnam. InBa Be district, 57 percent o f all households surveyedhad applied for a formal loan. And only 15 percent, or less than one third o f them, had seen their application turned down. From an international perspective, this relatively low extent o f credit rationing in a poor area is somewhat surprising, suggesting ahighavailability of funding at local levels. In the same vein, a study conducted in Yen Bai province shows that 67 percent o f the households classified as poor had accessed credit in 2000, compared too 41 percent o f the "average" group and 34 percent o f the "above average" group. Another study focused on small an medium enterprises from the private sector in four provinces (Ha Tay, Long An, Quang Nam and Phu Tho) and three cities (HCMC, Hanoi and Hai Phong). It revealed that rejected applicants and discouraged borrowers represented only 14 to 25 percent o f the sample, depending on the definition o f "discouraged". In the absence of credit constraints, the liabilities of the enterprises surveyed would have been8 to 24 percent higher. Access to credit was more likely among larger enterprises. It also appears that older enterprises get less credit, butthis may simply reflect amore limitedreliance on externalfinance, giventhat it is easier for them to mobilize retained earnings. The study also shows that SMEs obtain credit more frequently when the owner is a man. Consistent with the studies described above, the probability o f accessing credit is higher inrural areas. This suggests a deliberate intervention o f local governments to support economic development inthe poorer parts o fthe country. Finally, a database compiled by ADB to study the formalization o f economic activities in Vietnam was used to evaluate access to credit among household businesses and private enterprises. The ADB survey had included 20 household businesses and 20 private enterprises in each o f two rural provinces with relatively large formal sectors (Dong Thap and Ca Mau), two rural provinces with relatively large informal sectors (Thanh Hoa and H a Tay) and the two major urbancenters of Hanoi and HCMC. The results reveal that the possession o f property rights in the form o f LUCs has a large impact on access to credit. Only 12 percent o f f m s without an LUC find it easy or very easy to obtain a loan. The fraction increases to 21 percent for f m s with some legal documents on their land, and climbs further to 32 percent for f m s with a full LUC. "Connectedness", measured by the fact that the owner o f the businesshad ajob inthe state sector, has a large impact as well. Source: Based on Thomas Dufhues and others (2004), Sally P. Marsh and others (2004), John Rand (2005) and EdmundJ. Malesky and Markus Taussig (2005). 61 The characteristics of outstanding loans (as opposed to borrowers) can also be used to shed light on the operation o f the credit market in Vietnam. At the low end, credit from microfinance institutions is based on group-lending schemes, also known asjoint liability groups. In practice, however, individual borrowers are not heldliable for each other. If one member of the group defaults, the only consequence is that the group is deprived from access to credit inthe future. But expelling the defaulter from the group is often enoughto make it eligible again. This option reduces peer pressure considerably. The alternative to microfinance, for borrowers at this low-end of the market, i s to resort to private moneylenders. No collateral i s required in this case. But at 2 percent per month or more, the interest rate represents a substantial charge. By comparison, the interest rate on a formal loan is usually below 1percent per month. Households engaged in farming, forestry and fishing can borrow from VBARD without any collateral ifthe loan does not exceed 10 million dong. However, households are still required to submit their LUCs, which are then retained by the local bank branch as a means o f enhancing their "responsibility" to repay the loan. The importance of this type of arrangement at the local level i s revealed by the case o f the Nghi LOC branch of VBARD, where almost 90 percent of outstanding loans were extended in this way, without involving formal mortgages. In An Giang province, only 15 to 33 percent of the households which held agricultural land had mortgaged their LUCsto obtain loans from VBARD. At the other end, joint stock banks and SOCBs other than VBARD continue to rely heavily on collateral as the basis for their lending decisions. Only exceptionally i s credit appraisal based on an analysis of the applicant's cash flow, or its credit worthiness. However, banks are also aware of the practical difficulties to seize the borrowers' land in the event of a default. Instability in land prices and the inaccuracy of the land boundaries reported in many LUCs also encourage a cautious approach when it comes to valuing the collateral. Some banks have internal rules capping loans at 50 percent of the face value of LUCs. Others only lend against buildings, not against land. Requirements in terms o f collateral value appear to be higher for private firms than for SOEs. Based on the ICs conducted by the World Bank, the average value o f the collateral provided by private firms i s equivalent to 173 percent of the value of the loan, compared to 130 percent in the case of SOEs. There are signs that SOEs are finding it increasingly harder to get access to finance. According to the ICs, a quarter of SOEs were refused a loan the last time they applied, compared to 10 percent of private firms. However, it could simply be that private firms apply for credit less often, either because they lack the collateral or because they do not expect that commercial banks will support them. The bankingsector Considering all sources of credit, including private moneylenders as well as friends and relatives, the formal financial sector i s responsible for close to two thirds of the financial resources mobilized in Vietnam. Within the formal sector, the four major SOCBs retain a dominant position (Table 5.1). In addition, the DAF, which channels policy lending on behalf of the government, has become a major player, second only to VBARD in size. DAF provides subsidized credit to SOEs and private enterprises, supports the development o f infrastructure, and i s in charge of the on-lending o f the Official Development Assistance (ODA). Even without taking DAF into account, SOCBs still account for 70 percent o f total assets in the banking system, and of total banking credit as well. 62 Table 5.1: Relative Size of Financial Institutions Source: Based on data from IMF and SBV. Figures are in trillion dong. DAF credit includes state lending, short-term export loans and ODA on-lending. There are also 36joint stock banks (JSBs), accounting for about 15 percent ot total credit and 5 percent of the total chartered capital inthe banking system. Their customer base is mainly made of small and medium enterprises. In spite o f a period of restructuringaround 2001, several JSBs are weak. But others have become clear front-runners inthe market and are seen as models for bank transformation in Vietnam. These JSBs have also focused on improving operations, corporate governance, and transparency and have developed new products and services for specific market segments. It is estimated that their profits have increased by up to 30 percent recently. Confidence in their performance has been reflected in active trading o f their shares in the informal stock market, at two to four times their partvalue. Three groups of foreign credit institutions are operating inthe market as well. These are: branches of foreign banks, joint-venture banks and fully foreign-owned banks. Together, they account for just 10 percent of banking credit. Their limited development i s mainly due to restrictions on the type of clientele they can serve, and the amount and type of currencies they can hold and trade. The extent of these restrictions varies between branches of foreign banks. But these foreign credit institutions have clear advantages over local ones, including their technical expertise and the huge capital base of their parent banks. Untilrecently, foreign credit institutions were only allowedto set upjoint-venture banks, branches or representative offices. By now they can also contribute capital to existing credit institutions, or to purchase their shares. ANZ Bank was the first foreign-owned institution to gain approval for a 10 percent stake in a domestic joint-stock bank (Sacombank), for 27 million dollars. Shortly after, Standard Chartered paid 22 million dollars for an 8.6 percent stake inACB. Several other applications by foreign credit institutions to purchase shares o f Vietnamese banks are awaiting approval. The banking system of Vietnam is complemented by two policy lending institutions: DAF, which was mentionedearlier, and the Vietnam Bank for SocialPolicies (VBSP). The latter replaces the Vietnam Bank for the Poor (VBP), an organization which had no structures of its own below its head office and operated inpractice through VBARD branches and the facilities of mass organizations at commune and village levels. VBSP was created in 2002 to take over the small-scale programs of VBP and other directed lending programs previously administered by 63 SOCBs and government entities. It is currently in the process of absorbing the 14 million accounts (plus the duties, capital, assets, records and staff) of the VBP, as well as some operations o f VBARD. VBSP has branches in all provinces and 600 transaction offices throughout the country. A planto reform the bankingsector and improve the performance of SOCBs was adopted in2001. The plan included solving the backlog ofNPLs, making progress inreclassifying loans according to the quality o f their servicing, provisioning accordingly, and bringing the capital adequacy ratio up to international levels. Since then, SOCBs have also undertaken important changes to improve information management systems, loan approval procedures, credit risk control and the recovery process of delinquent loans. For instance, most o f them have much stronger information technology by now, and they have all issued credit manuals. Capital injections of roughly 10 trillion dong have boosted the equity of the four SOCBs. However, it i s not at all clear that the restructuring measures have made them truly profitable. The resolution o f old NPLs was accomplished mainly through debt write-offs. And in the meantime, the quality o f new lending did not improve substantially. Assessing the performance o f a loan portfolio is not easy in general; it is much more difficult in Vietnam, where until very recently the accounting standards used by SOCBs were not inline with internationalpractice. An attempt to provide estimates can be found below. But beyond the precise figures, it i s clear that new NPLs were being created more or less at the same speed at which the old ones were being solved (or rather, writtenoff), ifnot faster. The lack of a substantial improvement in the quality o f lending is due to the inability o f the restructuring plan to introduce a strong profit orientation in SOCBs. If management was determinedto maximize returns, it would make sure that credit risk is subject to close scrutiny, at all levels. This in turn would mitigate the potential pressures to lend for projects lacking merit from a business perspective. But this strong profit orientation has been missing at two levels. The most obvious one is at the top ofthe banking system, as the SBV remains heavily involved in the day-to-day management o f SOCBs. The SBV Governor i s responsible for the appointment o f Advisory Boards, top management positions and the Chairperson at each SOCB. This responsibility i s regularly exercised, to the point where more standard Annual Shareholder meetingshave not taken place in any of the SOCBs. At a more fundamental level, the situation simply reflects that SBV is not truly a central bank yet. It can rather be seen as similar to a ministry,both interms of rank and autonomy. Even more important is the lack of a strong profit orientation at the local level, where most lending decisions are made. Currently, the SBV has branches in all provinces. While in principle mandated to exercise banking supervision, the management of these branches i s close to both provincial People's Committees and the management o f the provincial branches o f SOCBs. This cozy relationship may result in credit decisions favoring the broader objectives of local authorities, rather than the profitability of the SOCBs. Decisions of this sort include assisting farmers in difficulty, by freezing their loan repayment obligations, or supportingthe attainment of growth targets, by funding the provincial SOEs in charge o f implementing local infrastructure projects. Such decisions would probably fail to get the approval of profit-oriented shareholders, but they are likelyto get the blessingof local authorities. The quality of bank loans Insufficient separation between the SBV and SOCBs, and the closeness between the provincial branches o f SOCBs and the corresponding local governments, have in practice 64 BANKING ANDFINANCE translated into weak mechanisms to assess the quality of loan applications. Risk and Credit Committees directly reporting to the SOCBs' boards have been created, but they are not vested with powers of credit decision. Equally important, there is no proper separation of commercial loan generation, credit risk assessment and loan approval. The same branch managers are involved all along. And they do have considerable decision-making power, as their lending approval limits vary between 80 and 100billion dong. The lack of a functioning credit scoring system also preventsa proper evaluation o f the creditworthiness o f borrowers. While the SBV does maintain a credit bureau, it only tracks information on large corporate borrowers and SOEs. On the other hand, some technical mechanisms to screen loan applications seem to be operating effectively. The required matching between the maturity of deposits and that o f loans falls in this category. Many investors would like to get medium- or long-term credit, but banks are unable to meet this demand because medium-and long-termdeposits only account for about athird ofthe total. On the surface, the repayment rate o f credits is strong in Vietnam. But this is to a large extent due to loan rescheduling. For instance, VBARD does not count frozen loans as non- performing, or overdue. Admittedly, as the seizure of collateral is difficult in practice, the rescheduling o f NPLs i s at times the only alternative left to banks. But the mere servicing o f interest payments becomes then a poor indicator of their quality. Differences inthe treatment of rescheduled loans are one of the mainreasons for the wide dispersion o f estimates of NPLs in Vietnam. One frequently used approach counts as non- performing any loan whose service is overdue by 90 days or more, as well as any loan which i s overdue by less than 90 days but has been rescheduled. At the risk of simplifying, this i s the approach underlying a series of independent audits conducted since the year 2000 on SOCBs, using international accounting standards (IAS). At the other end, Vietnamese accounting standards (VAS) ignore rescheduling. The main consequence of this difference in their underlyingapproachesis awideninggapbetweenIAS andVAS estimatesofNPLs (Figure 5.2). In practice, however, the differences between these two approaches are not as straightforward as the previous paragraph suggests. Loan classification in an emergingmarket i s an evolving process. Audits based on IAS usually contain important qualifications. This is due to the inability o f the auditors to assess the collateral o f the loans, to the lack of accurate and timely information on credit exposure, to insufficient record of off-balance sheet commitments, and to unavailable documentation on rescheduledand refinanced loans. The expanding share o f loans to small and medium enterprises, which typically entail a higher risk than large firms, i s an additional source of uncertainty. The issuance of Decision 493, in April of 2005, is an important step towards bringing Vietnam's loan classification and provisioning practices closer to international standards. Decision 493 i s indeed in accordance with IAS, although it will take three years to move from quantitative assessments (based on the number o f days the servicing of a loan i s overdue) to more elaborate, qualitative ones. The initial application of this decision yields a share of NPLs which i s much higher than under the old VAS, although not as high as the share estimated on the basis of IAS. A closer analysis of the gap would require an in-depth review of loan documentation, trying to identify where exactly do IAS auditors and the SBV officials in charge o f applying Decision 493 arrive at different conclusions. But there i s no doubt that Decision 493 i s a step in the right direction. Applied consistently, it should exercise considerable pressure on SOCBs to improve the quality of their lending. Such pressure was diluted under the much more lenient VAS approach. 65 ~~ Figure5.2: BadLoansas a Share of Total Credit 20 18 16 14 c c 12 0 10 2 0 2000 2001 2002 2003 2004 2005 1 I-IAS - - VAS A 493, I Source: Based on data from SBV and the World Bank. Figures are for NPLs o f the four main SOCBs, expressed in percentage of their total outstandingcredit. How costly to fix? The large share of NPLs in the portfolio of SOCBs creates a fiscal liability. Even assuming that a more determined restructuring of the banking system can be implemented inthe short to mediumterm, and leads to a rapid improvement inthe quality of new loans, someone will needto absorb the losses from poor lendingdecisions made inthe past. And ina way or another, that someone will be tax payer, through the budget. But how big would the budgetary burdenbe? While the share of NPLs in total credit is an important piece o f information to answer this question, it is clearly not sufficient. Classifying loans as performing or non-performing ignores the diversity in recovery rates, from total losses to partial losses. Moreover, any serious restructuring of the banking sector will require that the capital adequacy ratios of SOCBs be increased up to international levels, especially if they are going to be equitized. This will also entail an additional burden on the budget. Decision 493 provides a reasonable starting point to capture the gradation in recovery rates across loans with different degrees of non-performance. According to this decision, banks need to provision 5 percent o f the value of loans which are not current but are less than 90 days overdue. The provisioning rate increases to 20 percent for loans which are overdue by 90 to 180 days, to 50 percent for loans overdue up to 360 days, and to 100 percent above the one-year threshold. Loans which are being properly serviced but have been rescheduled are treated as if they were 90 to 180 days overdue. Provisioning rates aim to capture the expected loss (in probabilistic terms) for loans in each category. They may not really match the recovery rate that can be secured in Vietnam, which in turn may require that Decision 493 be revised from time to time. But they are based on international experience, and can therefore be used to produce a rough estimate of the lossesthe four main SOCBs can expect. The independent IAS audits conducted on SOCBs provide information on the structure of their loan portfolio, in terms o f delays in the servicing of debts. Unfortunately, the level of 66 BANKING FINANCE AND disaggregation does not always match the breakdown considered in Decision 493. For example, some o f the I A S audits would lump together loans which are overdue by less than 90 days, which loans which are overdue by 90 to 180 days. In those cases, a safe assumption i s to treat all the loans in these broader categories as if they belonged to the segment with the higher risk (90 to 180 days inthe example). Also, the most recent I A S audits of SOCBs are, in a majority o f cases, for the year 2003. The volume of credit has grown considerably since then, although at different paces ineach o f the SOCBs. To capture these trends, the loan structures o f the four SOCBs were weighted by their outstanding credit as of end-2004. The result was an (estimated) overall structure o f the SOCB portfolio, by loan quality, as of end-2004. Applying the provisioning rates established by Decision 493 to this estimated structure leads to an expected loss equivalent to 13.3 percent of outstanding credit. Note that this is less than the share of NPLs in outstanding credit, reflectingthe expectation that not all NPLs will be a total loss. This average provisioning rate was also applied to off-balance sheet items, assuming that such items amount to 15 percent of the figures captured by the IAS audits. And to err on the side o f caution, the total provisioning (including both balance sheet and off-balance sheet items) was increased by 50 percent. The direct fiscal burden from NPLs is represented by the gap between the required provisioning and the actual provisioning by SOCBs. But there is an additional burden represented by the gap between the actual capital levels of SOCBs, and the levels they should meet in order to comply with international standards (a capital adequacy ratio of 8 percent was used as the objective in this case). The sum o f the gaps in requiredprovisioning and in required capitalization amounts to roughly 8 percent o fVietnam's GDP. This is a considerable sum, which cannot be mobilized in just one year or two, even under a determinedeffort to reform the banking system. It is more reasonable to assume that the government would have to issue bonds and transfer the resources to the SOCBs as they restructure. This inturn, would increasethe overall public debt of Vietnam. A debt sustainability assessmentjointly conducted by the IMF and the World Bank tried to estimate by how much. It assumed that the issuance o f government bonds to support SOCB reform would be spread over the five-year period covered by the new Socio-Economic Development Plan (SEDP). The assessment also considered that the economy would continue growing as it did in recent years, at about 7.5 percent per year. Macroeconomic management was supposed to remainprudent, with the budget deficit being gradually brought down, until reaching 1.5 percent of GDP by 2010. Under these assumptions, public debt would increase from 41.2 percent o f GDP at present, to 48.7 percent in 2010 (Figure 5.3). By contrast, if SOCBs had been performing well, and not generated fiscal liabilities, the stock of public debt would remain almost stable, reaching 42.4 percent of GDP in2010. These numbers are to some extent deceiving, because a considerable portion of Vietnam's public debt is in concessional terms. This means that the net present value (NPV) of the long-term obligations faced by the government is considerably less than its face value. On a present-value basis, the burden of SOCB restructuring would increase public debt from 28 percent of GDP at present to 35.2 percent in2010. This figure is high, but it remains manageable. Delaying SOCB reform, on the other hand, could result in a much heavier burden. Even maintaining a stable share o f NPLs in total credit would translate into an increasing fiscal liability, because credit is growing much faster than GDP. The assessment o f fiscal liabilities from NPLs assumes that the provisioning ratios introduced by Decision 493 are a good estimate of the losses associated with loans in different performance categories. Whether this i s so will very much depend on the ability to bring delinquent debtors to foreclosure, and seize their assets. So far, the experience in this respect has been disappointing. 67 The SOCB restructuring effort launched in 2001 led to the creation o f Asset Management Corporations (AMC) at each SOCB. The goal o f these AMCs was to expedite recovery efforts and eventually lead to the liquidation o f bad debtors. But this approach has not been effective in practice. This i s because the AMCs face the same problem as the SOCBs themselves, namely they cannot seize the assets o f bad debtors, especially when those debtors are SOEs and "their" assets actually belongto the state. Figure 5.3: Public Debt over Time 55 I ~ I iI I ? E 35 l I Source: From IMF (2005). Figures based on selected scenarios from a joint Debt Sustainability Assessment by the IMF andthe World Bank. Yet, many bad debtors are probably solvable, and asset recovery might still be feasible, although through other mechanisms. Vietnamese firms generally have a low ratio o f debts to assets, which reflects their reliance on retained earnings as a source o f funding. This suggests that not all delinquent borrowers are actually unable to service their debts. Equally important, the experience o f several provincial branches o f SOCBs revealed successes in seizing assets. A study conducted in three provinces by MONRE and the Swedish International Development Agency (SIDA) showed that pressure on borrowers to sell the land themselves, combined with close cooperation with the local authorities, proved an efficient means o f ensuring repayment. The reluctance o f banks to resort to the judicial system highlights the need to rely on other, non- judicial means to enforce debt obligations and reach foreclosure ifneeded. Capital markets Vietnam's stock market remains relatively small, even by Southeast Asian standards (Table 5.2). It comprises two organized trading centers, in H C M C and Hanoi, both operating as non-business units attachedto the State Securities Commission (SSC). The H C M C trading center started its operations withjust two companies listed on it, in July 2000. It still had 30 companies, 281 bond issues, and one closed-end fund listed on it by late 2005. The total market 68 capitalization stood at 5.9 trillion dong for listed stocks, 33.7 trillion for bond issuances, and 0.3 trillion for the investment fund respectively. Established in March 2005, the Hanoitrading center i s intended to serve SMEs. It i s open for listing to firms which have at least 5 billion dong in chartered capital and 50 shareholders, and made profits in the preceding year. So far, this secondary center has conducted two auctions for initial public offerings and three for government bonds. Table 5.2: A Small Stock Market I Individual 2000 2001 2002 2003 2004 2005e trading accounts 2,870 8,703 13,398 15,569 21,402 24,000 Institutional trading accounts 38 71 122 166 193 250 Stocks listed 5 10 20 26 26 30 Market capitalization n.a. n.a. n.a. 2.3 3.8 5.9 Trading volume 3,641 19,028 35,795 28,024 72,894 82,300 Secondary market activity of the listed stocks has not started yet in Hanoi and has been overall modest in the HCMC. The exception was during the first year after the opening o f the HCMC trading center, when activity was buoyed by foreign investment. As a result, the composite index reached 571 on June 2001. But the boom came soon to an end, leading to a sharp market correction. Currently, the average daily trading i s about 3.9 billion dong in stocks, and 78.4 billion inbonds. O f the 30 companies listed on the HCMC trading center, 29 are former SOEs which enacted public share issuances as part o f the equitization program, before actually being listed. Few companies have used the stock market as a platform for raising funds through an initial public offering. Also, most market intermediaries are closely connected to state-owned financial institutions. The market is currently served by 13 securities firms, one fund management company, three custodial banks and one settlement bank. Of the 13 securities firms, seven are wholly owned by commercial banks and one by the state-owned insurance company. Meanwhile, a much more vibrant informal market for shares has developed. At present, it has an estimated value equivalent to approximately four to five times the organized trading centers. The informal market actually emerged even before the HCMC trading centers was created, as a by-product o f the equitization process. Over time, this process has created as many as 300,000 shareholders, mainly among SOE workers, without a market to make their shares liquid. According to one estimate, there are some 750 equitized companies, 30-or-so joint stock banks, and around 100 JSCs whose shares are beingactively traded inthe informalmarket. Two kinds of intermediaries operate in this informal market. The first one is made o f some 60 to 80 independent, unlicensed brokers. Some o f them deal in securities as a secondary job, but a majority works for financial institutions, including licensed brokerage firms. A typical unlicensed broker has about 10 regularly trading customers, out o f a clientele of about 100 investors. The second group comprises brokers who are licensed to intermediate in stocks in the any of the two official trading centers, but who nevertheless engage in brokering unlisted stocks inthe informalmarket. 69 Box 5.2: The Stock Exchangeversus the Informal Market The official stock exchange and the informal market differ in at least three ways in Vietnam. These are: the range and characteristics o f the companies whose stocks are traded, the extent o f disclosure on the performance o f those companies, and the rules guiding market intermediation. Current efforts by the government to develop the official stock exchange focus on the first aspect. Frequently invoked transparency considerations argue in favor of strengtheningthe second one. The third aspect has been somewhat neglected so far. It is generally agreed that the attractiveness of the 30 companies listed in the HCMC stock exchange i s not particularly high. On the other hand, there are many solid performers among the 6,700 JSCs registered so far in Vietnam. Some o f the JSCs trading in the informal market are among those considered "good" by the Vietnamese public, such as Vinamilk. Bringing more o f those "good" companies into the official stock market would thus help make it more attractive. This is what the government i s trying to accomplish at present. However, there are specific disclosure requirements that a company needs to meet in order to be listed inthe official stock exchange. Those disclosure requirements act as a deterrent for companies preferring to maintain opaque financial accounts, predominantly for tax reasons. A strong incentive for companies to participate in the official stock market will not exist as long as corporate transparency standards at large remain low. It could be argued that even more demanding listing requirements are needed to attract investors to the official stock market. However, unlicensed brokers in the informal market claim that they have not faced with any serious trouble related to poor disclosure. Raising corporate transparency standards at large (as opposed to disclosure requirements for listed companies) may thus more to boost the official stock market. Indeed, listing would then become less burdensome for enterprises, inrelative terms. Last but not least, the informal market is characterized by flexibility in the conduct o f business, whereas rigid rules apply in the official stock exchange. Transactions in the HCMC trading center are executed through periodic call auctions conducted on every weekday in two sessions, between 9:OO and 9:20 AM and between 1O:OO and 10:30 AM, and through over-the-counter transactions between 10:30 and 11:OO AM. The seller i s requiredto deposit the securities through a broker before placing an order, whereas the purchaser is required to deposit at least 70 percent (and sometimes 100 percent) o f the buy order. Inthe absence o f a match between deposits, transactions do not occur. As a result o f these stringent requirements, the official stock exchange is free from trade fails. But this is at the expense o fturnover efficiency andtrading volumes. Source: Based on DFC (2005), Nick J. Freeman (2004) and World Bank (2005). At present, the government is making a series of efforts to boost the activity of the official stock exchange. In August 2005, it instructed equitized SOEs to quickly list their shares at the two trading centers in Hanoi and HCMC. Licenses were also granted to six foreign- invested companies for their conversion into JSCs, on a trial basis. In addition, the government has committed t o facilitating the establishment and development of foreign and domestic investment funds in the country. And since September 2005, foreign organizations and individuals as a group can hold a maximum of 49 percent of a listed company's total outstanding shares, compared to 30 percent previously. While these are encouraging steps, understanding why the informal market has thrived while the official stock exchanges have languished requires a closer look at their differences (Box 5.2). 70 BANKING ANDFINANCE It should be recognized, however, that equity financing will not be a viable source of long-termfundingfor most of Vietnam's SMEs. The minimumcompany size neededto conduct a public share offering, or simply to be listed, i s markedlylargerthan that of the vast majority of private enterprises. For most SMEs, the banking sector, informal funding sources and possibly leasing will be the main ways to have access to finance. As regards larger enterprises, the development ofa well-functioningbondmarket shouldbe a priority. A first step inthis directionis the developmentof an active government securitiesmarket. Such market typically serves multiple purposes. It allows government to finance the budget deficit at a low cost; it providesthe investment communitywith safe and liquidmedium-to long- term savings instruments; and it is the foundation for the term structure of a country's interest rates. The controversialtransformationof SSC from an independent body into an agency under MOF might reflect the importance attached to the development of a government securities market. At roughly 3 percent of GDP, such market is already biggerthanthe equitymarket. For instance, in 2004 bond trading represented almost 90 percent of all transactions at the HCMC trading center. And some licensed securities firms actually make the majority of their profits on bondintermediation. By now, there are more than 100 issuancesper year oftreasury bills. DAF, the HCMC Investment Fund for Urban Development (HIFU) and Vietnam's Oil and Gas Corporation(PetroVietnam) are amongthe few other issuers of bonds inVietnam. The corporate bondmarket is still nearly non-existent. 71 6. THE LAND MARKET A land market has already emerged in Vietnam, despite the still partial development of formal titling. Many transactions take place inthe absence of proper legal documentation, simply because local officials know which piece of land or property "belongs" to whom, and arbitrate in case of conflict. From this perspective, the absence of formal property rights has not been incompatible with market development. But land titling is bound to increase efficiency. LUCs can be used as collateral to obtain credit, and insufficient access to finance i s arguably the main constraint to business development inVietnam. Property titles should also provide a more sound foundation for the development o f a real estate market. From a social perspective, they could help formalize the situation o f urban migrants, who continue to be registered as "temporary residents" despite having lived in the cities for years, if not decades. However, the dynamics of the property market inrecent years also show that titling alone is not enoughto ensure efficiency. Price "bubbles" in real estate need to be kept under control by making speculative transactions more burdensome, as the Vietnamese government did recently. And some o f the most difficult issues, such as facilitating the consolidation of agricultural land, restructuring state forest land, supporting community-based land management among ethnic minorities, and recovering idle land from urban SOEs, will still require specific policies, going beyond the issuance o f LUCs. Looking forward, one of the most difficult challenges ahead is the conversion o f agricultural land into residential and industrial land, which will be neededon a massive scale as Vietnam develops and urbanizes. Success will critically depend on fighting corruption in land conversion and properly compensating the affected populations. Land titling While only the state can own land in Vietnam, a private person or entity can also own "use rights" to a parcel o f land for a specific period of time. The allocation o f those use rights, launched with the Land Law of 1987 and accelerated after its 1993 revision, has been at the center of economic reform. The fraction of households using land in Vietnam is very large, because the economy is still to a large extent agrarian. According to the VHLSS 2004, the fraction could be as highas 86 percent. The allocation of use rights in urban areas, while affecting a smaller share of the population, presents additional difficulties. Allegedly temporary forms o f land occupation are common; inother instances, several households share the same dwelling. Because o f its scale and complexity, the process o f allocating land use rights in Vietnam can be seen as one o f the most ambitious privatizationprograms ever attempted The 2004 revision of the Land Law is an important milestone towards completing the land titling process. Under this law, LUCs can be leased, sub-leased, transferred or given in inheritance. They can also be used as a capital contributionto a business project, and mortgaged as a security for loans. It should be noted, however, that land titles are allocated for periods ranging from 20 to 50 years, and this raises questions as to how the values o f the LUCs will be affectedas their expiration date gets closer. For now, the oldest LUCs are only a dozen years old. The Land Law of 2004 suggests that the terms will be renewed"if the land users are still in need of such land and have strictly complied with land legislation during their period o f occupancy, and the use of such land is in accordance with the approved land use planning". But the specifics are unclear at this stage. Much progress has been accomplished in the issuance of LUCs for agricultural land, popularly known as "red books". This i s confirmed by both administrative data compiled by MONRE and household survey data (Table 6.1). Direct allocation by local authorities has been the most important allocation mechanism. In 2004, 48 percent of the agricultural land held by households had been obtained from communes, 16 percent has been purchased and 8 percent had been reclaimed. The remaining had been inherited; the mechanism through which the deceased had originally acquired the land is unfortunately unclear. There are important regional differences, however. Direct allocation by the communes was more important in the North, where it affected two thirdsof all agricultural land. Communes only played a marginal role inthe South, where inheritance and the market were much more important. Table 6.1:Progressinthe Titling o f Agricultural Land Source: MONRE and own calculations based on 2004 VHLSS by GSO. Figures refer to annual and perennial agricultural landonly. Field research conducted by ADB in six provinces across Vietnam shows considerable improvements in the land allocation process. Procedures are now relatively quick and simple, with most land transfers registered within one or two months. Since 2000, official transfer taxes and fees were reduced to a level that i s low by international standards. Because procedures are less onerous procedures, households increasingly seek to register land transactions formally. However, the extent to which land and urban property have been given proper titling varies considerably from province to province (Figure 6.1). Inthe South, many o f them were able to build on a reasonably well-developed system o f informal-customary law that established ownership at the village level. Cadastral maps also existed. As a result, disputes over ownership and boundaries were minimal. In the North, a large fraction of agricultural land i s covered by 73 new titles, whereas in the south LUCs have been used mainly for land that was not titled under the old system. In both cases, there i s a clear tendency by households to register previous, unofficialtransfers under the terms of the new LandLaw. Figure 6.1: Land Registrationacross Provinces Average Proportion of Average Proportionof Rural Households Urban Households That Have a LUC That Havea LUC for all of their plots of land for at least one plot of 1 (excluding residential land) Residential Land 55 65 - 165-75 175-80 Source: Own calculations based on 2004 VHLSS by GSO. Progress inthe issuance of land titles for urban property has beenmuch slower. The year 1994 saw the introduction of Building Ownership and Land-Use Right Certificates (BOLUCs), popularly know as "pink certificates". A BOLUC combines home ownership with the land-use right into a single legal document. Its possession amounts to having a fully-titled house. But obtaining a BOLUC is complicated. Depending on the occupancy date, specific legal documents have been identified as beingacceptable in claiming ownership. But even property owners with relatively long period of tenancy may not have those particular papers, and the regulations do not specify what to do in cases. Moreover, BOLUCs are not given to urban migrants without permanent household registration, nor when there are unresolved disputes over ownership. In Hanoi, disputes are aggravated by the sharing o f old, colonial-period dwellings by several households over many decades. In HCMC, problems were exacerbated by the incorporation of large portions of formally agricultural land on the peripheryinto five new urban districts, in 1997. As a result of all this, there is a considerable bureaucratic limbo hanging over BOLUCs. In 2005, a new decree introduced separate certificates, to be issued under the authority of the Ministryof Construction (MOC), to state owners' rights over buildings. Inparallel, a new Real Estate Registration Office is to be established in each province, under the management either of the Department of Justice, the Provincial People's Committee or the provincial Department of Natural Resources and the Environment (DONRE). This office would register all land and assets on land, as well as all real estate transactions. User or ownership rights would be registered on 74 the basis o f existing LUCs, building licenses or other documents. These initiatives raise the prospect o f potential inconsistencies, slowing down the completion of land titling inurban areas. The new Land Law also expanded the mechanisms by which enterprises can obtain land. In addition to the direct allocation or lease by People's Committees, and the transfer of LUCs among entities, it i s now possible to auction LUCs. Because this third mechanism is new, very few enterprises have benefited from it so far. To date the most common way of acquiring land is through direct transfer from a previous land user, at an agreed price. In the ICs sample, 56 percent o f manufacturing firms and 65 percent o f SMEs had used this mechanism for the most recent land acquisition. Direct allocation by People's Committee is more important for large- scale firms, often needing large plots of land. It i s uncommon (and probably difficult to process) inthe case o f small firms. The process to acquire land directly from People's Committees has been considerably streamlined. In the past, applicants had to submit a cadastral map, a feasibility study for the investmentproject, a plan for site clearance andthe compensation ofthe affectedpopulations, and the written opinions o f all the relevant agencies. By now, the documentary requirements have been streamlined to only three: an application form for land allocation or lease, an investment decision or certified copy of an investment license, and a certification letter from DONREon the effective record of land use by the applicant. A 30-day time limit exists to process applications on land that i s already cleared; the limit i s 45 days for sites which are not yet cleared. This compares with an actual average of more than eight months at present. The revised Land Law also requires DONRE to be the sole contact for applicants during the whole process, so that applicants do not need to be runningfrom one government agency to another anymore. Land prices There are at least three different "prices" for land in Vietnam. One o f them refers to the initial transaction, on allocated land, between the state and the recipient of the LUC. The life of this administrative price extends in practice beyond the initial transaction, as levies and fees on subsequent LUC transfers are based on it. This is also the price most often considered by banks, when LUCs are used as mortgage to obtain a loan. A second relevant price is the one used for transactions among households or enterprises, when transferring LUCs. And a third one i s the rental price o f land or property, again between households or enterprises. The first two prices concern the entire lifespan of the LUC; the third one covers a shorter period of time. Gaps between these three prices shed light on the operation o f the land market in Vietnam. Diverging trends between the first and the second one are bound to create distortions, and to provide incentives for corruption in land administration. The new Land Law represents an important attempt to bringthese two prices closer together, as it favors the use o f market-oriented mechanisms to set the price of LUCs, including auctioning, competitive bidding and appraisal by external evaluators. Diverging trends between the second and the third prices can be used to assess whether property markets are experiencing a "bubble". A booming market for LUC transfers, ifnot accompanied by a booming rental market, i s not sustainable inthe longrun. Estimating the actual level o f land prices, in transfer or rental transactions between households and enterprises, i s not an easy task. The role of land rental markets in agriculture i s still thin. In 2004, only 3.6 percent o f all agriculture land was rented, with 10.7 percent o f all rural households having rented-inland, and 6.0 percent rented-out land. Moreover, some of these transactions are not really at "arms length", as they often involved relatives and entail no payment. The development of land sales is limited too. In 2004, only 8.3 percent of all 75 households reported that they had bought land, but this is probably an over-estimate, including the acquisition o f LUCs from communes. At 2.7 percent, the fraction o f households reporting that they had sold land was actually much lower. The number o f observations available on actual "market" transactions i s thus small. Moreover, an estimation o f market prices for property in general would hide important disparities between provinces, depending on the supply o f land with proper titling. Danang is a case in point. In 2002 the provincial authorities launched a slew o f projects to fully exploit the availability of large amounts o f land around the city, creating thousands o f lots for sale. This resulted intemporary excess supply, and a downward price spiral. The trend was at odds with the steady increase in property prices observed at the aggregate level (Figure 6.2). Only now, when prices elsewhere are falling, i s Danang coming out o f its prolonged property market slump, as large investors show interest in developing tourism and residential projects. Figure 6.2: Property Prices for Urban and Rural Land - 1200 I l cb EF!34 1000 5 800 ` urban b i 1 (left axis) Q 600 31 CI) C rural 1 U0 -0 400 2 , (right axis) S m v) E0 3 200 1 0 0 1993-1996 1997-2000 2001-2004 1 Source: Own calculations based on 2004 VHLSS by GSO. Urban land prices are from 106 observations inninemajor cities. Rural landprices from 1,3 19 observations inrural areas. The trend is almost opposite inthe case o f rental prices (Figure 6.3). Information i s even more fragmentary in this case. Some, however, has been compiled on a systematic basis by real estate agencies catering to the upper end of the market. These agencies deal on a narrow set o f well-defined properties, such as offices meeting international standards o f service, and housing for expatriates. The trends are thus unlikely to be reflecting changes in quality over time. The dramatic decline in rental prices in the second half o f the 1990s reflects the steady increase in supply, as new buildings and residences came off the ground. The relative stability o f rental prices since the year 2000 or so i s clearly at odds with the boom in sale prices. However, there is clear evidence that the real estate bubble i s over by now. A major reduction in transaction started at the beginning o f 2005. Over the first six months o f the year, transactions registered at DONRE's property center in H C M C fell by 68 percent compared with the previous year. According to press reports, by end 2005 Hanoi still topped the country with a 76 median sales price of around 150,000 dollars for street-front homes. But this was 10 percent less than the previous year, and many listings were staying on the market for long periods o f time. Domestic demand for construction materials declined as a result, with the number of civil construction projects inHCMC down by more than 25 percent compared to 2004. Figure 6.3: RentalPrices for Upscale Housing and Offices 1 60 I I I I 25 I ! I 10 1 1996 1997 1998 1999 2000 2001 2002 2003 2004 I HCMCoffice - - - 'HCMChouse -Hanoi office- - - - - - .Hanoi house 1 I Source: Based on data from ChestertonPetty Vietnam. Figures correspond to international standard office buildings and residential units meetingexpatriate standards respectively. Two policy measures were instrumental in stopping the real estate bubble. First was the taxation of land transactions, at 5 percent of the value o f the LUC, introduced in early 2005 as a result of the new Land Law. The tax rate i s not high by international standards, and should not have a deterrent effect on non-speculative land transactions. But the re-evaluation of the administrative price of LUCs, on which the tax applies, implies that the tax amount i s not negligible anymore. Buying and selling again over a short period o f time, which was common duringthe bubble periodbecomesthen expensive. The other important measure was the ban imposed on real estate developers from selling small plots of land that do not have houses built on then. Now, land allocated to developers for the construction of residential property projects cannot be divided up into individual plots and sold prior to construction o f the infrastructure and housing outlined in the project plan. This measure blocked a loophole which allowed developers to reap significant profits by converting low-cost agricultural land into residential land without making any substantial investment. RuralLand The attempt to distribute land to rural households in an egalitarian way, so that each o f them would have a similar endowment in terms of both quantity and quality, has resulted in a considerable fragmentation o f land holdings. This objective, clearly stated in the Land Law on 1988, was attained through the allocation of sizeable numbers of non-contiguous to each 77 household. The problem i s particularly severe inthe North of Vietnam, where the average rural household has 6.5 plots of land. In some places, such as the Tu Son district of Bac Ninh province, the mean area per plot i s barely one sao, or 360 square meters. But it exists also inthe south, where the average is close to 3.4 plots per household. Fragmentation does not appear to be associated with a lower level o f mechanization, but it still affects agricultural production in other way. Households with more fragmented land endowments are more likely to cultivate rice, to spend more time on the farm, and to achieve a lower productivity. A crude estimate suggests that a doubling in the number o f land plots per household i s associated, other things equal, with a 15 percent lower productivity. The issuance o f transferable LUCs, under the revisedLand Law on 1993, was expected to solve the fragmentation problem, with market forces inducingspontaneous land consolidation. But this did not happen in practice. Market-based land consolidation requires multilateral and more or less simultaneous deals among several potential buyers and sellers. The presence of irrigation and transportation networks, whose shape depends on plot boundaries, can make those multilateral deals remarkably complex. Administrative constraints on crop choice represent another major obstacle. In the absence of crop choice, some of the possible economies of scale from land consolidation cannot materialize, which reduces the incentive for farmers to swap land with others. Credit constraints can also prevent or slow down market-based consolidation. As land fragmentation did not decline ina spontaneous manner, in 2003 the government adopted an explicit land consolidation policy. Localities where then encouraged to facilitate the transfer of LUCs and the conversion o f land-use purposes so that parcel sizes would be suitable for efficient production. Implementationi sjust beginning, as many local authorities are not even aware of the new policy. But some o f the ones who are have been testing various models for the consolidation on land plots, including voluntary exchanges coordinated by local authorities, compulsory exchanges, and market-based models inwhich households are encouraged to buy and sell plots. So far, land consolidation projects appear to have been slightly redistributive innature, with newly-marriedhouseholds beingthe main beneficiaries. Some provinces actually started land consolidation campaigns even before a national policy had been adopted. In Nghe An province, following a local Party directive o f 2001, such campaigns were conducted in333 out of 363 communes. Consolidation was conducted by village authorities, either through adjustment of adjacent plots or through complete re-organization and reallocation through the drawing o f lots. Participation inthe process was compulsory for farmers with annual crop land. The process was deemedto be successful, as it allowed shifting from rice to higher value peanut cultivation. For instance, inthe district o f Nghi LOC, the total number of plots was reduced from 382,948 to 165,139. However, LUCs issued prior to the land consolidation process have only been amended or reissued in 6 o f the 333 participating communes. Another area in which additional government activism i s needed i s the reform of State Forest Enterprises (SFEs), which still control about 40 percent o f all forest land in Vietnam. Much of this land is located in the poorest parts o f the country and managed inefficiently. Over the last decade, a series of steps have been adopted towards the restructuring of SFEs. The main objective underlying those steps i s to separate commercial forest activities, such as wood production, from public service functions, such as maintaining bio-diversity, and protecting watersheds. Inthe process, SFEs which carry out mainly public service functions will be turned into ProtectionForest Management Boards (PFMBs), whereas the land they use inefficiently will be transferred to local communities. A key decree was adopted in2004 to accelerate this process. As of May 2005, there still were 362 SFEs still in existence. Of these, 114 were to be converted into PFMBs, 27 into public utility enterprises, six were to be liquidated, and the rest 78 THELANDMARKET was to be converted into commercial enterprises. But the unclear boundaries between "protection", "special use" and "production" forests are delaying decisions on what land should be retained by PFMBs and SFEs, and what land should be reserved for allocation to households. Following a desire to maximize funding from government, provinces and SFEs have classified many areas as "protection" and "special use" forests, whereas in reality "production" would be more appropriate. A thorough re-classification might be necessary before final land allocation decisions can be made. Several other issues will needto be addressed, including the funding and staffing o f PFMBs, and the revision of the incentives faced by SFEs in provinces where logging bans are inforce. Beyond the specifics, it i s expected that 500,000 to one million hectares o f forest land could be released to communities through this process. This should benefit the almost one million households who use forest land to support their livelihoods, most o f whom are poor. However, an effective transfer may require the use of customary land titling, a possibility first formalized in Vietnam by the revised Land Law of 2003, but piloted before that in several provinces, including Dak Lak and Son La. Those pilots amounted to the recognition, by local authorities, o f the long tradition of sustainable land management by ethnic minorities, especially in relation to community-held forest land and so-called "unused land" (Box 6.1). In Dak Lak, between 1999 and 2002 around 16,000 hectares of forest land were allocated to ethnic minorities, of which one third went to villages and one third to groups of households. In Son La, between 2001 and 2003, some 685,000 of forest land were allocated to local people, of which half went directly to village communities. Currently, communities throughout the country are de facto managing some 2.5 millionhectares of forest land. UrbanLand Even if all land titles for residential property could be issued, urban areas would still be confronted with the need to re-allocate their commercial and industrial land. At present, such land is mainly in the hands of SOEs, whereas non-state enterprises face considerable difficulties ingetting accessto land. The ICs sample gives a sense of the magnitude of the problem. Among enterprises with 250 employees or more, SOEs have on average five times more land than private firms. Across all institutional types and sizes, only 54 percent o f respondents to the ICs declare that they have the right to sell or mortgage the piece of land they acquired more recently. And only 10 percent of them are located inwhat they considered to be their own land. To some extent, limited access to industrial land simply reflects the overwhelming population density and the ensuing shortage of idle space within urban boundaries. Most o f the existingindustrial land is already occupied by SOEs, whereas the conversion of agricultural land into industrial land at the fringes of the cities is a slow and often controversial process. Thus, the supply of available land coming on to the market is too small to meet the demand. But in addition, SOEs often get priority in the allocation o f whichever land i s available. The case o f Hanoi illustrates the point. In the period from 1994 to 2002, only 428 new rental contracts covering about 3 million square meters were issued. More than half of these contracts were to SOEs. By now, an overwhelming 95 percent of land under lease to organizations in Hanoi i s in the hands of SOEs, leavingless than 5 percent to private firms, cooperatives and other production groups. 79 Box 6.1: Land Relations in Black Thai Villages The Black Thai communities o fChieng Dong have resisted landallocation vigorously. Not only did they ignore and circumventedthe new regulations on the issuance of LUCs: they also continued to modify plot sizes and exchange plots among one another, ignoring their registration and the associated legal duty to report changes in landholdings. Whole villages disregarded land classifications commanded by the state, preserving overlapping and seasonally changing uses o f land for cultivation, animal husbandry and forestry. Black Thai resistance to land allocation even included open protest against the registration o f some fields, forcing local authorities to exclude them from LUCs. Traditionally, a local lord (or phia) was considered the formal owner o f Black Thai land; yet his actual powers were mainly symbolic. At the next tier, village communities exerted collective control rights over the land, but could not intervene in its day-to-day management. At the bottom o f the hierarchy, households heldrightsto cultivate wet rice fields and upland fields, to graze water buffaloes and to collect forest products. In exchange for these rights households owed shares o f labor and products to the village collectives and thephia. The balance of individual entitlements and collective control varied between different land uses. Village collectives exercised strong control over the highly pricedwet rice fields. They periodically adjusted allocations in response to changes in household and village demography. On the other hand, collective control did not extend into the uplands, where households were free to clear suitable forest for new fields provided that no other household had worked the land previously. Similarly, all households enjoyed the rights to graze their water buffaloes and collect forest products inthe uplands, with no village boundaries applying. The Black Thai of Chieng Dong also resisted the complete collectivization o f agriculture during the periodunder central planning. Collective production remained limitedto wet rice and some upland rice fields. Households continued to grow rice and other supplementary crops, and to raise pigs and poultry, on their own account. The Black Thai also opposed the formation o f larger cooperatives, including more than one village. Inthe end, collectivization reinforced some ofthe key features oflandrelations among the Black Thai. The district authorities and cooperative leadership took over the estates o f administration previously held by the phia and village collectives. Access to wet rice land remained contingent upon the fulfillment of obligations towards the village, now cooperative. The leadership o f cooperatives allocated labor duties, hence shares in total product, to all adult villagers. Access to land thus became contingent upon the number o f main laborers in a household, unifying the allocation formula across villages. During the late 1970s and 1980s the balance of power gradually shifted away from the cooperatives and the district administration. Villagers expanded individual upland fields up the slopes and increased their engagement in own-account activities. The village collectives did not stop this expansion o f individual cultivation. By the early 1990s land relations in Chieng Dong resembled those that had existed before collectivization. This i s when land allocation took place. District authorities then asked villagers to declare their wet rice fields. Butthe requestwas met with protest, because long- term allocation would conflict with the to adjust plot sizes to changes in household need. In response, the district leadership instructedthe cadastral office to leave the fields out o f the land certificates. Inthe years following land allocation, all villages in Chieng Dong continued to reallocate their wet rice fields. When asked about their motivations to continuously reallocate land, village leaders unanimously referred to a concern for equity. Source: Based on Thomas Sikor (2004). 80 Officially, SOEs are not allowed to sub-lease their land to other enterprises, but they do so in practice. Inthe small-scale enterprise survey conducted by the World Bank, 11percent of the respondents had rented land from SOEs. Leaseshad most typically a short- or medium-term horizon, with 10 years beingthe maximum contract length. In addition, as the rent i s unofficial, the private enterprise is inavulnerable position, should a disputearise. Unofficial sub-lease arrangements of this sort may appear to be a convenient, market- based solution to address the shortage of land, but they are far from ideal. The combination o f relatively short lease terms with the overall insecurity associated with unofficial arrangements could be a deterrent for investment on the land by the lessees. Profitability i s further eroded by the fact that the rent a private firm pays to an SOEs cannot be considered a tax-deductible expense. Conversely, revenue generated from sub-leasing the land i s not recorded in the annual accounts o f SOEs. This i s a potentially important source of corruption. The new Land Law mandates SOEs to return to the state any landthat is not requiredfor the conduct o f the business operations. Inpractice, few have done so. The process ofrecovering idle SOE land is in fact quite laborious. The lack o f a precise definition of what constitutes idle land, or an effective mechanism for land surrender, has led different provincial governments to try different approaches. Danang, for instance, has relied on the authority of the Party to recover considerable amounts o f SOE land for its urban development projects. Hanoi i s by now auctioning state land, and houses in the inner city, in a move which could generate up to 100 million dollars inrevenue for the city in 2005. But the case of HCMC highlightsthe magnitude of the challenges. The Prime Minister had issued a decision to solve and rearrange state-owned land and houses in HCMC in 2001, creating a legal ground for their proper and use by enterprises and individuals inaccordance with city approved plans. Since then, the HCMC Department of Finance (DOF) has been able to inspect about 2,000 sites, out of the 6,812 plots of state-owned land being used by enterprises or individuals. These inspections show that most often land has not been used for the intended purpose, and has instead been leased to a third party. The rent charged to the third party is substantially higher than the fees the official users have to pay for the state-owned land. Enterprises and individuals have often built houses or hotels on the land without permission, while others have simply failed to pay their fees to the state. Zoning and conversion The state allocates land-use rights, but it also enjoys the powers to recover them in prescribed circumstances. With Vietnam's rapid economic development and urbanization, the recovery of agricultural land for residential, commercial, industrial and infrastructure projects has become increasingly common. In many cases, the recovery i s aimed at building industrial zones. By the end of 2004, there were 192 major industrial zones and clusters in 13 provinces, with a total surface close to 30 thousand hectares o f land. Originally, some 84 percent o f this land was agricultural. It was recovered from over 100 thousand households. The most common recovery procedure was the compulsory recall of farmers' LUCs by the state. The most common allocation mechanism was the discretionary leasing o f recovered land to enterprises. However, the process of land conversion has proved especially controversial. In some cases, land conversion has been met with resistance by the affected households, leading to social unrest and considerable delays in land clearance. Land conversion has also been marred with corruption cases. Given the gap between the administrative price at which the state allocates the 81 converted land and the price at which it can be sold, capital gains can be sizeable. And this in turncreatespervert incentives for the government officials involved inthe conversionprocess. A recent assessment ofthe implementationofthe new LandLaw by MONRE included a review o f 17,480 letters of denunciation. Out o f this total, more than 70 percent were related to site clearance and compensation. The largest numbers o f denunciation letters were from HCMC, Hanoi, Phu Tho, DienBien, Bac Kan, Ha Tay, KhanhHoa and Quang Ninh. Lack o f proper legal documentation on the reclaimed land was identified as the main cause for conflict and discontent. But this is not the only problem. In spite of the introduction of market-based mechanisms to set the administrative price of land, the latter are still consider insufficient. Many complaints arise from the fact that neighboring, similar pieces of land may be valued differently if they lie within different administrativejurisdictions. Even when administrative prices are close to the market price of equivalent agricultural land, they are substantially lower than the price it can fetch once it i s converted into commercial, industrial or residential land. The users of the agricultural land being reclaimed do not benefit from the resultingcapital gain and often complain about it. Problems are aggravated by corruption and poor communication. Several studies put land and housing agencies at the top of the corruption ranking inVietnam, or close to it. The list includes the ICs of the World Bank, the pilot corruption diagnostic study conducted by the Internal Affairs Committee of the Party in seven cities, and a report cards study done infour cities by the World Bank with support from the Ministry of Home Affairs (MOHA) yield all a consistent message in this respect. Payment o f informal fees i s common, although there are important regional differences. For instance, based on the ICs, less than 20 percent o f firms make additional payments to receive a LUC in the South Central Coast, compared to nearly 60 percent inthe RedRiver Delta. As for communication, the report cards study mentionedabove uncovered a highdemand for information on land-use plans and zoning decisions. But less than one fifth o f respondents were aware of them. An additional complication comes from the different timeframes for zone planning cycles and land leases. Under the new Land Law, long-term land use plans are to be drafted for a period of ten years, broken down into more detailed five-year plans. But the timeframe for land leases is from 20 to 50 years. At present, continuation rights are not well spelledout. Neither are the redemption rightsfor enterprises whose LUC is withdrawn as a result of land re-classification. Disagreements over the compensation o f affected populations have resulted in considerable delays in the conversion of agricultural land. A publicized case concerns H a Tay province. In 2002, Kim Bai company submitted a request to lease 6,442 square meters o f pond land behind the company's head office, in Thanh Oai. The provincial People's Committee approved the application in 2004, after a two-year consultation and bargaining process with the affected households. The company agreed to pay 11,000 dong per square meter in compensation, and to contribute 3.3 million dong per suo in support o f infrastructure building by the People's Committee. However, shortly after the local residents preventedthe company from filling inthe ponds. They were afraid that the new factory would pollute their stream. The dispute continued despite several new rounds of negotiations and assurances by the company that its factory would have an international-standard waste water filter system. Since 2004, provincial People's Committees are requiredto establish resettlement zones to provide housing for households whose land has been recovered bythe state. Inpractice, few o f these zones have been created, and complaints abound on the lack of provision of proper housing for households whose land has been reclaimed. And since 2005, foreign investors are no longer requiredto pay for the reimbursement of existing tenants of the land they are allocated. Under 82 THE LAND MARKET the revised policy, investors pay the lease and the provincial government takes charge of all resettlement and ground clearing expenses. If the investor incurs such expenses, it can deduct them from the cost o f the lease. 83 7. THE LABOR MARKET Vietnam is well-known for its disciplined, hard-working, fast-learning population. This is the country's main asset in its drive for economic development. Low wage costs also make it one of the main sources of its international competitiveness. Not surprisingly, the surge in business activity associated with economic reforms has led to a huge increase in the demand for labor. This has translated into a double mobility: occupational, from agriculturaljobs to non-farm employment; and geographical, from rural to urban areas. But wage employment i s unevenly distributedacross the country. And the same is true o f labor earnings, which vary considerably from province to province. Disparities in earnings have been reduced with economic growth, as also has the earnings gap betweenmen and women. Butthe labor market reward to education has increased, opening the prospect of a different kind of inequality as Vietnam develops. Improvements in labor earnings have not been associated with labor conflict. Strikes have been uncommon, and have basically bypassed the official industrial relations system. Overall, Vietnam i s characterized by rather decent working conditions, and corporate social responsibility (CSR) principles are gradually beingacceptedinkey export sectors. Unemployment has not been a major issue either, with most o f the jobless being young, relatively educated entrants to the labor market. Over-staffing is still common in the state sector. It i s rather the excessive turnover of qualified workers that creates a problem for business. But the main weakness of Vietnam's labor market i s the social protection system built on it. Traditionally conceived for public sector workers only, this system still needs to complete the transition to a market economy, so as provide affordable insurance against major risks to workers who are increasingly mobile, occupationally and geographically. Employmentand migration Labor force participation rates are unusually high in Vietnam, both for men and for women, whereas unemployment rates are unusually low. From the onset, this establishes a clear difference with other countries, where the most pressinglabor market issue i s often to understand why so many people are inactive or out of ajob. InVietnam, a large majority o fthe working-age population i s active, and most often employed (Table 7.1). Understanding the characteristics of thejobs available as the economy develops, or the determinants o f the earnings they provide, are thusthe most important labor market issue. Rapid economic growth has resulted in two, related forms of job mobility. From an occupational perspective, an increasingly large share o f the population works out of agriculture. Farmingjobs absorb less than 40 percent of the working-age population nowadays, compared to roughly half inthe 1990s. This decline has been compensated by an increase inthe share of wage jobs, relatively minor in the state sector, but considerable in the private sector. By now, household businesses, registered domestic private enterprises and foreign invested companies provide wage employment to more than 18 percent o f the working age population. This is slightly higher than the share o f self-employment, and more than twice the share of the state sector (government and SOEs combined). Ifanything, these figures may under-estimate the role of the private sector, as they are based on household survey data. Individuals working for LABOR MARKET equitized enterprises and for joint ventures involving the state sector may declare that they work for SOEs, thus inflating the share of the latter inthe working-age population. Table 7.1:The Structure of the Labor Force 1993 1998 2002 2004 Inactive 19.42 15.32 16.69 17.17 Active 80.58 84.68 83.31 82.83 Employed Government 3.08 3.55 4.44 5.25 SOEs 2.50 2.57 3.30 3.14 Private enterprises 10.78 0.14 15.71 16.99 FDIcompanies 0.10 1.12 0.80 1.33 Non-farmself-employment 14.67 6.52 19.05 16.52 Farmers 49.46 50.15 38.2 38.77 Unemployed 0.63 1.so 0.83 Total unemployment rate 0.74 2.16 1.oo Urbanunemployment rate 1.44 3.31 1.96 Urbanunemployment rate (MOLISA) 6.85 6.01 5.60 Source: Own calculations basedon data from GSO and MOLISA. All figures are in percent o f the populationaged 15 to 64, except for unemployment rates, which are in percent of the active population. The last row is based on MOLISA's labor force survey. All other figures are from VLSS and VHLSS. Householdbusinessesare treated as private enterprises for their hiring o f wage workers, and as a source of non-fann self- employment for thejobs they provideto householdmembers on an unpaidbasis. The second form of mobility characterizing the Vietnamese labor market in recent years i s geographical. Some people engage in non-farm employment without leaving the rural areas they live in. But for a large share of the population, moving from farming to wage employment also entails moving from rural to urban areas. Migrants now make up a large proportion o f urban residents. Migration into large cities, most notably Hanoi and HCMC, has been happening at a faster pace. The extent of this phenomenon is probably under-estimated, as data from population censuses gets quickly outdated. The recent, mid-term population census of HCMC uncovered 420,000 more people living in the city than had been predicted by GSO. It also appeared that roughly 30 percent of HCMC's population i s made of migrants who have non-permanent registrations. InVietnam, the right of individuals to freely choose their place of residence and work is enshrined in the Constitution. As the migration flow puts additional pressures on already overloaded infrastructures and public services in major cities, local authorities have established implicit barriers to labor movements leading to additional costs borne by migrants, apart from natural relocation costs. The registration system, whereby migrants who do not have a place o f residence do not get access to some basic services, is the main administrative barrier to 85 geographical mobility. But it has not deterred a massive move to the provinces, districts and communes where thejobs are. A recent study of migration issues in Thanh HoaandNgheAn provinces, by ADB, GSO and VASS, shows that 42 percent of the households count a migrant among their members. In more than 80 percent of the cases, these migrant members work in a different province. And in most cases, they contribute to higher living standards at home. About 36 percent o f the respondents inthis study declare that migration has ledto a substantial increase infamily income, and an additional 61 percent report a marginal increase. Less than 4 percent consider that their situation i s worse due to the migration of a household member. Migration is partly driven by the huge geographical dispersion of wage jobs (Figure 7.1; left panel). Almost two thirds o f those at work in HCMC do it for a wage or a salary, and the share i s around 60 percent in Danang and Hanoi. This i s where the opportunities for work are. At the other end, the share ofwage employment inthe total is below 20 percent inDienBien, Son La, Lang Son and Lao Cai. These figures should be interpreted with some caution, as the VHLSS i s not precise enough in the case o f small provinces, especially if the number o f wage workers i s low. Weak precision makes comparisons between two consecutive surveys hazardous. But the gaps inthe shares of wage employment across provinces, inany particular point intime, are large enough to be significant. It clearly appears that wage employment is more prevalent in some areas, and especially inthe South East and the north east corridor to China. Figure7.1:Where Are the Jobs? And how much D o They Pay? Average LabourEarnings ~ Share Of The Working pee Populati PerSalaried Worker, 2004 HoldingASalariedJob, 2004 C ~ V f J o ) Lessthan20 1Lessthan4500 at030 4500to 5000 %to35 5000to 5700 s t o a FA) at045 -45to55 -Morethan55 u * Source: Own calculations based on GSO data. 86 LABOR MARKET Wages and the cost of labor The average wage in Vietnam, based on the VHLSS 2004, is about 824 thousand dong per month, or roughly55 dollars. This figure is an average computedover both salariedworkers and casual laborers, including the value of in-kindbenefits. The figure is admittedlycrude; for instance, in-kind benefits are most likely under-estimated. But even if it is not highly precise, it confirms that labor is cheaper in Vietnam than in neighboringcountries. The picture is similar when data from FDI companies is considered instead. According to a recent report by MPI, the averagewage of a laborer inthis sector is around75 to 80 dollarsper month, the averagesalary of an engineer is about 220 to 250 dollars, and that of an administrative officer is close to 500 dollars. And this is despite the fact that wages inFDIcompanies are considerablyhigherthan in domestic enterprises operatinginthe same sectors. Labor costs are by and large unaffectedby minimumwages. At 350 thousand dong per month, the most generally applicableminimum wage is below what most registered enterprises pay; and substantiallybelowthe earnings of skilledworkers. A higherminimumwage applies to foreign companies, but the dual regimewill need to be abolished once Vietnam enters the WTO, as it contradicts the "national treatment" principle. In fact, the minimum wage is relevant in Vietnam mainly because of its implicationsfor public sector pay, and potentially for the pension obligations of the government. But it can hardly be considered as a binding constraint from a labormarket perspective. Moreover, focusing on remunerationlevelsonly leads to an over-estimationof the "true" cost of labor in Vietnam. It amounts to ignoring the remarkable work ethic of Vietnamese workers. In many countries, one of the main problems faced by business is how to elicit effort fromtheir employees. But shirkingis muchless prevalent ina populationwhich canbe described as workaholic. This unusual willingness to work hard i s confirmed by the WVS. Many respondents inVietnam declare that work is an importantpart o f their lives, which is common in many other societies around the world. What is uncommon, however, is the attitude towards leisure. Only 7 percent of Vietnamese respondents see it as an importantpart of their lives. This i s a remarkably low percentage even by the standards of other hard-working societies, such as Japan(40 percent)or the UnitedStates (43 percent). However, there are considerable differences in remunerationlevels within Vietnam, and this is another importantdriver of migration(Figure 7.1; right panel). Again, usingdata fromthe VHLSS 2004, the averagemonthlywage inHCMC was close to 1.5 million dong per month, and it exceeded 1.2 million in Quang Ninh and Hanoi. At the other end, monthly wages in Hau Giang, Tra Vinh, SOCTrang, HaNamandAn Giangwere below 0.5 million. It is also interesting to note the contrast between the Mekong Delta and the North East mountains. Inthe former, a large fraction of those at work are wage earners, but they do not make much. This is because of the prevalence of casual, agriculturalwork. Inthe latter, wage earners are a small fraction of all employed, but their earnings are high. This is because most wage earners work for the public sector, inone way or another, whereas privatesector employment i s notmuchdeveloped. Remuneration levels depend not only on the province workers live in, but also on their own characteristics. Educationalattainment and work experience are among the most important determinants of earnings at the individuallevel. But the labor market does not necessarily treat equal individuals in an equal way. Important gaps exist acrossjobs, and especiallybetween the public and the private sectors. Pay levels in the former are not necessarily aligned with the market, which may result in some public sector workers beingprivilegedand others beingunder- 87 paid compared to their private sector counterparts. There are also important gaps betweenjobs in urban and in rural areas. And women tend to earn less than men in every society, with the magnitude o f the gender gap in pay saying something relevant about the extent of inequality among sexes. Standard statistical tools allow estimating these gaps on the basis o f individual data from surveys such as the VHLSS. The estimates can be interpreted as the percentage change in earnings associated with a change inthe individual or job characteristic considered. For instance: by how much does remuneration increase ifthe person has one more year o f education, or ifthe person lives in an urban area, or works for government, or i s a male? While such interpretation involves some important methodological caveats, it still provides a telling picture o f the Vietnamese labor market, and how it has changed over time. Figure 7.2: Determinants of Labor Earnings I I - 7 , I 0 2 6 38 c -5 5 .-n 8& -10 c (I) -15 m -20 2n 0 -25 -30 0 1993 1998 2002 2004 . -35 One more year of education Female worker I 50 I " 45 c 3B 40 35 30 (I) p 25 2 20 0 `ii '5 g 10 g 5 0 1993 1998 2002 2004 Government employee Works in a major city Source: Own calculations basedon GSO data. The labor market premiumto education has become more important over time. In 1993, one additional year of schooling was associated with a 1 percent increase in wages, everything else equal; ten years later it was associated with a 6 percent increase (Figure 7.2). This apparently 88 LABOR MARKET minor gap makes a substantial difference when compounded over several years o f schooling. In 1993, an employed person with a university degree earned, on average, 10 percent more than someone with primary education only. By 2002, the average gap had widened to 80 percent. On the other hand, women are less disadvantaged than in the past. Inthe early 1990s, the average woman worker earned 32 percent less than a man o f the same age, education and experience living inthe same area. By2004 the gender gap had fallen to 17percent. The relative position of government employees also changed considerably during this period. By 1998, a civil servant earned 28 percent less than an otherwise identical wage earner out o f government. However, as a result o f the public administration reform agenda and its associated pay increases, by 2004 the situation had been reverted, and civil servants earned 30 percent more than their equivalents out of government. It should be noted, however, that these figures are averages across all government employees. Pay tends to be more compressed in the public sector than out of it. Therefore, civil servants in professional and technical occupations could still be underpaid compared to their private sector counterparts. Civil servants at the low end o fthe hierarchy, however, are most likely overpaid. Finally, the results indicate a substantial wage premium associated with employment in major cities, compared to rural areas. Except for the year 2002, for a worker with average characteristics interms of age, education and experience, the gap was between 33 and 44 percent. The smaller gap estimated for 2002 may say more about the quality of household survey data on that particular year than about the wage gap. Admittedly, living in cities i s more expensive as well. Butthe magnitude ofthe gap is revealing ofthe migratory attraction large cities might exert on rural populations. On the other hand, the wage premium associated with small cities, not reported here, i s much more modest. Several other studies, relying on different methodologies, add important nuances to these findings. An analysis of wage inequality conducted by ILSSA, using data from the 1998 VLSS and the 2002 VHLSS, shows that the payoff to an additional year of education varies across groups of workers. The returns to schooling are indeed higher in urban areas than in rural areas. They are highest inFDI companies, followed by the domestic private sector. Interestingly, they are also higherfor female workers than for male. Data from the three-wave round o f SMEs, also by ILSSA, reveals a remarkable equalization o f wages over time. In the 1991 round, there were large differences in labor remuneration between the north and the south, between rural and urban areas, and between men and women. By the 1997 round, these differences had diminished greatly, with the notable exception of the rural-urban gap. For instance, in the 1991 round, average wages in HCMC were roughly twice those in Hanoi and Haiphong; by 1997, the gap was in the range o f 30 to 50 percent. Wage differences across ownership forms also diminishedconsiderably. For instance, by 1997 remuneration levels in household business had more or less caught up with pay in registeredenterprises. The gap in remuneration levels between urban and rural areas is suggestive o f labor market segmentation. Barriers to geographical mobility, from transportation costs to registration permits, make segmentation likely. In practice, this means that urbanjobs are "better" than the jobs available to those living inrural areas. Evidenceon this gap injob "quality" i s reported in a study by Mekong Economics, showing that four out of five factory workers were satisfied, and would promote working in a factory to their friends and families. Further proof i s provided by the willingness of many workers to pay a recruitment fee to get a factory job. Interestingly, the study found that more than 11percent of women pay for theirjobs, comparedto only 7 percent of men. The average payment to work in a foreign company was 274,000 dong; roughly half o fthat i s paid for an SOE job. There i s also evidence that migrant workers "pay" for these better jobs 89 BUSINESS through lower levels of remuneration, compared to their non-migrant colleagues, as if their migrant status putthem ina weaker positionwhen bargainingwith employers (Box 7.1). Box 7.1: Migrants in the Textile and Garment Sectors Do migrant workers earn less than what can be expected on the basis of their education and experience? And are they less likely to receive training than similar non-migrantworkers? A survey of 150 textile and garment f m s , including individual interviewswith eight workers in each, can be usedto attempt an answer to these questions. Overall, 15 percent of the interviewedworkers were migrants; and three quarters ofthe migrants were female. Migrant workers inthe survey are 25 years old on average, as opposedto 30 years for non-migrant workers. They have less education too: only 26 percent has completed upper secondary school, against 55 percent of the non-migrantworkers. And they also work for less than their non-migrantcounterparts. On average, they make 877,000 dong per month, compared to 937,000 in the case of non-migrants. But do they really earn less after taking into account the fact that they are younger and less educated? Not on the surface. Standard statisticaltools suggest that they do earn more. But this is a spurious result, due to very nature of geographical mobility. Migrantstend to move to the regions and f m s that pay relatively well. Once the overall "premium" for firm and location are taken into account, the results suggest that migrantworkers earn 13 percent less, everythingless equal, than non-migrantworkers. The same methodology, involving individual characteristics but controlling for fm and location, canbe usedto evaluate access to training. Inthis case, it appears that migrantworkers do better than non- migrants, as they receive more training. However, in spite of this additional training, they are less to work inskilledjobs comparedto similar, non-migrantworkers. Source: Basedon Remco Oostendorp (2004). Other common forms of labor market segmentation appear to be less prevalent in Vietnam. In particular, a recent study by ADB did not uncover significant gaps in the labor market payoffs to individual characteristics between formal and informal sectorjobs, nor between farming and off-farm jobs. Industrial relations There is limitedindustrial conflict inVietnam. While the number o f work stoppages has increased inrecent years, data collected by the Vietnam General Confederation of Labor (VGCL) and the ILO reveals barely more than 700 strikes since the introduction o f the Labor Code, in 1994 (Figure 7.3). Most of these stoppages were in foreign companies in the South. But even there, strikes remain a rare event. There were only 100 o f them in 2003, at a time when there were about 2,650 foreign companies in operation. The domestic private sector is even less disruptedby industrial strife. In2003, at a time when the number of private domestic enterprises reached 65,000, there were still less than 40 strikes. Reports suggest that most strikes have resulted from legal or contractual breaches, including failure to pay wages and benefits, failure to pay social insurance contributions, and failure to pay severance pay at termination. In some cases, industrial conflict was prompted by demands by workers to do overtime beyond what i s allowed by law. SOEs seem to comply with 90 LABOR MARKET labor regulations more thoroughly than private firms. For instance, they report a larger share o f their workforce for social insurance and health insurance. Figure 7.3: Number o f Strikes by Enterprise Ownership Numberof Strikes in Viet Nam by EnterpriseOwnership 1995 to 2005 100 - i State-Owned Enterprises Foreign-Invested Enterprises Private Vietnamese) ~ Owned Enterprises 1 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 20/5/05 Year Source: Jan Jung-MinSunoo (2005) and Simon Clarke (2004). However, formal compliance i s not always reflected in better working conditions. For instance, in a series of formal interviews conducted in the garment and footwear sectors, by Mekong Economics, workers at SOEs reported more access to first aid, protective clothing, drinkingwater and wash rooms than in other factories. Workers employed by foreign companies reported the least access. Butthis contradicted the findings from the focus group discussions and the observations carried out for the study. In practice, infrastructure in SOEs was found to be poor compared to foreign companies. Working conditions in the latter also appear to be better than intheir counterparts inneighboring countries (Box 7.2). The 2002 revision of the labor code sets out the procedures to follow incase o f collective labor disputes. But these procedures are clearly not followed inpractice. Inprinciple, in case of conflict a written request for conciliation must be submittedto the Labor Conciliation Council in the enterprise, which must react within seven days. Ifeither part rejects the Labor Conciliation Council's proposalto resolve the dispute, it can refer the dispute to the Labor Arbitration Council, which begins work within 10 days of receipt o f a written submission. Ifeither part disagrees with the proposed solution, the matter can be referred to the People's Court. But at this stage, the Executive Committee of the trade union can decide to strike, based on a secret ballot. A strike which occurs while this reconciliation process is underway, or which is not endorsed by more than half the union members, i s considered illegal. Although all strikes have been technically illegal, sanctions which can be used against unlawful strikers have almost never been invoked. Police action against striking workers has not been part o f the government response either. In practice, collective labor disputes may be resolved through other channels, such as the human resources manager of the enterprise calling DOLISA to play a mediating role, despite this not beingpart of its formal mandate. 91 BUSINESS Most factories have trade unions, but many of these do not operate in practice, and some exist in name only. Trade unions are more active in SOEs. But only one third o f foreign companies and 15 percent o f domestic private enterprises have a collective agreement in place. The study on the garment and footwear industryby Mekong Economics, mentioned above, shows that trade unions mainly provide social services to their members, such as organizing social events, sporting activities and field trips. In some cases they provide interest-free loans to their members, as well as bonuses and gifts on special occasions such as Tet, International Women's Day or Children's Day. In a similar vein, a survey conducted by ILSSA in 24 enterprises found that only 16 percent of workers felt that the trade union had a role indispute resolution. Box 7.2: TaiwaneseFirms in China and Vietnam How is human resource managementaffectedby the host-countryenvironment? A recent study tried to answer this question by interviewing local officials, trade union representatives and expatriate middle-levelmanagers in36 Taiwanese enterprises. The study concludedthat working conditions were better in Vietnam than in China, with labor standards being more rigorously upheld, especially in relationto working hours andpenalties. Taiwanese managers said that the "militaristic" style of management that they apply in China was not effective in Vietnam and that they were therefore adopting a "soft" management technique. The following quotes illustratedthe difference in approaches and emphasize the workers' awareness of their rightsas an important factor: "You cannot even touch Vietnameseworkers, let alone abuse them. In China, we have used a Taiwanese management style. When we began our operations in China, we frequently resorted to punishment. Physical punishments were very common, including even hitting, like in the military. " "Their human rights awareness is very high.... That is why I think Taiwanese who are into shoemaking here have toface a lot of labor disturbances and strikes. They easily stage mass protests. I n our company this happened last year on the night of the Moon Festival. Their labor and democratic consciousness is very high. .... This is not just a problem only at my factory; this is a problem of the entire society. I n Vietnam their protection of labor rights is too stringent. This is something totally unexpected.'I The study identified other factors which have been important in improving working conditions, relative to China. Of primary importance was the different application of the household registration system. In China, employers may withhold documents, effectively preventingworkers from changing jobs. InVietnam, by contrast, employers are not involved inhouseholdregistration. The provision of company dormitories, which is common practice in China but not in Vietnam, also allowed greater control over workers' movements. The study also argues that the VGCL, though still a state-controlled organization, enjoys more autonomy, and some support from the international labor movement. It claims, finally, that differences in the decentralizationprocess have reduced government control over labor standards more quickly inChinathan inVietnam. Source: Anita Chan andHongzenWang (2004). Labor inspections are not a particularly effective mechanism to enforce labor standards either. As of 2004, the labor inspectorate o f MOLISA had a staff of about 300 inspectors and related officers. Since 1995, inspections have been carried out in only 1,635 enterprises. Respondentsto the ICs by the World Bank are ambivalent about visits by labor inspectors, which 92 LABOR MARKET are regarded as being neither particularly helpful nor unhelpful; perhaps because there are not many of them. Nor are they especially demanding: ICs respondents reported that gifts or unofficial payments were expected in about one quarter of inspections by MOLISA, which i s considerably less than with other government agencies. On the other hand, the concept of CSR has been embraced by some private enterprises in Vietnam. CSR i s based on the notion that businesses can voluntarily adopt social, environmental and ethical practices which are both profitable and good for economic development. The incentive to introduce CSR is stronger among multinational companies with valuable brands, among labor-intensive firms working under contract and in enterprises with a high visibility (publicly traded, or in high-impact sectors such as oil or mining), among others. In Vietnam, adoption o f CSR principles is emerging in garments and footwear, where enterprises have been requiredor encouragedto adopt one or more codes of conduct (Box 7.3). These codes o f conduct normally force compliance with the Labor Code and with a set of principles covered by the ILO "core conventions" on forced labor, child labor, discrimination, freedom o f association and collective bargaining. They also typically cover health and safety, wages and hours of work. Box 7.3: Corporate SocialResponsibility inVietnam Almost all Vietnamese suppliers to the global chain in textile and footwear production first confiont CSR through codes of conduct developed by individual buyers. Through these codes, the buyer specifies social and environmental standards to be met and acts as the monitoring partner (either directly or through third parties). Suppliers usually accept the terms in the hope of improving their reputationand attractinglarger andmore stable contracts. About 1,000 codes are in operation globally, almost all developed by individual companies focusingof their businessneeds. Not all of these codes are beingusedinVietnam. But many suppliers in Vietnam are working with more than one buyer, so that they have to comply with more than one code. Some large suppliers deal with up to six ofthem. Significantnumbersof workers are coveredby some company codes of conduct. For example, Nike, a producer of athletic shoes and sports apparel, estimates that 84,000 workers are coveredby its code of conduct across the 34 Vietnamese f m s in its supply chain. Adidas-Salomon, a competitor, reports nearly 43,000 workers under its own code. These codes appear to be enforced. In 2004, Adidas-Salomonterminated business relations with one of supplier due to excessive working hours, insufficient time-recording systems, incorrect payment of wages and benefits, improper worker- managementcommunicationandpoorhealth, safety and environment conditions. Basic information on CSR is providedby business associations, such as VITAS inthe case of textile, and LEFASO inthe case of footwear. But the specifics of the codes are detailedby the buyer. This top-down relationship between buyers and suppliers means that most Vietnamese enterprises engaged inCSR are passive implementers. Few ofthem are developingCSRpracticesontheir own, in the hope of increasing their market share. However, suppliers which were initially responding passively to buyer requirements now see CSR investments as a way to increase workers' satisfaction andreduceturnover andreportedillhealth. In parallel, a growing number of enterprises are choosing to demonstrate commitment to responsible practices by complying with internationally-recognizedfactory certification schemes, such as WRAP and SA8000. Vietnam had 27 SA8000-certified facilities in September 2005, covering nearly 28,000 workers. This placed it in fifth place, worldwide (after China, India, Italy and Brazil), in terms of number of workers covered. Source: World Bank (2003), ILSSA (2004). 93 Unemployment The unemployment rate of Vietnam is low by international standards. Admittedly, household surveys and labor force surveys yield estimates which are not strictly comparable (Table 7.1). The former have a more representative coverage of the population, whereas the latter allow computing the unemployment rate in exactly the same way as the ILO would recommend. But for the country as a whole, there is no doubt that the unemployment rate is very low. For instance, the Survey on Employment and Labor, conducted by MOLISA in 2003 reports 0.95 million unemployed, at a time when they were 41.31 million labor force participants. This i s equivalent to an unemployment rate of 2.30 percent, very close to the 2.16 percent rate estimated basedon household survey data. Low unemployment rates are not surprisingin a still predominantly rural economy. The problem in rural areas is seasonal underemployment, rather than open unemployment. But even for urban areas, unemployment rates are quite low, ranging between 2 and 6 percent depending on the data source used. Moreover, detailed information on the characteristics of the unemployed, obtained from the VHLSS 2004, shows that they are mainly young and relatively educated (Figure 7.4). Needless to say, joblessness can be a painful experience, even for first-time job seekers who enjoy considerable family support; and it may become a source of frustration and social tensions. Also, a low unemployment rate may hide considerable under-employment, especially in rural areas during periods of low agricultural activity. But based on this key statistic, open unemployment i s not ofVietnam's main problems. Figure7.4: KeyCharacteristics o fthe Unemployed I EdEmployed Unemployed 1 I tinployed Uneqloyed Source: Own estimatesbased on data from GSO. Figuresare for 2004. The flip side of a low unemployment rate is usually a high vacancy rate, and this is probably what respondents to the ICs by the World Bank complain about, when they identify skills and education of available workers as a major or severe constraint. A high vacancy rate i s associated with high labor turnover, with the best workers voluntarily moving to (or being poached by) other enterprises. A recent study on FDI, by CIEM, i s quite compelling in this respect. It shows that over the period from 2001 to 2003, the labor turnover rate among foreign companies reached a stunning 43.4 percent. Turnover is highest in textiles, garments and 94 footwear. Among those changing jobs, 42 percent are skilled. According to information provided by the companies which lost workers, 32 percent of them moved to other foreign companies, 23 went on to establish their own business and 18 percent took jobs in domestic enterprises (information about the rest is not available). Labor turnover also appears to be highest in the south. The three-round survey of SMEs conducted by ILSSA found a marked difference between Hanoi and Haiphong on the one hand, and HCMC on the other. InHanoi a little more than 10 percent of the enterprises surveyed inthe second round had lost any wage workers, and almost as few had recruited any new workers. In HCMC, by contrast, almost half of the enterprises had lost workers and an equal proportion had hirednew workers. It also appearedthat this highturnover was mainly voluntary, with very few instances o f workers being fired. Inthe vast majority of cases the initiative seems to have come from the worker, not from the enterprise. On the other hand, there is a latent unemployment problem in SOEs, many o f which inherited a bloated workforce from the period under central planning. The extent o f labor redundancy i s difficult to estimate in practice. However, there are several indications to suggest that labor redundancy in SOEs i s still considerable, although its distribution across activities i s probably quite uneven. One indication comes from the ICs conducted by the World Bank, as it directly asks managers what they would do if they were allowed to cut their workforce without restrictions. All firms responded that they would downsize to some degree, but SOEs would go on average for a 14 percent cut, compared to only 3 percent for domestic private firms. Yet, SOE directors may not have an incentive to fully report the extent of over-staffing, as a bloated workforce could be seen as reflecting poor management. Another piece o f evidence comes from the operation of the social safety net fund for redundant SOE workers, institutedunder MOF in 2002 to assist the equitization process. This safety net fund provides more generous compensation for job loss than stipulated in the Labor Code. Workers who are either separated from their jobs or volunteer to leave are offered two months o f basic salary per year o f service. In addition, they receive a training allowance and a job-seeking subsidy together equivalent to 12 months of salary, and a lump-sum o f 5 million dong. Two tracer studies of retrenched SOE workers assisted by this fund indicate a relatively high level of satisfaction, suggesting that SOEs may not be afraid of downsizing through this mechanism. On average, 49 workers were separatedfrom the 2,228 SOEs supported by the fund, which corresponds to an average redundancy rate of 19 percent. But this figure is an under- estimate too, as it does not include workers separated through early retirement, or workers hired after the beginningof the equitization process, who are not eligible for assistanceby the fund. Social protection Before the approval of the Labor Code, in 1994, only public sector workers were entitled to formal social protection inVietnam. The Labor Code stipulated that all workers who engaged in labor contract longer than three months or with an indefinite term were to be covered by a compulsory social insurance program. The total contribution rate towards this program amounts to 20 percent of the salary, formally split between 15 percent to be paid by employers and 5 percent by workers. The Labor Code also places the responsibility on businesses engaged in agriculture, forestry, fisheries and salt production to participate in forms o f social insurance suitable to their production and employment characteristics. In 2005, coverage was extendedto all workers inregistered enterprises, regardless o ftheir type o f contract. 95 The social insurance system provides a series of short- and long-term benefits, the most important o f which is old-age pension. Menare eligible to retire at age 60 and women at age 55, provided that they have contributed for a minimumo f 20 years. The pension level i s definedby a formula involving the average wage over a certain period of time and a replacement rate. The latter varies with the number of years of contribution, from 45 to 75 percent. The former i s computed over the last five years for public sector workers, but over the entire enrollment span for private sector workers. Because there is no indexation mechanism in the formula, the "average" wage to which the replacement rate i s applied falls short o f the average real wage o f the retiring workers. In the case of the private sector, where the average is computed over the entire contribution period, it does so by a considerable margin. Those who have contributed for less than 20 years are entitled to a lump-sumpayment, much lower than the present value o f the contributions made. Given the relatively low coverage of social insurance in Vietnam, workers changingjobs runthe risk o f losing their entitlement to old-age pension. The expansion in the number o f enterprises which are formally registered and the sustained growth in wage employment had led to an increased formalization of the economy (Figure 7.5). This is reflected in the growing share o f labor relationships involving enrollment with social insurance, a long-term horizon, written contracts or other elements o f protection, as captured by household surveys and other statistical instruments focusing on individuals. But the trend is also captured through the administrative records of the Vietnam Social Insurance (VSI) agency. Moreover, a substantial increase informality i s expected in 2005, due to the extension o f coverage to all salaried workers inregistered enterprises, regardless of their type of contract. Figure7.5: A gradual Formalizationo fEmployment I I 6.0 5.5 5.0 - .-.- ;4.5 -Ec 4.0 3.5 3.0 2.5 1996 1997 1998 1999 2000, 2001 2002 2003 2004 I Source: Basedon data from VSI. The formalization processcould hide, for quite a longtime, the fact that the current social insurance program of Vietnam is not financially viable. At present, and setting aside retirees from the program that was inplace until 1995, the ratio o f contributors to beneficiaries i s close to 17, which is abnormally high by international standards. Moreover, this ratio only decline slowly. Becausethe population is young, over a long period the cohorts entering the labor market will be much larger than those reaching retirement age. Also, becausethe economy will become 96 more urban and more formal, the share of the labor force contributing to social insurance will increase steadily. As a result, the ratio o f contributors to retirees could be roughly two by the middle of the century. Yet, in spite o f this favorable context, the current social insurance system could be expected to move from surplus to deficit in about three decades. Under realistic assumptions, the surplus would never exceed 1.5 percent of GDP; by 2050 the deficit could be in the range o f 3 to 5 percent of GDP. Either budget support or much higher contribution rates would be neededto fill the gap. The financial weakness of the current program is aggravated by its benefit formula, which discourages long contribution periods. On the surface, the compulsory program i s generous, as the payable pension increasesby 2 percent o f the average salary for every additional year o f contribution inthe case of men; and by 3 percent inthe case of women. But this apparent generosity applies to an "average" salary computed without any indexation, which brings its value down compared to the actual salary at retirement age. This combination of nominal generosity and real tightness creates a considerable wedge compared to a "purely" actuarial formula. Pension benefits inVietnam are higher than actuarial benefitswhen the number o f years of contribution is low, butthey become lower as the number of years increases. Last but not least, the gradual expansion o f the compulsory system through the formalization of employment only will fail to provide social protection to a large share o f the population, including the poor. Even the introduction of voluntary social insurance programs could fail in this respect, despite the fact that the poor stand to lose more than other groups from adverse shocks. But precisely because they are poor, they might be unable to afford even the minimum level of contributions needed to participate in a voluntary program. Laying the foundations for a universal system requires that specific mechanisms be considered to reach this group. Those mechanisms necessarily involve an element of subsidy, to be paid out o f general tax revenue. Fortunately, the government of Vietnam i s fully aware of these shortcomings o f the current social insurance system, and a new law is under preparation to address them. The specifics may still evolve, but the intention i s to make the system viable, to expand the coverage of the compulsory program, and to lay the foundations o f a universal social protection system. 97 8. INFRASTRUCTURESERVICES Enormous progress in infrastructure development has been accomplished in Vietnam in recent years. The country i s by now investing more than one third o f its GDP. About a tenth o f GDP goes into infrastructure; as a result, it is rapidly catching up with its neighbors in terms o f availability and cost o f services. Electrification and telephone penetration have seen the most remarkable improvements, and the road network has expanded considerably as well. Still, enterprises in Vietnam complain about insufficient transport infrastructure, and excessively expensive electricity and telephone services. Making further progress in infrastructure development requires a diversification of funding sources and improved transparency in resource mobilization, especially at local levels. The adoption of sound regulatory frameworks, facilitating cost recovery and promoting competition, could also help attract private participation in infrastructure and further contribute to business development inVietnam. There have been a few promising initiatives in this respect, across most infrastructure sub-sectors, especially in terms o f increasing competition among suppliers. However, most business participants are SOEs. Looking forward, private sector participation cannot be relied upon to address, alone, the infrastructure needs o f Vietnam in the coming years. Nor even to cover a large share of those needs. In this area, business development crucially depends on the quality of public investment, and on the appropriate pricingo fthe services it provides. A regionalperspective There has been impressive growth in access to infrastructure services since the early 1990s. All urban areas inVietnam are electrified by now. Inrural areas, electrification expanded from 51 percent of all households in 1996 to 88 percent in 2004. The length of the road network increasedfrom 96,100 km in 1990 to 224,500 km in 2004. Inthe case o f national-level roads, the increase was from 15,100 km in 1997 to 17,300 km. By 2002, 45 percent were deemed in good condition, compared to 37 percent in 1997. Access to improved water grew from around 35 percent of the population in 1993 to 87 percent in 2004, and access to hygienic latrines from 10 to 32 percent. The number of fixed and mobile lines per 100 people increased from 1.1 in 1995 to 9.2 in 2002. It i s on target to reach a total of 10 million lines in 2006-2007, achieving one o f the fastest growth rates of tele-density on record. Regional comparisons confirm how far Vietnam has come in terms o f infrastructure development. In some sub-sectors, such as electricity and water, access to infrastructure services approaches those of richer countries in the region. In others, like sanitation and telephones, access levels are some distance behind the best regional performers. But at comparable development levels, Vietnam does better than average for sanitation and about average for telephones. Overall, the increased availability of infrastructure has resulted inprices increasingly aligned to those inthe region (Table 8.1). Perceptions by entrepreneurs are not necessarily that positive, however. The ICs by the World Bank uncovers concerns about both the quality and the prices o f the available infrastructure services (Figure 8.1). Insufficient roads and an excessive cost o f electricity and INFRASTRUCTURESERVICES telephone services are among the main concerns o f respondents. Interestingly, it i s the SOEs which are most concerned about electricity prices, with much fewer domestic private firms considering it a severe or major constraint. Table 8.1: CatchingUp with the Neighbors Hanoi HCMC Bangkok Jakarta Manila K.Lumpur Industrial electricity (centskWh) 5.5 5.5 4.2 5.0 10.0 5.0 40-feet container to Yokohama (dollars) 1630 1150 1300 990 950 725 Call to Japan (dollars/three minutes) 1.95 1.95 1.49 3.78 1.20 1.42 ADSL connection (dollars/month) 76.3 76.3 14.6 782.1 25.4 162.6 Source: JETRO (2005). Figuresrefer to 2004. When are asked which infrastructure improvements would have the greatest impact on their business, more than 40 percent o f I C s respondents agree that improvements to roads and electricity would be most beneficial. In contrast, firms felt that improvements to railways and airports were much less important for their business. But even in the case o f roads, there are interesting regional differences. Respondents inthe Southeast and the South Central Coast would give priority to upgrading the national network; those in the Red River Delta, on the other hand, favor better inter-provincial roads. Figure 8.1: Problems with Infrastructure: the Views o f Entrepreneurs ::::% ~ 60.0 50.0 mPoor quality 40.0 0Slow service 30.0 OToo expensive ~ ~ 20.0 ~ 10.0 I 0.0 National Provincial Electricity Telephone roads roads Source: Basedon ICs survey. 99 The funding for infrastructure At more than one third of GDP, Vietnam's investment rate is already very high by international standards. Although it i s difficult to estimate precisely the share of infrastructure in this total, at present Vietnam could be devoting close to 9 percent of GDP to improvements in electricity, transport, water and sanitation and telecommunications (Table 8.2). The figure would be substantially higher ifinvestments inoil and gas were counted as well. 1999 2000 2001 2002 2003 Electricity 2.81 3.39 3.23 Transport 2.81 2.64 3.71 4.03 Water and sanitation 0.53 0.52 0.53 0.52 Telecommunications 0.80 While fast accumulation boosts future productivity, the cost is a diminution of current consumption, possibly leading to slower poverty reduction. It would be difficult to lift the investment rate much above its current level. But even without thinking o f an increase, just fundingthe current investmenteffort represents a formidable challenge, interms of bothresource mobilization and transparency. The resources mobilized in the process come from a variety o f sources, including the government budget at both central and local levels, state development credit provided by DAF, the issuance of infrastructure bonds, borrowing from SOCBs, own-source revenues, and partnerships with the private sector (Table 8.3). Apart from the budget, the key public financing institution for infrastructure i s DAF. Established in 2000, in an attempt to take policy lending activities out of SOCBs, DAF i s a specialized facility operated by MOF, with branches throughout the country. It serves as a mechanism for aggregating and managingfunds from a variety of public sources, and on-lend to a range o f projects. Its interest rates are lower than the cost of raising funds, with the government subsidizing the difference. In 2004, about 40 percent of DAF's resources were from ODA, 15 from the issuance of government bonds, 12 percent from domestic trust funds, 10 percent from social security fund, and 6 percent from postal savings, among others. Ofthe resources mobilized domestically, about 80 percent has been on-lent to SOEs, although this is often for the implementationof public infrastructure projects. In2004, the DAF's outstanding loans amounted to 11percent of GDP. More recently, the government started issuing bonds for specific infrastructure undertakings. For instance, during the next five-year cycle bond issuance will be used to finance projects such as the north-south Ho Chi Minh Highway, roads along the borders with China and Cambodia, and irrigation works in the disaster-prone central provinces. In 2004 the government issued bonds for about 0.7 percent o f GDP. In 2005 it issued Vietnam's first sovereign bond, for 100 INFRASTRUCTURE SERVICES roughly 1.5 percent o f GDP. Bonds are kept off-budget, to comply with the State Budget Law, which limits the budget deficit at 5 percent of GDP. Table 8.3: Where Do the ResourcesCome From? Percent o f GDP Source Percent o f Transport Electricity Telecoms I Water All total Budget 0.1 1.o 11 ODA 0.3 3.2 36 ~ Govt. Bonds 1.2 14 SOCBs 0.1 1 Users 0.9 0.8 0.1 1.8 20 BOT 0.2 1.2 1.4 16 Private 0.1 0.1 1 Total 4.0 3.4 0.8 0.6 8.8 100 Source: World Bank (forthcoming). Figures are estimates around the years 2002 and 2003. ODA is channeledthrough DAF. At the provincial level, the first municipal bonds were issued by the HCMC government in 2003. Provincial governments are allowed to borrow up to 30 percent of their annual budget for development investments; but the cap is 100 percent inthe case of Hanoi and HCMC. Other rules for issuanceo f municipal debt are not yet clear; for instance; the tenure and the coupon rates of the bonds are largely determinedby MOF, rather than the concerned province. Issuance o f corporate bonds for infrastructure development was pioneered by EVN, in 2001. One of the main reasons to issue the sovereign bond in 2005 was actually to provide a benchmark for corporate bonds by state conglomerates such as Vinashin. This is in fact the ultimate beneficiary o f the resources raised through the sovereign bond. Thejustification for this use of the proceeds comes from the inability of Vinashin to borrow on a long-tem basis from commercial banks, despite having large export orders, due to the short-term nature o f their deposits. From this perspective, the sovereign bond issuance can be seen as the combination of a corporate bond issuance and a government guarantee. Except for construction SOEs working for transport projects transport sector, borrowing from SOCBs has not been a common way to finance infrastructure development. In fact, SOE borrowing illustrates the failure of SOCBs to award credit on a strictly commercial basis. Between 1999 and 2002, about 35 percent of transport commitments had been approved by the Prime Minister's office had not been allocated funding. Notwithstanding, the MOC and the Ministry of Transport (MOT) contracted their affiliated SOEs to undertake the work on the promise of subsequent reimbursement. SOCBs were encouraged to lend to these SOEs. Eventually, SOCBs had no choice but to grant loan rollovers, as the interest payments due exceeded enterprise capitalization. Arrears amounted to one trillion dong in MOC, and 1.2 trillion dong in MOT. More recently, Local Development InvestmentFunds (LDIFs) have gained importance at the local level. In principle, LDIFs offer an operational and legal structure for the provincial 101 governments t o focus o n infrastructure, including the ability to enter into contracts with the private sector (Box 8.1). Box 8.1: Local Development InvestmentFunds In light of the enormous infrastructure development needs at the local level, provincial governments havebeenencouragedto take greater responsibility for financing their own investmentprojects. LDIFswere thus establishedin 13provinces, withthe approval andsupport ofthe central government. They have become increasingly important in Binh Duong, Dong Nai, Hanoi and HCMC. In 2004, the corresponding provincial governmentschanneled 7, 9, 33 and 13 percent o f local investments through these funds. The HIFU, established in 1996, is the oldest o f all. There has been a ten-fold increase in the combined operational capital o f LDIFs since then. Meanwhile, their activities have expanded from lending to establishing joint stock companies engaged ininfrastructure development. Yet, the legal basis for the creation, organization and operation o f LDIFs is still weak. They are not covered by the State Budget Law, and no national regulations govern their operation functions, including accounting and personnel management. A decree on LDIFs i s still indraft form. Standards in terms of sector focus, risk weight, or portfolio quality are non-existent. As a result, provincial governments have significant discretion over their operations. Depending on the provinces, LDIFs combine in various ways elements o f the accounting, financial management and personnel regimes o f DAF, SOEs and SOCBs. Inpractice, LDIFsplay the role of a contracting vehicle with a unique legal standing, allowing the provincial government to directly partner with the private sector through joint ventures and other financial arrangements. They are also a legal entity that can be usedto mobilize funds for infrastructure development from the market and other sources. From a budgeting perspective, they are used as a tool to earmark resources. Provinces such as HCMC and DongNai are also taking advantage o ftheir LDIFs as unitswith dedicated staff specializing ininfrastructure projects. LDIFs support the development of transportation, water, health, education, housing and industrial park projects. This support i s conducted in various forms, such as the direct financing o f projects, joint investments with other business entities, and contribution o f capital to establish enterprises which in turn develop and implement infrastructure projects. However, lending constitutes the bulk o f LDIF activities so far. For instance, during its eight years o f operation, HIFU it has provided 2.9 trillion dong inloans and only 0.2 trillion indirect investment. Lending tends to fall under any o f three main modalities. First, there are syndicated loans, inwhich LDIFs take the lead in calling for capital from local banks, and especially SOCBs. Second, there is lending in combination with central government development investment credit. And third, there i s independent lendingfor priority projects inthe province. The funds managed by LDIFs are a combination o f direct contributions from the state budget, proceeds from the sale o f state-owned housing and borrowing from SOCBs and DAF. In the case o f HIFU,they also include the issuance of municipal bonds. The direct participation ofthe private sector inHIFU, either as an owner or operator ofinfrastructure assets, remains minimal. Butit could increase thanks to a recently established framework for contracting. Source: World Bank (forthcoming) and ADB (2005). Large utilities are to an extent self-financing their investments. In the electricity sector, EVNcovers its operating and maintenance costs, and makes a contribution to capital costs. Over the period 1997-2002, roughly a third of EVN's investments were funded through retained 102 [NFRASTRUCTURE SERVICES earnings. EVNfurther diversified its finance sources in 2004, when it started issuingbonds and selling part o f its equity. Intransport, ports and railways earn enough revenue to roughly cover their operating costs. Water utilities, on the other hand, have struggled to cover even operating expenses but a new tariff-setting directive encourages full cost-recovery including capital costs. In the telecommunications sector, the Vietnam Post and Telecommunications Administration (VNPT) is entirely self-financing. It funds its own investments from retained earnings and by borrowing, and makes enough profits to contribute to the State budget with dividends. Private participation Energy has beenthe principal recipient o f private investments, mainly under the form o f operators involved in small-scale generation and distribution. There are approximately 100 o f these operators, jointly providing close to 1000 M W of power. The largest i s the 375 MW Hiep Phuoc plant built by a Taiwanese company to supply an industrial zone south o f HCMC. Only two projects involving private participation in infrastructure development have been completedto date. These are Phu My 2.2 and 3, both of which are generating power from gas. These two projects, the first to be done under Vietnam's BOT laws, were funded by foreign consortiums and ODA loans. Both were structured with government guarantees, with Phu My 2.2 receiving in addition a World Bank partial credit guarantee. A third project o f this i s under development at OhMon 2, with investment from Unocal and Mitsui,among others. There has also been progress in the introduction of private equity into the energy sector through the equitization program. In 2005 EVN sold 35 percent of shares in a hydro-electric subsidiary company, raising about 30 million dollars. There are plans to equitize seven generating plants although the timetable i s not clear. Private participation in roads i s very limited. In Quang Linh, a private sector company built a segment of road in exchange for LUCs along the road corridor, a model which has been replicated in other locations. But with the exception o f the road to the Tan Thuan export processing zone, all other toll roads currently operational in the country are owned and managed by SOEs or LDIFs, most notably in HCMC. Complex land acquisition procedures and the difficulty to recover costs, given current tariff levels, have deterred private consortia from developing new toll roads. A BOT scheme will be used for the Phu My bridge, linkingdistricts 2 and 7 in HCMC, but it involves a consortium of SOEs. However, there are a few instances o f private companies obtaining construction and maintenance contracts. Looking forward, the government is seeking private participation in four highway BOTs, linking HCMC and Can Tho, Danang and Quang Ngai, Noi Bai and Ha Long, and Hanoi and Haiphong. It i s not clear whether the proposed private BOT deals will be financially viable without counterpart funding from government. A fifth BOT, for the highways linking HCMC, Long Thanh, Dau Day and Vung Tau, i s scheduled to be completed by the state-owned Vietnam Expressway Corporation. There has been no significant involvement o f the private sector in the development o f railways and airports to date. Yet, MPI has issued a list o f four rail corridors in which the government is seeking BOT participation. Meanwhile, HCMC has made progress regarding an unsolicited BOT project proposal to build an underground metro rail. In airports, private participation has been limited to small scale supply and construction contracts, but no significant concessions, management contracts or BOTs have been developed. Opportunities for private sector participation inthis sector remain limited due to the definition of strategic sectors in which full state ownership isjustified, underDecision 155. 103 Developments have been more encouraging in ports. Despite the dominance of public operators inthe sector, three private ports have been developed under greenfield BOT schemes: a bulk cargo facility at Baria Serese, a container handling terminal in HCMC, and a small private terminal serving the industrial zone at Dinh Vu. Two other private port projects are under development at Cai Map and Cailan. The-government i s also seeking private sector participation intwo large transshipment ports, at BenDinh-SaoMaiandVan Phong. Water and sanitation sectors are decentralized, with the central government setting policies and approving the largest projects, whereas provincial governments are actually responsible for delivering water supply services. This is done through water resource companies which are in fact provincially controlled SOEs. Maximum tariffs are set by the State Price Committee and typically cover operating expenses only, not maintenance or investment costs. Low revenue makes private sector participation unattractive. So far, it remains limitedto small- scale water retailers and some waste removal businesses. Private sector investment should be viable inbulkwater supply, but only two attempts have beenmade to negotiate BOTSinthis area, and they both failed. Intelecommunications, the liberalization of the sector has allowed multiple operators to enter the market. The leading player is VNPT, a government agency which receives financial support from the state budget. However, there i s a clear intention to promote competition. For instance, the two mobile phone operators controlled by VNPT, namely Mobiphone and Vinaphone, do compete with each other, perhaps in part because they business cooperation contracts with different foreign partners. The planned restructuring o f VNPT into three separate regional telecommunication companies, for the North, Central and South regions, is expected to further promote competition. Several enterprises have been licensed to enter the telecommunications market so far. Three of them, namely Electricity Telecommunications and Information Company (ETIC), Hanoi Telecom and the Military Communications Company (Vietel), are SOEs; one is semi-private (Saigon Postel) and one i s fully private (S-Fone). Some o f these enterprises had a small presence in the fixed-line market to begin with, becausethey operated their own networks. The mobile market is still controlled by Mobiphone and Vinafone. But the five competitors in operation are gradually gaining market share. In addition, there is private sector participation through business cooperation contracts in the mobile, internet and paging lines o f business. In Internet services, the dominant providers are state-owned. The Vietnam Data-communications Company (VDC) controls 62 percent of the market, followed by the Corporation for Financing and Promoting Technology (FPT), which controls 23 percent. However, the sector is open to private investment, both domestic and foreign, except for Internet exchange provision, which is still reserved for SOEs or enterprises inwhich the government holds absolute majority shares. Quality and prices Even if the relative share o f the private sector in infrastructure development was to double in the coming years, the public sector would remain in charge o f about two thirds o f infrastructure investments. And the hypothesis of a doubling i s probably over-optimistic. Between 1997 and 2003, private projects amounted to roughly 15 percent o f all infrastructure spending. But if a single large project for the development o f the Nam Con Son gas field and its pipeline is excluded, commitments fall to about 8.5 percent o f the total. The private share increased to about 16 percent in recent years, mainly due to the Phu My 2.2 and Phu My 3 generation projects, and the equitization of power plants could boost this figure. Further 104 increasing this share will difficult. Therefore, in the coming years the access to quality infrastructure at reasonable prices will still depend, crucially, on public investment. Improving the process through which infrastructure sectors are regulated and investment projects are selected i s a top priority in the coming years. The gradual phasing out o f ODA funding in concessional terms will make this priority all the more important. Assessing the quality of investment projects was less crucial in earlier stages o f the transition, for the simple reason that returns were bound to be very high in almost all circumstances, due to the dramatic shortage of modern infrastructure. As the most glaring gaps are being filled, and Vietnam catches up with its neighbors, the risk of making poor decisions becomes higher. Yet, the preparation of feasibility studies for public investment projects i s still characterized by important weaknesses (Box 8.2). In many cases, these feasibility studies serve relatively narrow objectives, such as securing funding from donors, or providing estimates of the required budget support. More rigorous analyses o f financial and economic returns, or the contribution o f a project to the attainment o f broader development objectives, or the risks associated with the project, or its potential impacts on the environment or on vulnerable social groups, are generally absent from Vietnamese feasibility studies. The process reflects a concern not for efficiency inthe allocation o f resources, but rather for getting the most value for money once budgets are allocated. This concern i s understandable in an economy where resource shortages were the norm. But as a result of it, project scrutiny before approval i s less developed than it should, given Vietnam's current development stage. As regards the pricing of infrastructure services, there is a wide diversity across sub- sectors. Much remains to be done inwater and sanitation. Issues are admittedly complex. Social and public health considerations may justify that relatively low prices be charged to households, and especially to the poor. But in this case a key issue i s to determine who should cover the difference. If water resources companies are not financially viable, it will be difficult to attract private participation to the sector and, more generally, to fund the investments needed to extend coverage (and it i s mainly the poor who currently lack coverage). User prices below cost recovery thus call for a subsidy from either central or local government. Public health considerations are even more prominent in the case o f sanitation, although to a different extent across waste types and service users. Thus, solid waste disposal, hazardous waste management and industrial wastewater treatment require a specific pricing policy each. And depending on the gap between prices and costs, each of these policies may also involve transfers from or to the budget. There are also challenging technical issues. For instance, water and wastewater services tend to be tightly linked, as many households use both. This often calls for a single "water" bill, with a single fee covering both services. However, in a developing country many households have access to water but not to sanitation, so that a single fee makes them pay for a service they do not receive. At the other end, much progress has been accomplished in telecommunications and electricity. In both cases, progress is facilitated by the fact that most providers o f infrastructure services (power companies, for instance) need to connect to a common network. Access protocols, involving technical specifications, pricingrules and dispute resolution mechanisms, are therefore the foundation for the development o f these two sub-sectors. But there are still important issues involved. The price o f transactions between the network operator and service providers can be spelled out in the protocol or determinedon the spot. As for the price charged by service providers to end-users, modalities vary across sub-sectors. In telecommunications, where users can switch service providers easily, competition is the norm, so that government 105 intervention is not required. Electricity, on the other hand, tends to involve more price regulation, as well as obligations interms of service coverage. Box 8.2: Common Weaknesses of Vietnamese Feasibility Studies Appraisal requirements Most frequent shortcomings Country and sector issues Lack o f national and sectoral perspective, especially inrelation to master plans. Development objectives Usually not derivedfrom high-level plans and strategies; generally narrow infocus. Funding options Top-down approach; possible overlap with other projects incase of donor fimding. Key indicators Logical frameworks linking actions and indicators are not commonly used. Alternatives considered Limitedanalysis of options based on technical, economic, social and environmental criteria. C. IMPLEMENTATIONAND OPERATION Institutional arrangements - -[Insufficient analysis o fthe commitment and I capacity o fthe project executing agency. Sustainability Less attention than warranted to operation and maintenance aspects o f the project. Monitoring and evaluation Usually based on outputs, not involving baselines or impact assessments. Critical risks Insufficient analysis o frisks (e.g. difficulties in land acquisition) and mitigation strategies. D.PROJECT SUMMARY Technical Limiteddetail on standards, operational aspects and environmental implications. Economic and financial Estimates drivenby cost norms; unreliable methodology for estimation o freturns. Fiduciary Assessment o f fiduciary risks and corruption mitigation strategies usually absent. Social Insufficient consultation and assessment o f potential impacts on vulnerable groups. Environmental Limiteduse o f environmental impact assessmentsand safeguards. Source: Adapted from KFW (2005). The Electricity Law, which came into effect in July 2005, is an important step towards efficient regulation. The Law basically calls for the unbundlingof the electricity sector and the 106 INFRASTRUCTURE SERVICES creation o f a competitive power supply market. This ambitious objective i s to be phased inover a period o f two decades, beginning with a single buyer model and eventually developing into wholesale and finally retail competition. However, critical strategic choices will need to be made over the next couple o f years, and major restructuring will need to be completed before 2009, when full competition in generation i s envisaged. Under this so-called "electricity market roadmap", it is expected that transmission assets will remain under state control. Inthe near term, the transformationo f the sector will require the divesting o f EVN generating units, so that they can compete in equal terms with private generators. This will be particularly challenging from a corporate governance point o f view, as EVNwill remainthe buyer o f electricity inthe nearterm. The plan calls for the introduction o f greater competition inthree phases. Inthe first one, generators would bid to sell electricity into a power pool, with EVN as the single buyer. This would lead to a simulated market, determining dispatch order o f generating plants, but without real financial transfers (most generators would still be owned by EVN; as for the others, they would be paid according to their private participation agreements). In the second phase, the simulated transactions would become real. And in the third phase, large consumers such as distribution companies or major industrial firms would be given the right to make bilateral contracts with generators, avoiding EVNas the single buyer. A number o f issues will need to be addressed to implement this plan, including the mechanics o f spot trading, the roles and responsibilities o f the electricity regulator, and the conditions and procedures for licensing each electricity activity. The Ministryo f Industry(MOI), will place these responsibilities with a new regulator. Recommendations on electricity retail tariffs will be made by MOI with the assistance o f the regulator and approved by the Prime Minister. Wholesale tariffs for electricity, fees for transmission and distribution, and for auxiliary services, will be proposed by the entities involved in each o f these activities, evaluated by the regulator and approved by MOI. 107 PARTIll: POLICIES FOR BUSINESS 9. GLOBAL INTEGRATION Integration inthe world economy has been one o f the main drivers o f economic reform in Vietnam. Fromparticipation in AFTA to the implementation o f the USBTA, it has made markets more competitive and has forced domestic enterprises to restructure. But the impending accession to the WTO takes the agenda even farther, because o f the encompassing nature o f the commitments it implies. This is not to defend the accession process, which puts increasingly harder demands on newcomers. But regardless o f the virtues o f the process, WTO accession will entail important changes both at the border and behind the border. On the one hand, much the same as participation in AFTA, it will lead to a reduction in trade barriers. In spite o f the considerable progress accomplished so far in removing import quotas and reducing tariffs and subsidies, protection rates remainhighin Vietnam. Further liberalization should lead to increased efficiency. The process needs to be supplemented by substantive improvements intrade logistics. Increasingly congested ports, with relatively expensive services, and a customs department focused on control (legally or otherwise), rather than on trade facilitation, are obstacles to further international integration. On the other hand, much the same as adherence to the USBTA, accession to the WTO i s about institutional changes, from a more level playing field among enterprises, to competition in key services such as banking, to improved sanitary standards to strengthened intellectual property rights. All these changes should boost productivity in the medium to long term, but they will represent an important challenge for Vietnamese enterprises (including SOCBs) inthe medium term. The WTO accession process Vietnam's road to membership in the WTO began in 1995, when a formal request for accession was submitted. In the following year, a comprehensive memorandum on the foreign trade regime was prepared. Regular meetings o f the Working Party started in 1998, but Vietnam only submitted its first market access offer for both goods and services in 2002, to a lukewarm reception. In 2003, the highest authorities decided to accelerate the process, trying to enter the WTO as early as 2005, if possible. The completion o f China's own accession negotiation, the prospect o f remaining subject to textile quotas at the expiration o f the Multi-Fiber Agreement, the successful experience with export growth under the USBTA, and the difficulty to oppose protectionist measures abroad, may all have increased the attractiveness o fjoining the WTO, thus explaining this renewed determination. After several rounds o f negotiations not resulting in substantive progress, the more ambitious market access offer made by Vietnam at the eighth Working Party meeting, in June 2004, represented a breakthrough. At that point, the 28 negotiating partners agreed to move forward and draft the Working Party Report o f Vietnam's accession, thus paving the way to bilateral negotiations. These negotiations are often subject to criticism, as an acceding country can be requested to meet more stringent requirements than those applying to current member countries (Box 9.1). But Vietnam has shown a remarkable willingness to address the concerns o f the negotiatingpartners. Box 9.1: Incumbentsversus Newcomersinthe WTO Accession to the WTO i s guided by rules that make it differentfiomjoining other multilateral organizations. In addition to a common body of commitments, known as the single undertaking, the aspiring country needsto satisfy all the member countries who may want to ask for further concessions, in return for their support for the application. Without the support of key WTO members, a country cannotjoin. Over time, this approachhas ledto increasingly higher standards for the accession of new entrants. Vietnam, for instance,may needto makemore sweepingchangesthan China. The primary motive for developing countries inseekingto join the WTO i s the boost that they hope membership will give to their exports, thanks to their improved access to international markets. Vietnam expects expanding sales of agricultural and fishery products, and textiles and garments. But WTO membership also makes a country more attractive to foreign investors. And it gives access to the WTO dispute-settlement mechanism, which enforces international trading rules. As a WTO member, Vietnam would also have a say inshapingthose rules. One controversial access of the WTO accession process, in Vietnam's case, concerns agricultural protection. Despite the fact that the majority of population (and a vast majority of the poor) lives in rural areas, WTO members are asking Vietnam to further liberalize its agricultural sector. Vietnam's market access offer to the eighth Working Party meeting set the average agricultural bound tariff at 25 per cent. This is substantially lower than the levels of neighboring countries such as Thailand and the Philippines. Another concern is the "non-market-economy" status that Vietnam may get when it enters the WTO. This status was one of the so-called "WTO plus" commitments China had to agree to during its accession negotiations. It implies a special methodology for assessing potential subsidies in anti- dumping cases brought against Chinese companies, which considerably reduces the burden o f proof. China is already the target for one fifth ofthe anti-dumping measuresworldwide. Vietnam may have to accept similar terms. Source: Oxfam International (2004). Vietnam's market access offer includes a commitment to cut tariffs to an average o f about 18 percent, with industrial tariffs o f around 17 percent and agricultural tariffs bound at 25 percent on average. There i s also a commitment to cease export subsidies for coffee on the date o f entry, and for rice, pork and vegetables in three following years. Upon accession, Vietnam will also adopt WTO-compliant rules on intellectual property rights, investment incentives, customs valuation, technical barriers to trade, sanitary and phyto-sanitary (SPS) standards, import licensingprovisions, anti-dumping procedures, and rules of origin. Accession to the WTO has important legal implications as well. About 100 major government documents need to be issued or revised. This has forced the National Assembly to speed up its legislative program, in the hope o f passing as many WTO-related laws as possible. The seventh session, held in May and June 2005, dealt with the amended Civil Code, the amended Commercial Law, the Law on State Auditing, the Law on signing o f Accession to and Implementation o f International Treaties, the amended Customs Law, the amended L a w on Export and Import Duties, the amended Mineral Law and the amended L a w on Complaints and Denunciations. The eighth session, in October and November, debated the L a w on Intellectual Property, the Law on Electronic Transactions, the Unified Enterprise Law, the Common Investment Law, the amended L a w on Value Added Tax, the amended L a w on Special 112 GLOBAL INTEGRATION Consumption Tax. Looking forward, an "omnibus law" may be needed to ensure the prevalence o f WTO commitments over existing legislation incase o f conflict. There is some contradiction between the encompassing nature of the commitments made by Vietnam and the limited awareness of the business community regarding the implications o f WTO accession in particular, and of global integration more generally. The three-round survey of SMEs conducted by ILSSA asked, in its most recent edition, what were the risks and challenges associated with further market liberalization. Responses were analyzed as part o f a project bythe SIDA. A somewhat worrying finding is that only 12 percent of respondents had positive expectations. Limited liability companies andjoint stock enterprises (the most modern corporate governance forms) were more optimistic than sole proprietorship companies and household businesses. However, the number of respondents holding a negative view was small as well. A majority responded that they did not know what further liberalization will mean, or that they did not expect any notable changes. One possible explanation for this finding i s that SMEs have a limited experience o f internationalization: many o f them are not in formal contact with foreign firms and, for a large part, they do not compete with imports either. Among those with some international exposure, expectations were more positive. For instance, 53 percent of exporters expected benefits from further trade liberalization, compared to 12 percent of non-exporters. But for a vast majority o f Vietnamese entrepreneurs, the implications of WTO accession probably remain unclear. Trade distortions Some of the most immediately visible implications o f WTO accession are related to trade liberalization. Vietnam has made commitments concerning both import duties and export subsidies. But predicting the consequencesof those commitments is not straightforward. Trade regimes are a complex web of wedges affectingthe prices o fboth inputsand final products. They clearly cannot protect all sectors at the same time, because they basically redistribute resources across the economy. But understanding which sectors benefit from the trade regime, and which ones are at a disadvantage i s technically challenging. Still, this i s an important first step to assess what the implications of WTO accession could be. The number of products subject to import quotas inVietnam has declined since 1999. By end 2005, there were quotas only on sugar and petroleum. However, seven other agricultural commodities were brought under tariff rate quotas in2003. These are raw milk, condensed milk, eggs, maize, raw tobacco, salt, and cotton. Tariff rate quotas are not quantitative restrictions, but they allow applying a higher duty once imports exceed some threshold. The Ministry o f Trade was given the authority to activate this mechanism if the conditions of domestic production and foreign trade make it necessary. There i s also a list of items for which trade i s currently prohibited. It includes military equipment, toxic chemicals, antiquities, narcotics, firecrackers, poisonous toys, cigarettes, used consumer goods and right-hand driving automobiles. In addition, a considerable number o f products require approval from relevant ministries before they can be traded. Some pharmaceuticals, chemicals, food products, fertilizer and recording and broadcasting equipment are part of the list. As in other countries, the need for approval i sjustified on heath and security grounds. The commitment to comply with WTO rules on technical barriers to trade, SPS 113 standards and import licensing provisions, should introduce more transparency in this area, making it less prone to disguised protectionism or abuse. Tariffs have been declining too (Table 9.1). Their maximum level was brought down from 200 percent in 1997to 120 percent in 2001 and to 113 percent in 2003. By then, less than one hundredthof tariff lines, accounting for around one twentieth o f all import value was subject to tariff rates above 50 percent. High tariff rates remain concentrated in a few areas such as beverages, spirits, tobacco and cigarettes, usedclothes, vehicles and vehicle parts. There has also been a reduction in the dispersion of tariffs, mainly through a compression at the low end o f the distribution. For instance, the number of tariff lines in the range o f 1 to10 percent declined from about one quarter of the total in 1995to 13 percent in2003. Table 9.1: Trade Barriers and the Effective Protection Rate 1 Weighted average 1997 2001 2003 Nominal 1Effective Nominal I Effective Nominal IEffective Agriculture 8.1 7.7 6.3 7.4 11.1 12.5 Mining 9.4 6.1 8.9 16.4 3.6 -0.0 Manufacturing 30.6 121.5 25.3 96.0 29.2 43.9 Total 21.0 72.2 17.9 58.5 18.2 24.9 Simple average 23.3 59.5 20.1 54.1 20.0 26.2 Rate dispersion 133.8 156.0 149.9 172.3 106.5 134.9 Source: Based on Premachandra Athukorala (forthcoming). The nominal rate of protection is estimated based on most-favorednationtariffs. The effective rate of protection measures the gap invalue added computed at domestic prices and at borderprices. It is estimatedusingthe input-output table for the year 2000. Rate dispersion is measured as the ratio o f standard deviationover mean, inpercent. As a result of all these changes, the averagetariff rate of Vietnam is close to 18 percent at present. This is the figure Vietnam committed to in its market access offer to the eighth Working Party meeting. It is comparable to China's average before its accession to the WTO. But it i s considerably higher than the average tariffs of Indonesia, Malaysia and the Philippines. And much higher than China's average tariff after its accession to the WTO. As in China's case, Vietnam's averagetariff can be expected to decline further as a result o f the bilateral negotiations to accedeto the WTO. A reduction in the dispersion of tariffs, like that observed in Vietnam in recent years, bringsdown the overall transfer made to the manufacturing sector. This is becausetariffs on final products (typically the highest) become lower while tariffs on inputs (usually the lowest) become higher. The overall transfer to a sector is measured through the so-called "effective rate" of protection, which indicates how much higher the value added o f an industry i s compared to a situation where internationalprices would prevail. Inthe case of Vietnam, the lower dispersion in tariffs brought about a dramatic decline in the effective protection rate for the manufacturing sector between 2001 and 2003. The reduction in peak rates committed during WTO negotiations should reduce effective protection rates even further. 114 Tariffs had also been used in recent years in ways which were not consistent with WTO principles. At stake were motor vehicles and a wide range of mechanical, electric and electronic products, including color television sets. In all these cases, an implicit subsidy was delivered under the form of a preferential tariff on imported final products and components linked to meetingmandatory local content targets. Arrangements ofthis sort are not allowed under WTO's agreement on trade-related incentives. They were thus terminated for motorbikes in 2003; for cars they will be terminated upon accession. The new CIL includes an article on opening markets and investments related to trade, which i s designedto ensure compatibility with WTO rules. It stipulates that government will not require investors to give priority to the purchase or use of domestic goods and services, or to purchase them from a specific domestic supplier. Firms will not be compelled to export a fixed percentage o f goods or services, and restrictions will not apply to the quantity, value or type of goods and services which may be exported. There will be no requirement to balance import needs with the quantity or value exported. Investors will not be subjected to localization ratios during manufacture o f goods, to supply goods or provide services in a specific region, or to establish their head office at a particular location. More explicit subsidies had proliferated too. An Export Reward Fund was established under MOF in 1999, to assist Vietnamese exporters. It initially focused on agricultural products such as rice, coffee, pork, canned fruits and vegetables. Assistance was provided under the form of interest rate support, direct financial assistance to first-time exporters, and export bonuses. Cash rewards were also paid to exporters provided that they would meet a minimum exports volume or a minimum exports growth rate. More recently, subsidies were extended to other agricultural products, including beef, poultry meat, semi-processed fruits and vegetables, tea, peanuts, pepper and processed cashew nuts. And they also reached some industrial products, such as plastic and mechanical goods, handicrafts, and rattan and bamboo products. Other, more indirect transfers can be counted as subsidies as well. For instance, there i s a sizeable price support for the electricity used in irrigation. And it has been suggested that there are significant subsidiesto several agricultural inputs, including fertilizers and seeds. Given the multiplicity of mechanisms at work, computing the net subsidy on a product- by-product basis is difficult. However, a study conducted by ADB and MOF concluded that only two products, cotton and sugar, were receiving a substantial transfer. Measured in percent o f value added, and using a concept akin to the "effective protection" rate, the implicit subsidy rate was estimated at roughly 45 percent for sugar, and at 7 to 11 percent for cotton. For all other products, the subsidy equivalent was probably lower than 5 percent. For rice and pork, it was insignificant. Working Party members adopted the position that Vietnam would not be entitled to the concessions allowed to the poorest countries, so that subsidies had to be abolished upon accession to the WTO. However, as a developing country, Vietnam can still retain support measures for up to 10 percent of the value of production. In practice, this means that only the subsidy for sugar will needto be reduced. When putting all the pieces o f this complex web of trade distortions together, it appears that protection is highest in Vietnam for industries which are dominated by SOEs, foreign companies, or both. For instance, as of 2003 effective protection rates exceeded 100 percent for liquor, beer and processed rice. And they fell in the range of 50 to 88 for tea, bricks and tiles, home appliances, textiles, clothing, carpets, plastic products, home appliances, motorcycles, bicycles and motor vehicles. At the other end, most of the industries with substantial participation of domestic private enterprises, such as leather, rubber and plastic products and furniture, were operating underrelatively low protection. 115 BUSINESS Admittedly, sectoral analyses ofthis sort needto be interpreted with caution. An obvious question, when looking at these results is: how did Vietnam achieve rapid export growth in sectors where selling domestically allowed 50 to 88 percent more value added than selling abroad? Part of the answer is that effective protection rates refer to averages across sectors. Even within a well-defined activity like clothing, there is great variation across firms in terms o f product variety, quality, scale and technology. For instance, the relatively small firms that produce the type of clothing favored by local consumers may bear little resemblance to the large- scale clothing manufacturers that often use advanced technology to produce a limited variety o f clothing items under contract to buyers inthe world market. International comparisons are difficult as well, becauseof differences inthe treatment o f various elements of the trade regime in each country. However, it appears that the effective rate of protection of Vietnam's manufacturing sector, as of 2003, was above that of other countries in the region, Itwas twice as highas in Indonesia, Malaysia or the Philippines inthe 1990s. It was comparable to that of Thailand, but this was because of the high protection given to automobile industry, which has been scaled down inrecent years. And it was also higher than the effective protection enjoyed by manufacturers inKorea at the early stage of its industrialization. Trade logistics While quotas, tariffs and subsidies can all distort the trade patterns of a country, some o f the most important barriers to global integration are sometimes found elsewhere. Inefficient ports, cumbersome customs procedures, rent extraction from corrupt government officials or powerful trading companies, can all discourage participation in world markets. These aspects of the global integration agenda are only partially covered by WTO negotiations. But the broader trade logistics needto be tackled to make the most o fthe accessionprocess. Consider ports first. Vietnam has seven international seaports, and five special ports through which only oil and coal are shipped. The most important ones are Haiphong inthe north, Danang in the centre and HCMC in the south, with the latter accounting for 70 percent o f all traffic. A significant amount of investment has gone into upgrading these ports over the past decade. As a result, by now they can handle approximately 14 million tons o f freight per year (excluding crude oil), compared to only 4.5 million tons in 1993. Despite these upgrades, however, the main HCMC port i s considered substandard by comparison to regional and international competitors. Its scale could prove insufficient to meet demand by as early as 2006. Most of the other major ports inthe country are congested, poorly maintained and managed, and have inadequate access by water channels and roads. The current state o f the ports results in relatively high unit costs for shipping lines, and reduced competitiveness for Vietnamese enterprises. This is clearly one area where private sector participation is needed, not only to mobilize capital for investment, but also to foster competition and bringunit costs down. Ports inVietnam are managed by several state organizations. The Vietnam National Maritime Group (Vinamarine), under the auspices of MOT, i s responsible for the planning, construction and management of ports in general. It also operates the country's three major ports. Another seven large national ports are operated by Vinaline, a General Corporation reporting directly to the Prime Minister. The country's 60 other ports are operated by local SOEs and various ministries. But tariff levels for port services are determined by MOT and are the same for all ports in the country. Only container charges are set by the individual ports. 116 GLOBAL INTEGRATION The government has already developed a framework for third-party management of ports. And Vinaline already signed, in 2004, a formal lease with Vinamarine for the management of Cailan port in 2004. In principle, this leasing framework could be replicated with private sector participation. But current regulations do not allow it. The list o f so-called strategic sectors currently in force prescribes that the management and maintenance o f large ports is to remain under state ownership. Thus, despite the fact that three private ports have already been developed under greenfield BOT schemes and two others are under construction, the full benefits o f competition in port services are not being reaped. Clearing goods through customs can also add to the cost o f importing or exporting goods, thus operating as another barrier to global integration. An efficient and ethical customs administration can facilitate trade, raise trade-related revenue for the government, generate reliable trade statistics, and protect society from crimes such as counterfeiting, smuggling or narcotics trafficking. For now, however, Vietnamese customs are still considered slow, unresponsive, inconsistent and vulnerable to corruption (Box 9.2). Box 9.2: Customs Reform:Trade, Governanceor Both? Current customs procedures are based on verification before clearance, heavily relying on physical examinations. Audits conducted after the release o f goods are mainly seen as an enforcement measure, not as a compliance management mechanism, and they are not regularly carried out anyway. Progress has been made regarding the valuation o f goods. In the past, to minimize duty evasion, minimumprices were applied. Their removal is a requirement underthe WTO accession process. But alternative mechanisms to verify the correct value o f goods cleared through customs are not in place yet. Information systems are weak too. Well-hnctioning procedures are inplace inseveral offices, but not across the board. Meanwhile, both the ICs o f the World Bank and the Corruption Diagnostic Study of the Internal Affairs Committee o f the Party identify the customs department as one o f the agencies most exposed to graft. The recent passage o f the revised Customs Law by the National Assembly provides a sound legal basis for reform inthis area. The key to success lies on revamping businessprocesses, more than on technology. Verifications prior to clearance and physical examinations favor rent-seeking behavior by the officials in charge. Instead, a risk management approach is needed. Statistical regularities can be used to identify possible cases o f fiaud, whereas randomized procedures can support physical examinations (for instance, regarding which part o f a container should be inspected). In the general case, however, compliance with the law should be assumed. Technology should be put at the service o f the change in business processes. A risk management approach requires reliable databases on previous clearances and audits, so as to infer the statistical regularities associated with fiaud. It also relies on effective inspection equipment, such as X- ray and other scanning hardware, so as to process examinations in a rapid and non-intrusive manner. Matching records on the outcome o f examinations with clearance records, inturn, can help identify and deter misbehavior by customs officers. Appropriate performance indicators will be neededto monitor that customs reform is on track. Statistics on average clearance times at each customs office, on declarations processed per staff, or on revenue collected, should be part o f the monitoring system. User surveys, in the spirit o f report cards, can also provide feedback on the operation o f customs offices around the country. Source: Based on World Bank (2005~). 117 BUSINESS A moot question is whether state-owned trading companies are a serious obstacle to global integration as well. The state monopoly over foreign trade was formally abolished in 1989, but license requirements applied before an enterprise could import or export. Only in 1998 were those requirements removed. Since 2001, private companies as well as SOEs are allowed to export most products without any license. Inprinciple, private enterprises have the same rights as SOEs, but the latter tend to have a closer relationship with the ministry in charge o f the sector, which puts them in a more advantageous position. Several years later, state trading companies still handle roughly half of Vietnam's foreign trade (Table 9.2). They remain dominant for the export of important agricultural commodities, such as rice, coffee, rubber and tea. For instance, state trading companies control almost 100 percent o f the rice export market, whereas SOEs affiliated with Vinacafe control 35 to 40 percent of the export market for coffee. But does implythat they have market power and can extract rents from producers? The answer is not clear. 2000 2001 2002 2003 Numberof enterprises 17,547 20,722 24,794 28,396 State 1,194 1,064 1,047 896 Non-state 16,308 19,593 23,682 27,437 Foreign-invested 45 65 65 63 Total turnover (trillion dong) 345 376 515 585 State 210 214 303 301 Non-state 130 156 205 273 Foreign-invested 4 6 7 11 Source: Basedon data from GSO. There are numerous state trading companies operating in each sector, so that collusion among them would be needed to push producer prices down. The experience elsewhere, from textiles to telecommunications, suggests that Vietnamese SOEs do compete with each other. An indication that state trading companies do not exploit their potential market power comes from their low profitability. Measured in relation to turnover, their profits are below those of private traders, domestic or foreign. In the case of rice, value chain studies indicate that Vietnamese farmers do appropriate a very high share of the world price, which again i s incompatible with substantial rent extraction. Ifanything, the problemwith these state trading companies is reminiscent ofthe conflict of interests plaguing other, non-equitized SOEs. There i s anecdotal evidence that SOEs affiliated with Vinacafe were used to implicitly subsidize farmers when coffee prices were low in world markets. Another state trading company is allegedly content to export at small or even negative margins in order to secure foreign exchange for the government. Non-commercial objectives, determined by the authorities the state trading companies report to, allow reconciling market dominance with low profitability. 118 GLOBALINTEGRATION Behindthe border Some o f the most important changes promptedby WTO accessiondo not actually happen at the border betweenthe acceding country and the rest of the world, but rather behind its border. The national treatment principle is crucial in this respect. It basically means that foreign companies cannot be subject to more stringent rules, or askedto pay higher fees, or charged more for services, than domestic companies. This makes WTO accession a formidable tool to level the playing field. Businesses of all sorts have to be treated equally by the government. And this is what many o f the legislative changes beingfrantically introducedby the National Assembly in its recent sessions are about. One o f the implications o f the national treatment principle i s that government support to domestic enterprises needs to be discontinued. In Vietnam, this support i s channeled by DAF, underthe form of subsidized loans for investment,interest rate refundson commercial borrowing, and investment credit guarantees. DAF targets disadvantaged and priority areas with its assistance. It also funds an export credit program for "priority industries and products", which includes fishery products, sugar, and mechanical products. To the extent that only domestic enterprises are eligible, DAF support i s not compatible with WTO rules. DAF will thus need to betransformed into a WTO-compliant policy lendinginstitutions; for instance underthe form o f a provider of credit to exporters and importers, regardless of their nationality (in the spirit o f the "eximbanks" o f many other countries). It could be argued that WTO accession then amounts to preventing any "industrial policy", while this was one of the main levers usedby richer East Asian countries to support rapid economic growth. And there i s some truth to that argument. However, not all industrial policies are barred by WTO rules. Investment incentives are allowed provided that they are not linked to export or import subsidies, or to local-content schemes. So-called "green box" measures are allowed as well, without any limit. They include budgetary support for infrastructure services, regional assistance programs and extension and advisory services. As part of the Doha Ministerial Conference o f 2001, WTO members agreed that developing countries could use subsidies to achieve goals such as regional growth, technology research and development funding, production diversification and development and implementation of environmentally sustainable methods of production. Thus, for instance, investinginresearch, and in fostering linksbetween universities and enterprises, would qualify as a WTO-compliant industrial policy. Given the quality of human resources inVietnam, this could be an alternative worth exploring. On the other hand, the experience with the kind o f direct support to enterprises that has been available inVietnam i s not that encouraging (Box 9.3). Another important way in which WTO accession affects the workings o f the economy behindits borders is by fostering competition in services. In this case, reform is about ensuring that all players have access to the market under the same conditions. This aspect o f WTO accession has important implications intelecommunications and inbanking. Inthese two sectors, the USBTA gives some indication of what can be expected from the WTO accession process. Intelecommunications, under the USBTA American companies will be able to enter the mobile and internet market by the end o f 2005 and the fixed-line services market by the end of 2007. U S ownership inthese two services is capped at 49 percent and 45 percent respectively. 119 Box 9.3: Government Assistance and Enterprise Survival Data fkom the three-round survey o f SMEs conducted by ILSSA can be used to assess the impact o f assistance during start-up on subsequent enterprise growth. Based on the interviews, assistance to SMEs can take several forms. The government can give advice or recommendations on the line o f business to adopt, it can help in completing applications and registration procedures, it can provide subsidized credit, and it can allow tax exemptions. The analysis compared the performance o f SMEs across the three rounds o fthe survey, depending on whether they had benefited from any o f these forms o f assistance. Performance was measured through survival and growth rates between two consecutive rounds o f the survey. Statistical tools were used to estimate how different variables, including government assistance, affected the probability o f survival, or the rate o f growth. The variables considered in the analysis included the individual characteristics o f the entrepreneur, such as age, gender and educational attainment. Among other things, the study found that smaller enterprises tend to grow faster, which is a common result around the world. When it comes to government assistance, the study found that government credit was indeed an important determinant o f enterprise growth. However, a closer look revealed important differences over time. Government credit was positively relatedto enterprise growth inthe 1990s, but there was no significant impact o f this kind o f government support on the growth rates o f new enterprises interviewed inthe 2002 round o f the survey. The contrast between the two sub-periods probably says more about the economic conditions prevailing at the time than about the effectiveness o f government support. The 1990s were characterized by considerably more turbulence. Survival rates between the first and the second rounds o f the survey are much lower than between the second and the third one. Large changes in enterprises size (either considerable growth or decline) are more common in the early years as well. It could then be that subsidized government credit really makes a difference in turbulent times, but is unlikely to produce "champions" innormal times. Source: Based on Henrik Hansen and others (2005). Considerable competition has already been introduced in the telecommunications sector of Vietnam, and teledensity has been growing faster than in almost any other country in the world. WTO accession could further increase efficiency, but radical changes should not be expected. Things are different in banking, where important restrictions still apply on the type o f business foreign banks are allowed to conduct, which might be one o f the reasons why SOCBs remain dominant (Box 9.4). SPS standards are another important area affected by WTO accession, despite being "behind the border". In this case, it is the entire farm-to-table chain that is affected, with implications for both economic efficiency and public health. 120 Box 9.4: Will CustomersAbandon Domestic Banks? A survey o f 335 individuals and 60 enterprises was conductedto predict customer reactions to banking liberalization. Almost half o f the individuals had incomes between 2 and 5 millions dong per month; one third was below 2 million. Their level o f financial sophistication varied considerably. Savings deposits were the most popular service they used, followed by ATMs, money transfers, and loans. The use o f securities, e-banking, and consulting was limited among individuals. As for enterprises, 42 percent were domestic private f m s , 36 percent were SOEs and the rest were foreign companies. About a quarter o f the enterprises made more than 75 percent o f their sales abroad, but 46 percent exported very little, or nothing at all. Among this sample o f corporate customers, money transfers and payment services were the most widely used financial services. One quarter o f the f m s inthe sample hadoutstanding loans from banks at the time ofthe interview. Participants in the survey were asked about their preferences in using banking services, and their reaction to the expanded choices that would open to them as a result of banking liberalization. The questionnaire was designedunderthe assumption that foreign entry would be allowed on a national treatment basis. The survey revealed striking differences between customers' intentions and bankers' perceptions. Most bankers still believe that their depositors will not migrate to foreign banks, because o f stronger trust in Vietnamese institutions and a tighter cultural link to them. But based on the responses to the survey, customers could prove them wrong, as roughly half o f them are willing to switch allegiances and migrate to foreign banks. Simpler procedures were the most importantjustification for the declared intention to switch in the case o f corporate customers; professionalism was cited by individual customers. Interest rate considerations only ranked third inimportance for both groups o f respondents. Trustworthiness didnot appear to matter much for any o f the two groups. I Enterprises I Switch I Donot I Enterprises I Switch I Donot I II Loans in Loansindong 43 57 Depositsindong 47 53 dollars II 44 II 56 I1Deposits indollars I1 58 II 42 I1 Individuals Switch Do not Individuals Switch D o not Loans indong 47 53 Deposits in dong 52 48 Loans indollars 41 59 Deposits indollars 57 43 The timing is actually right t o improve SPS standards, as Vietnam's agricultural sector is undergoing major structural changes. Production and trade o f high-value commodities such as fruit, vegetables, meat, fish and seafood products has grown considerably over the last decade, whereas that of traditional bulk products, such as rice, rubber and sugar cane, has stabilized. However, in the case of fruit, vegetables and meat, growth remains almost entirely driven by domestic demand. International market access is hampered by Vietnam's difficulty in complying 121 with quality and health standards. High levels o f food-borne pathogens and toxic residues, as well as the presence o f plant pests and animal diseases, are amongthe most common problems. An action plan has been preparedto make progress on this front. It includes measures for disease and pest exclusion, leading to investments in quarantine facilities. It also includes the development o f active surveillance systems and early response measures. It requires an improvement in diagnostic capacity and an upgrading o f the laboratory infrastructure. And it involves the expansion o f certification mechanisms. One of the advantages of WTO accession is forcing an effective coordination between all these activities. Last but not least, WTO accession should also lead to a stronger enforcement o f intellectual property rights. This is sometimes considered a mixed blessing from the perspective o f developing countries. For instance, stronger enforcement will no doubt make software use more expensive for domestic SMEs. For now, these enterprises take advantage o f widespread software piracy. Indeed, the international Business Software Association estimates the piracy rate inVietnam to be 92 percent, the highest inthe world. More worryingly, a stronger enforcement o f property rights could reduce access to generic drugs, with adverse consequences for public health. The most obvious example in this respect concerns retroviral drugs for the treatment o f HIV/AIDS. But adverse implications could also be expected in other, less notorious cases. As the bilateral negotiations for WTO accession are coming to an end, it i s hoped that Vietnam will be given a treatment which i s consistent with its developing country status. These legitimate concerns should not overshadow the interest o f Vietnamese enterprises inhavingtheir patents and trademarks protected. This interest is illustratedbythe cases o f Trung Nguyen Coffee and PetroVietnam which, upon attempting to get their trade names and trademarks recognized inthe US, found that they had already been registered by other entities. In the domestic market, the brand names and logos o f L a Vie (drinkingwater) and the Golden Pig (animal feed) have been repeatedly copied and imitated. Even in relation to software, Vietnam could soon be concerned about piracy by others. According to the Vietnam Software Association, there are 1,000 companies which name software development as one o f their businesses. It i s estimated that about 400 o f them are actually engaged inthis activity. Furthermore, the H C M C Association for Information Processing lists 50 software enterprises with at least 30 employees in their payroll and annual revenues exceeding 100,000 dollars. Lower piracy levels are needed for this sector to thrive. As Vietnamese businesses increasingly realize the importance o f trademark protection, the number o f applications for registration is increasing rapidly. In 2004 alone, the National Office for Intellectual Property (NOIP) receivedalmost 15,000 o f them, and granted 7,600. More than two thirds o f the applications were by Vietnamese. 122 I O . If global integration is one of the main drivers of economic reforms in Vietnam, their engine i s clearly homegrown. Since the beginningof the renovation process, almost two decades ago, the government has been steadily pushing for change across all policy areas, with the objective o f building a market economy with a socialist orientation. The five-year plans articulating this vision are also changing along the way. They rely less on quantitative targets for material production, and focus more on policies and development outcomes. Consultation with stakeholders at large i s also becoming part of the process, and business associations are having increasingly more clout. The next five-year cycle offers the prospect for Vietnam to complete its transition, across several dimensions: towards increasingly relying on market mechanisms, towards full membership inthe global economy, and towards middle-income country status. But success in these multiple transitions will also require a policy gear-up. Vietnam needs to complete the first generation o f reforms, the ones that allowed the emergence o f businesses of all sorts and a spectacular reduction in poverty. But it also needs to embark into a second generation o f reforms, adapted to the increasingly complex problems it faces as it prospers. Inthe economic front, it has to improve the efficiency of capital mobilization, by reforming its financial sector and better managing state assets. In the social front, it should lay the foundations o f more modern universal systems, combining market efficiency with broad coverage in education, health and social insurance. In the institutional front, it has to strive for transparency and accountability, further strengthening public financial management, reducing corruption and creating a modern judiciary. Progress across all these fronts would contribute much to business development. Planningand consultation The preparation of the SEDP for the period 2006-2010 provides an opportunity to articulate the domestic reform agenda so that it complements the global integration process. In a way, commitments towards the international community in the context o f WTO accession can be seen as the flip side of commitments to the Vietnamese people through the SEDP. From this perspective, the timing could hardly be better, as the implementation of both WTO obligations and SEDP contents will start more or less simultaneously. At the end o f this five-year period, if it all goes well, there is a reasonable chance that Vietnam will be at the verge o f becoming a middle-income country. Ensuringthat the reforms supported by the SEDP allow getting there i s thus offundamental importance. Consultation is one o f the key mechanisms to improve the relevance and quality of planning documents. And consultation with stakeholders at large, not only across government agencies and with mass organizations, i s becoming more common in Vietnam. As part of the transition to a market economy, there is a move away from setting production and investment targets, towards relying on development outcomes as motivators for public action. In the preparation of the SEDP for 2006-2010, there is also a noticeable openness to seek feedback from communities, non-government organizations and business associations, in a way that is reminiscentof the 2001-2002 consultations to inform the Comprehensive Poverty Reduction and GrowthStrategy (CPRGS). Infact, the CPRGSprincipleshave been officially integrated intothe preparation of the SEDP for 2006-2010. While fully moving away of the traditional command approach to planning will take time, a more strategic and consultative approach is already emerging fromthis integration. Box 10.1: BusinessAssociationsin Vietnam Some o fthe most relevant business associations are: VCCI: Established by the government in 1960, its original objective was to promote economic relations between Vietnam and the rest o f the world. According to its current statutes, it is "an independent, non-governmental organization" operating on the basis o f financial autonomy, but carrying out its activities "with the support and under the supervision o f the Vietnamese state". VCCI has 5,270 members, o f which 1,394 are SOEs. Its officials are public sector employees, and about a fifth o f the hnding is providedby the government. Vietnam Cooperative Alliance (VCA): It is also sponsored by the government, and has its staff appointed from government ranks. VCA has branches in all provinces. Any cooperative can become a member. The Union o f Associations o f Industry and Commerce (UAIC): This is the heir o f business organizations which existed in the south prior to reunification, such as the General Association o f Industrialists o f the former Saigon. In 1989, with the encouragement o f the Fatherland Front and the permission of the provincial government, the Association of Industrialists of HCMC was established by a group o f business people with a trajectory before 1975. This was the first business association to include members from the private sector. It was later reorganized into a series o f sectoral organization; there are 12 o f them at present. Originally the UAIC membership consisted o f private enterprises only, but in recent years the association has opened its doors to SOEs. Currently, the UAIC has approximately 1,800 members. Young Business Associations and Clubs: They are active in many provinces. The majority o f their members are private companies. The first one was established by the HCMC Youth Union; others started very informally. The Young Entrepreneur'sClub in Hanoi, for example, was launched by about ten o f the first private company directors o f Vietnam, inthe early 1990s. The group began to meet mainly to network and socialize. Gradually, it became a forum in which members could discuss experiences and make businessconnections. There are also a number o f sectoral associations, whose members are predominantly from the state sector. The Vietnam Leather and Footwear Association (LEFASO), for example, was created out o f an earlier production alliance by SOEs, whereas the Vietnam Textiles Association (VITAS) emerged from a GC. In addition, there are about 20 associations involving foreign companies, such as the American, Australian, British and European Chambers o f Commerce, which meet regularly to advance the interests o f their constituencies. Source: Based on Trang Quynh PhuongNguyen and Jonathan A. Stromseth (2002) and ADB (2005d). However, business associations are still weak in Vietnam. The current legal framework allows entrepreneurs to organize on their own, but the formation o f an association still requires government approval. A study by MPDF, which included a survey o f 414 enterprises, revealed 124 DOMESTIC REFORMS that only half o f them were formal members of a business club or association. Participation was more frequent among larger enterprises, among those in the manufacturing sector, and among those exporting some of their production. Moreover, the study revealed that only a few o f the many business associations registered were actually providing any services to their members. The diversity o ftheir nature and membershipbecame apparent as well (Box 10.1). The active business associations active are performing fairly well at providing information to their members on policies and legal issues at delivering or organizing short-term training courses. On the other hand, their provision of business services tends to be weak. And a majority o f the associations do not consider themselves capable o f engaging in a policy dialogue with the government on behalfoftheir members. At the risk of simplifying,the dialogue betweenthe business community and government agencies takes place at three levels. The highest one brings together representatives from a number of business associations and a range of line ministries and government agencies. It is organized twice a year, under the name o f Vietnam Business Forum (VBF). These regular gatherings have led to the identification of a series of policy priorities for business development. Progress in addressing these priorities i s monitored and evaluated at the bi-annual VBF meetings. Participation by the highest national authorities allows the business community to convey its views and concerns inan effective manner. A second level of policy dialogue is between specific ministries or agencies and the relevant stakeholders inthe business community. These exchangesoccur mainly inthe process o f drafting new laws or decrees, and are becoming more regular over time. One o f their venues i s the Monthly Business Forum organized in Hanoi and HCMC by VCCI and the Asia Foundation. The third level involves provincial authorities and the corresponding business community. These exchanges are most generally aimed at improving the local investment environment and attracting investors to the province. Meanwhile, the foreign business community has established working groups focusing on specific issues, such as banking, manufacturing, distribution and infrastructure. These groups meet with varying frequency throughout the year and report on their activities and concerns to the VBF. Overall, it is fair to say that foreign business associations tend to have better access to policy makers than domestic ones. Among the latter, participation inthe policy dialogue tends to be limited to large organizations such as VCCI and the UAICs from Hanoi and HCMC. These groups often contribute comments and opinions to the drafting committees in charge o f formulating new laws and regulations. In the case of VCCI, they occasionally serve on the drafting committees themselves. Sometimes this participation has had a tangible positive effect on policies affecting the private sector. But a stronger voice from representatives o f SMEs (and even household enterprises) would be welcome, as it would ensure a better balance between the views of businesses of all sorts. Structuralreforms The structural reform agenda is broad, as it traditionally includes objectives such as trade liberalization, SOE restructuring, financial sector modernization and private sector development. In a transition economy like Vietnam, much needs to be done in each of these areas, even after almost two decades into the renovation process. Yet, at this stage the most crucial components o f the structural reform agenda relate to capital mobilization. The reform of the financial sector needs to be completed, so that it can swiftly channel resources to the most productive businesses. 125 And the management o f state assets needs to strengthened, so that whichever capital is in the hands o f government i s used in an efficient and transparent manner. M a n y o f the technical aspects o f banking reform are already in place, from an improved payments system to modern credit manuals. But in spite o f these efforts, SOCBs remain weak. The share of NPLs in their total credit does not show any clear sign o f abating; rather the contrary. A s a result, the contingent liability for the government could by now be as high as 8 percent o f GDP, if not more. At a superficial level, bad SOCB portfolios result from still weak operational, credit appraisal and risk management procedures. At a more fundamental level, the problem i s that SOCBs remain too exposed to implicit pressures for non-commercial or politically-preferred lending, especially in their provincial branches. Supporting SOEs, helping farmers when commodity prices are low, and financing infrastructure projects for which no budget allocation i s available, are among the most common sources o f NPLs. Clearly, those who make lending decisions o f this sort are not trying to maximize returns, but rather to do "the right thing", as defined by the relevant authorities. However, at a time when the investment needs o f Vietnam are enormous, allocating the available capital to the projects with the highest returns would be a preferable choice. In the absence o f determined banking reform, the situation o f SOCBs is bound to deteriorate, thus increasing the burden from contingent liabilities. Accession to the WTO implies that foreign banks will be able to bringtheir sophisticated technology and streamlined procedures to bear on Vietnam. This will surely benefit the economy as a whole; but it just as surely represents a threat to the SOCBs. The most profitable and fonvard-looking enterprises, especially those involved in international trade, will be tempted to switch to foreign banks. Skilled and promising SOCB staff will be attracted to work for them, as they offer more interesting challenges and more attractive remuneration packages. If SOCBs retain their existing goals and incentive structures, they will not be able to compete successfully with the newcomers. The transparent classification o f loans based on IAS, as required by Decision 493, i s an encouraging first step towards benchmarking banks' performance and setting milestones for the gradual resolution o f NPLs. But transparency alone will not be enough. The flow o f new NPLs can only be contained ifthe management o f the SOCBs is heldaccountable for their profitability. Equitization i s another potentially important step. However, selling equity shares to the public at large will not be enough either. Bringing in strategic investors would to more to strengthen the profit orientation o f SOCBs, and keep lending at local levels under control. A key complementto reforming the SOCBs is to re-positionthe SBV as a modern central bank, in charge o f supervision and monetary policy. Ensuring financial stability in an increasingly open environment will require that the supervision function o f the SBV be considerably upgraded. Not by checking individual transactions, with overly detailed legal requirements, but rather by appraising the banks' own risk management and strategies. Also, the recent inflationary shocks experienced by Vietnam show how important it i s to develop the capacity o f SBV as the leading institution for monetary policy. Fortunately, the government o f Vietnam i s fully aware o f these problems, and the recently adopted roadmap for banking reform contains the right ingredients to tackle them (Box 10.2). Its implementationwill be challenging, however. Banking reform should be complemented by the strengthening of capital markets, as this i s the other main vehicle to channel resources to the most productive businesses. Inthis case, the problems are more technical in nature. The rigid rules for the conduct o f business applying to the official stock exchange ensure that trade fails do not occur, but they adversely affect turnover efficiency and trading volumes. At the other end, there are neither rules nor information systems for the informal market, which raises the prospect o f abuse o f investors' trust. 126 Box 10.2: A Roadmapfor BankingReform Inearly 2005 the SBV preparedan ambitious submission on bankingreform, as an input for the SEDP for 2006-2010. While this submission was based on a close consultation with the Prime Minister's office, it was considered a departure fiom the bankingreform strategy followed since 2001, thus warranting consideration by the top leadership of the country. The terms o f the submission were endorsed, indeedstrengthened, inthe summer of 2005. This ledto the preparation of a decisionby the PrimeMinister, giving the overallsense of directionfor the new strategy. It will take some time for the various building blocks of this new strategy to be spelled out in detail and to be translated into concrete policy actions. But the current roadmap for banking reform appears to revolvearoundfour maintasks: Revision of the laws on SBV and credit institutions: These laws will be drafted in2006 and issued in2007. They shouldtransformthe SBV into an independentcentralbank, andbe consistent with WTO commitments and other internationalbest-practice principles. Supervisionwill be the center of SBV's activities, with first the creationof a supervisionboardand subsequently its transformation into an independent body within SBV. There will also be a strengthening of the monetary policy functions ofthe SBV. Separation of SBV from SOCBs: Any decisions related to the management of SOCBs will be moved out o f SBV. The shareholding rights of the state will be exercised out of SOCB, either by MOF or by a specialized agency. To reinforce the separation between SBV and SOCB at local levels, SBV branches will be re-organizedon a regionalbasis. Regionalbrancheswill have similar responsibilities to those of the central office, except for monetary policy. The functions and responsibilitiesof existing, provincial brancheswill be gradually curtailed. Speedingup the SOCB equitizationprocess: VCB and MekongHousingBank shouldbe equitized in 2006. The Bank for Investmentand Developmentof Vietnam (BIDV) should follow in 2007, and the Industrial and Commercial Bank (Incombank) in 2008-2009. By 2010, there should be progress towards the equitization of the VBARD. Following equitization, these banks will issue shares with the intention o f bringing state ownership down, to as low as WTO commitments establish. Upgradinghumanresources and I T technology: this will also require adapting the overall business process, focusing on banking supervision and monetary policy. Efforts will go into developing supervision manuals and training programs for supervision staff which are adapted to the characteristics of the banking business inVietnam, including the nature of the information systems ofjoint stock banks. A salary reform will aim at making SBV competitivewith them. Source: Basedon informationfrom SBV. Both an active official stock market and a vibrant market of unlisted securities are needed inorder to support business development. Inthe official stock market, the rigid rules governing transactions need to be replaced by reliable information systems, on both securities and brokers. As for the informal market, securities listing and registration should be dissociated, so as to increase the transparency o f transactions occurring out of the official stock exchange. Moreover, current systems do not provide information on brokers and dealers who are not members of the official stock exchange. It is thus difficult to holdthem accountable incase o f misconduct. Efficiency is also needed in the allocation o f state capital. Some of this capital has been divested through the equitization program. Equitization has also resulted in better enterprise performance, by reducing the power of the line ministries and provinces the SOEs are affiliated 127 BUSINESS with. By the time the equitization process is completed, some 4,000 SOEs will have gone through this kind of transformation. The newly approved Common Enterprise Law is bound to scale up this approach, by giving a maximum of four years to all remaining SOEs to move under its corporate governance rules. Yet, even ifthe equitizationprocesscould be completed on schedule, which is unlikelyto happen, there would still be some 1,800 SOEs in Vietnam by 2007. Because many of these enterprises are quite large, their combined capital i s considerable. A crude estimate suggests that their equity could be close to 200 trillion dong, or 25 percent of Vietnam's GDP. Their total capital, partly mobilized through long-term debt, could be twice as high. Proper management o f the state shareholder rights on this capital is fundamental to increasethe overall efficiency of the economy and to ensure transparency. Unfortunately, this i s easier said than done. There are not too many "models" where the state has retained control over a large portion of the capital of an economy, and succeeded at managing it efficiently. Arguably, Singapore is one of them, and this might be the reason why Vietnam seems keen to learn from the experience of Temasek (Singapore's agency for the management of state assets). For now, there are discussions about fully divesting many more small SOEs, so as to retain state ownership in "only" 1,000 enterprises. These would be organized around five to seven economic groups, operating in a narrow set o f strategic sectors from electricity to air transport. It would also seem that a multiple ownership structure would be adopted, so that none of those economic groups would be purely owned by the state. Further evidence o f the move towards private-style management i s provided by the recent decision to allow five GCs to hire outsiders as chief executive officers (CEOs). Up to now, senior management of state-owned corporations had been appointed by government. Vinashin, the biggest shipbuilder in Vietnam and a very dynamic enterprise, i s the first state corporation inthe country to hire a foreign CEO. This move could be followed by about 100 other entities with state ownership inthe coming times. More important than the precise number of enterprises to be retained, or their sectors of activity, i s the institutional architecture to support an efficient and transparent management o f state capital. The lack of planning documents articulating the government's intentions inthis area makes it difficult to predict how this institutional architecture will look like. However, it will most probably revolve around three components. The first one is a management unit making the key decisions on capital allocation. This role should be played by the State Capital Investment Corporation (SCIC), created by the Prime Ministerin July 2005. The SCIC will absorb state ownership rights inequitized SOEs from line ministries and government agencies, thus consolidating all shareholdings in one institution. By moving ownership rights out of ministries, it should alleviate the conflict o f interest between ownership and regulation. Finally, it will also have a charter capital to allocate selectively across its affiliated enterprises; for now this charter capital has been set at five trillion dong. Whether the SCIC would also manage the state ownership rights on SOCBs after their equitization will only become clear after the details o f the banking reform roadmap are spelled out. The second component of this institutional architecture is a performance evaluation mechanism, to determine which enterprises deserve support, and which ones need to be restructured. Initially, this role could be played by Decision 271 o f 2003, which leads to the classification of SOE performance in one of three categories, on an annual basis. Arrears in the servicing o f debts automatically put the enterprise in the lowest performance category. Two consecutive years of poor performance can lead to the restructuringo f the enterprise and the removal of its director. For now, the implementation of Decision 271 by MOF has been slow, but 128 DOMESTIC REFORMS it is expected to accelerate inthe coming months. Looking forward, the planned listing of many equitized SOEs inthe stock market will provide another mechanismto assess their performance. The third component is a restructuringunit, in charge of handling the assets and debts of poor performers. The Debt and Asset Trading Corporation (DATC), created on a pilot basis in 2003, under MOF too, should play this role. DATC has spent most of the time since its inception dealing with organizational, procedural and legal issues. Operations started in earnest in late 2005, when one SOCB signed a factoring contract with the DATC in an attempt to speed up the collecting o f unpaid debts with two debtors. On the surface, DATC could face the same difficulties as the asset management companies o f the SOCBs. However, by beingdirectly under MOF it might bebetter positionedto reach foreclosure and seize assets ifneeded. The recent handling of Pacific Airlines by MOF i s suggestive of how this emerging institutional architecture could be operating in the future (Box 10.3). It is also telling regarding its associated governance risks, if insufficient attention is paid to the transparency o f SCIC's dealings and its overall accountability as a government agency. Box 10.3: FosteringCompetitionin Air Transport? Pacific Airlines was formed in 1992 as ajoint stock company between state-owned Vietnam Airlines and other SOEs, includingSaigon Tourist, VASCO, SASCVO and Airimex. Vietnam Airlines held 86 percent o fthe capital. With a staff of 561 employees and four leasedplanes, Pacific Airlines offers similar standards of service to Vietnam Airlines, charges similar prices, operates from the same airports and owns no routes. In practice, the two companies complement each other. But Pacific Airlines is nonetheless unprofitable; in 2003 its losses reached 19 billion dong. This made it an exception amongjoint stock companies with a majority state share. Poor load factors, too many passengers not paying fares, and inadequate aircraft utilization could be among the reasons underlying this poor performance. At the end of 2003 the capital of Pacific Airlines totaled 230 billion dong, and its liabilities 213 billion. By 2004 ithadpresumablybecome insolvent, andwas not payingits maintenancebills anymore. In April 2005, MOF reclaimed all the shares Vietnam Airlines had in Pacific Airlines, and hired Temasek to diagnose the company and propose a restructuring plan. The assessment was conducted in four months, and subsequently Temasek (which is the owner of Singapore Airlines) agreed to buy 30 percent of the shares of Pacific Airlines, for 50 million dollars. MOF retained the unsold shares. After the deal was signed, MOT announcedthat it would not allow athird carrier inthe domestic market. However, during 2005 two regional low-cost carriers were authorized to operate flights i?om Vietnam to Bangkok and Singapore. Source: Basedon information fiom MOF. Social inclusion Vietnam managed to build inclusive basic health and education systems in the period under central planning, reaching a vast majority o fthe population and delivering social indicators comparable to those o f middle-income countries. But these systems did not cope well with the transition to a market economy, and they were anyway ill-suitedto addressthe more complex and expensive needs of an increasingly more prosperous population. The provision of services for a 129 BUSINESS fee allowed coping with the demands of the most affluent, but ended up penalizing the poor. Meanwhile, Vietnam had not managed to build an equally inclusive social protection system. Assistance to the poor was targeted and delivered quite effectively at local levels, but more modern social insurance mechanisms were only available to those working in government or SOEs or, more recently, inthe formal private sector. The challenge for Vietnam now is to combine the efficiency of market mechanisms with the broad coverage that characterized education and health services in the period under central planning. Success in this combination could lay the foundations of modern universal systems. In a paradoxical way, the relatively good quality o f targeting mechanisms makes it possible to use them as a tool to attain universal coverage. Not by providing transfers and subsidized credit, as they still do to some extent, but rather by giving the poor access to services which are otherwise delivered for a fee. Targeting and "socialization" policies can then go hand inhand. But proper attention to incentives i s neededfor this combination to work. On the surface, education is the easiest case. Targeted programs have been used for quite some time to deliver exemptions o f education fees to the poor. Exemptions allow all children, regardless o f family income, to attend the same schools. Empirical assessments o f the performance o f these exemptions show, consistently, that they are effective at increasing school enrollment among the poor. Inprinciple, differentiation inthe nature of the services accessible to households only occurs at higher education levels. For instance, private providers are by now authorized in tertiary education, and public universities can increasingly charge fees for the coursesthey offer. However, this does not mean that universalism has already been reached in basic education. There is still a fraction of children who do not attend primary school, mainly among the ethnic minorities. Andthe quality ofthe education impartedvaries considerably. The poorest districts and communes have worse schools, resulting inmuch lower achievement in standardized tests. Local fees for school maintenance, and unofficial charges (for instance, under the form of nominally voluntary private lessons after school) also fall on the poor. Enforcement o f quality standards, larger budget allocations for poorer communes and districts, and a shift to full-day education are thus part of the social inclusion agenda inthis area. Inhealth, targeting mechanisms hadtraditionally beenusedto deliver health care "cards" to the poor. Under such scheme, public health facilities were mandated to provide some services to card holders for free; in exchange, they were supposedly refunded by the government. In practice, services were not always delivered, as the health facilities lacked an incentive to do more than the bare minimum for poor, non-paying patients. A breakthrough happened in 2002, with the creation of provincial health care funds for the poor (HCFPs). Targeted mechanisms were to be usedto deliver health insurance cards to the poor, of the same type available to formal sector workers. But despite a substantial allocation o f resources from the central government to HCFPs, most provinces continued reimbursingthe public health facilities directly. However, conditions are now better to aim for universal coverage. Health insurance has been restructured, with the introduction of a voluntary program offering benefits similar to those accessible to formal sector workers. The HCFPs have been reorganized as well, now forcing provinces to provide health insurance cards to the poor, and banningthe direct reimbursement to public health facilities. This way, targeting mechanisms can be used to deliver the same kind o f health insurance cards available under the voluntary program. MOF i s working on the increased budget allocations neededto support the revamped HCFPs. What is pending for health reform to be successful is the strengthening o f incentives for the delivery of quality services at reasonable cost. For now, Vietnam Social Insurance (VSI) simply uses its resources to reimburse public health facilities. It does not yet behave as a 130 strategic, wholesale purchaser o f health care packages from public and private providers. But it is in a good position to do so, as it is bound to manage an increasingly large volume of resources, both from those paying a premiumfor their health insurance, and from the budget, which pays for the premium on behalf of the poor. The social inclusion agenda in health is then to effectively combine modern insurance principles, effective targeting mechanisms, and the development o f a market for wholesale health service packages. The biggest challenges are in social insurance, where Vietnam has much less o f a trajectory. Not surprisingly for an agrarian economy, protection against poverty in old age was not delivered through a formal pension system. The latter was only accessible to the minority of the labor force working for government or SOEs, and it did not involve any contribution by the beneficiaries. In 1995, a social insurance system was created, with its coverage extending to formal sector workers in the private sector. Workers and employers had then to pay social insurance contributions, to defray the cost o f their future benefits. But the resulting compulsory program had serious weaknesses that need to be addressed. In 2005, the government identified the key priorities for its reform, and the first ever Social Insurance Law should be issued in2006. The drafting of this law raises difficult technical issues, and its implementation is likely to be demanding as well. The first priority identified by the government is to revamp the compulsory social insurance program for formal sector workers, in a way that makes it financially viable in the long term. As currently designed, the program is unlikely to generate a surplus in excess o f 1.5 percent of GDP in any year, and will start runninga deficit inabout three decades. This i s despite a young and rapidly formalizing labor force, which implies that the number o f contributors will vastly exceed the number o f beneficiaries for quite a long time. To make the compulsory program financially viable, the formulas settingthe level o f pensions need to be much closer to an actuarial benchmark. The latter indicates the income flow that could be derived from past contributions, had they been invested in a sound manner. Currently, different pension formulas apply to men and women, and especially to public and private sector workers. And they are all quite far removedfrom the actuarialbenchmark. A second, closely related priority is to extend the coverage o f the compulsory program, But for this to happen, it is necessary to modify the incentives for participation by private sector workers. Under its current design, the program discourages long periods o f contribution, or the full reporting of earnings after some number of years. This is again due to inadequate benefit formulas. Past some point, the increase inthe pension level i s simply not commensurate with the contribution effort (Box 10.4). Lack of attention to incentives will leadto insufficient revenue for the social insurance system and, eventually, to an increase in the contribution rate in order to cover the deficit. This would in turn discourage employers from declaring their workers, further undermining the chances o f extending the coverage of the system. A third priority identified by the government is to set up a voluntary social insurance program, available to farmers and the self-employed, so that those inthe informal sector can have access to a basic set o f benefits as well. The reform approach adopted in the health sector could serve as a model inthis respect. In the health sector there are not different programs, but rather a single program (health insurance) with three access modalities: compulsory, voluntary or through targeting mechanisms. But such model would be out of reach in social insurance if the compulsory and the voluntary programs were too far apart. The "portability" of past contributions between the compulsory and the voluntary programs, and the similarity o f the benefits those contributions purchase, i s key to support labor mobility (occupational and geographical) in an economy where the informal sector i s likely to remain dominant for quite some time. 131 Box 10.4: Social Insurance: Getting the Incentives Right The generosity o f the benefits offered by a social insurance system is often difficult to gauge. 3nthe surface, it is clearly spelled out ina formula. For instance, under Vietnam's current regulations, :he payable pension increases by two percent o f the average salary for every additional year o f :ontribution in the case o f men; and by three percent inthe case o f women. This i s very generous by international standards. But for private sector workers, the "average" salary i s computed based on all the work history of each individual, without any indexation. For civil servants, only the last five years Eount. Inan economy where real incomes (including wages) canbe expected to grow substantially over the next few decades, this lack of indexation pushes the level of the average salary down, compared to the actual salary at retirement age; especially so for private sector workers. This combination o f nominal generosity and real tightness produces unexpected outcomes. One practical way to see this is to compare the resulting benefit with a hypothetical, actuarial benchmark. The two broken lines inthe figure indicate the level o f old-age pension as a fraction o f the salary at retirement under the current social insurance regulations o f Vietnam. They are computed in the case o f a man, depending on the years o f contribution, and on whether he works for the public sector (solid line) or the private sector (dotted line). They are drawn under the assumption that incomes grow by 6 percent per year in real terms, and the inflation rate i s 3.5 percent. The smooth line, inturn, i s the old-age pension an actuarial system would pay if its real rate of return on past contributions were 4 percent per year. Pension as Percentage o f Salary at RetirementAge 60% , I 40% I _ . . . .. . I . . . . I - . . , I * 15% -* . . - * - 10%- , / _ e . - - - , , , I , I , , , , 1 1 , There are three striking results inthis figure. First,workers lose most o fwhat they contributed ifthey remainenrolledfor lessthan 20 years (the two brokenlines lay clearly belowthe actuarial benchmark). This i s because they get a lump sum payment which is too small to sustain an income flow, even if invested profitably. Second, public sector workers get a much better deal than under an actuarial system (the solid, broken line i s substantially above the actuarial benchmark). And third, it is not inthe interest o f private sector workers to contribute beyond some 22 years. They clearly lose after the 30" year. This is because the additional benefit from an extra year is offset by a decline in the "average" salary it applies to. As the work history on which the average is computed becomes longer, older wages bring the real value o f the average down. Source: Based on Paulette Caste1and Martin Rama (2005). 132 However, the poor are unlikely to enroll in a voluntary social insurance program no matter how vulnerable they are to risks, simply because they cannot afford it. The social inclusion agenda, inthis case, amounts to setting up a mechanism to provide them with some o f the social insurance benefits that will increasingly be available to the population at large. One option inthis respect i s to pay a minimumpension to the poor intheir old age, even ifthey never contributed. Another alternative, more similar in spirit to the HCFP approach, is for the government to make a minimumcontribution to voluntary social insurance on behalf of the poor, on an annual basis, much the same as it does with health insurance cards. The relationship between this social inclusion agenda and business development might look thin at first. But it would be a mistake to downplay the links between the two. There are at least four o f those links. First, businesses should become important suppliers of social services over time. Public funding should not be confused with public provision. Tertiary education, health care services and, maybe in a not-so-distant future, pension programs, can all be supplied by non-state organizations, including private enterprises, cooperatives andNGOs. A second link concerns the social insurance system specifically. If properly reformed, social insurance could accumulate a massive amount o f resources before the active workers of today start retiring. Those resources will need to be invested; if managed in a sound and transparent manner, they could make a major contribution to capital accumulation. Third, and most obviously, good human resources cannot be dissociated from a healthy and educated population. The development of the education system, inparticular, is the key to tap all the talent available ineach cohort of Vietnamese children, and to build up the skills the business community needs. Last but not least, inclusive development i s a major foundation of social stability. And this has been one of the main advantages of Vietnam over its competitors. Modern governance As Vietnam gradually gears up to middle-income status, the governance agenda is bound to become increasingly important. In the early stages of development, the economy could function on relatively rudimentary but well-tested systems. Land transactions took place in the absence of titling, because local authorities "knew" what belonged to whom; the lack o f a legal system was not an obstacle to business transactions, because direct knowledge o f counterparts allowed operating on a trust basis; corruption was rampant, but the amounts at stake were seldom large, simply becausethe population was poor. But reliance on market mechanisms requires that more elaborate transactions be sustained, in interactions often involving anonymous partners and considerable resources. This may not happen without further institutional development, aimed at fostering transparency and accountability. Completing the modernization of public financial management is a central part o f the governance agenda. Much progress has been made regarding public expenditures. The full disclosure ofthe budget, the increasing say of the National Assembly and People's Councils at all levels in allocations, the creation o f a single Treasury account, the introduction o f international accounting standards, the upgrading of the functions and independence o f the State Audit o f Vietnam (SAV) and the adoption of modern information management systems are all steps towards reducing fiduciary risk. The increasingly large number o f donors delivering aid to Vietnam under the form o f general budget support is a testimony o f the progress accomplished in this respect. The agenda emerging from the recent Public Expenditure Review-Integrated Fiduciary Assessment (PER-IFA), conducted by MOF and the World Bank with the participation of several donors, has charted a clear agenda for the next steps of this process. 133 Progress has been much more modest in the management of state assets and liabilities. The need to strengthen asset management was discussed above, when describing the structural reform agenda. But from a governance perspective, the gradual centralization o f government ownership rights in the hands o f the SCIC is a double-edged sword. On the one hand, it could substantially increase efficiency in the use of capital, and be combined with the development o f key markets, from electricity to air transport. On the other hand, it raises the spectrum of non- transparent deals involving colossal sums of money. Strengthening accountability in the management o f state assets shouldthus be considered part of the governance agenda, and not only of the structuralreform agenda. Better control over liabilities is important as well. At present, the government o f Vietnam does not have a centralized information system for public debt. More worryingly, it has limitedclarity on its contingent liabilities, which arejust beginningto be assessed systematically. For now, the most important liabilities are associatedwith the provisioning for bad loans in DAF and in SOCBs, with the need to recapitalize the latter in view o f their equitization, and with the fulfillment o f pension obligations towards civil servants who did not contribute to social insurance. Over time, however, especial attention will need to be paid to provincial finances. In their drive to develop local infrastructure, many local governments are embarking in "creative" deals through their local development investment funds (LDIFs). Mobilizing syndicated loans and SOE resources to create joint ventures with the private sector and implement transport and sanitation projects might be necessary for rapid economic growth. But it could also be a source of important losses for the provincial governments and, eventually, and eventually a burden on the nationalbudget. Figure 10.1:Tax Revenueand the Emergenceof Small andMediumEnterprises I Simulated decline in revenue as size distribution of firms changes i 100 i I 96 ~ 92 I I 84 , I 2003 2005 2007 2009 2011 2013 2015 Source: Own estimates, basedon datafrom GSO and MOF. The distribution of enterprises by size represented in Figure 1.8 i s used for the simulations. The ratio of tax revenue over sales for each size group is assumed constant. Tax revenue includes CIT, VAT, trade taxes, naturalresourcestax and special consumptiontax. 134 The public financial management agenda for the coming years also includes tax administration reform. Integration with the global economy may not reduce government revenue immediately, as the volume o f trade will grow faster than tariffs rates will decline, at least for some time. At present, trade-related revenue represents 12 percent of all government revenue, or slightly more than 3 percent of GDP. But in a few years this share will start declining, and the shortfall will needto be replaced by revenue from other sources. Equally important, the high "productivity" of taxes such as CIT, and especially VAT, i s to a large extent due to the unusual distribution by size of Vietnamese businesses. Taxing a few large enterprises, be them SOEs or foreign companies, is easier than going after a myriad o f small and medium ones. As the "missing middle" of the size distribution gets increasingly filled, revenue from CIT and VAT will be affected. A simple way to visualize this i s to assume that the ratio o f tax revenue to sales remains unchanged for enterprises in each size segment, but the distribution o f enterprises by size changes inthe way the enterprise censuses suggest it will. The result, with unchanged tax capacity collection, would be a decline in total tax revenue by 7 to 9 percent in a decade (Figure 10.1). This simple exercise shows the urgency o f developing a tax administration that can effectively deal with large numberso f SMEs. Finally, there i s the need to make public markets more competitive. Nationwide open competitive biddinginpublic procurement was introduced in 1996. This system was improved in 1999 and further strengthened in 2003, when the use o f standard bidding documents became mandatory. The first of these documents, for the procurement of goods, was issued in late 2005. Another tool for increased transparency, the public procurement bulletin, has been in use since early 2005. Both bidding opportunities and information on the awarding o f large contracts must be reported in it. The development of an e-procurement bulletin is at an advanced stage, and the recently passed Law on Procurement should help consolidate a still fragmented legal framework. However, lack o f competition in public markets i s not always due to regulatory imperfections. The limited number of suppliers, especially for civil works, opens the prospect of collusion among bidders(mainly SOEs). Another important part of the governance agenda, closely related to the previous point, is to first contain and then reduce corruption. Apart from public markets, SOEs, land conversion and some social services are particularly vulnerable to it. The ICs survey by the World Bank suggests that corruption modalities directly affecting businesses are not as burdensome as often claimed, which offers some hope to tackle them. But seemingly minor bribes, from a business perspective, often amount to the equivalent o f a month o f salary. Seen from a household perspective, corruption is not so petty then. Failure to tackle it could undermine public morality, and eventually affect the business community as well. Both the ICs survey and the corruption diagnostic study by the Internal Affairs Committee of the Party also suggest that the extent of corruption varies considerably across government agencies (Figure 10.2). Some o f them appear to be much "cleaner" than i s usually acknowledged. Others clearly require determined government action. But determined i s not synonymous of punitive. While sanctions are a central component of an anti-corruption strategy, they alone will not be enough to overcome corruption inVietnam. An effective anti-corruption strategy will have to combine a few other elements. Simplification of procedures, so as to reduce the discretionary power o f government officials, is one of them. The Public Administration Reform Agenda has measures going in this direction, such as the generalization o f OSSs and the development o f electronic interfaces and other forms of e-government. Itwould also be important, for agencies such as the traffic police, to set fines at levels which are not unreasonably high, so that those breaking the rules have less incentive to pay bribes. The strengthening of government systems is another important element in an anti- 135 BUS!ESS N corruption strategy. This i s already happening in public financial management, and should also happen in government agencies such as the customs department, of the tax administration authority. Monitoring the assets o f civil servants and their immediate families can also help, although the effectiveness of this component very much depends on the way it i s implemented. There is no way for the government to monitor the assets of all civil servants. On the other hand, focusing on senior government officials (and especially on those whose responsibilities make them more vulnerable to corruption) may have higher payoffs. This calls for a phased-in introduction o f this component, with low coverage but close scrutiny in the initial stages. Last but not least, investigative media can help uncover instances of corruption, and act as a deterrent for misbehavior. Civil servants need to be protected against defamation, but journalists need to be protected ifthey do theirjob honestly. Figure 10.2:Corruption byAgency: the Good, the Badandthe Ugly Traffic police Customs department Tax authority Market controller Department of construction Municipal police Land administration agency Import license authority Business registration District Wople's Committee 0% 20% 40% 60% 80% 100% 1w No corruption Isolatedcases Gifts expected faGats required w Widespread Source: Basedon ICs survey by the World Bank. Finally, a functioning legal system supported by a judiciary capable of interpreting and enforcing the law is essential to support business development. In its absence, there i s no assurance that contracts will be honored, which in turn results in high risks and lower competitiveness. Admittedly, Vietnam has done relatively well so far, despite the fact that firms can not effectively solve disputes incourt, and hold business partners or government accountable for their commitments. For instance, in Hanoi there were only 70 cases brought to the Economic Court in 2003, and 20 to the International Arbitration Center operated by VCCI. Firms generally avoid the courts because the process i s lengthy and expensive, decisions are considered arbitrary and enforcement mechanisms are ineffective. But the current institutional arrangements work well mainly for firms at both ends o f the size distribution. Household businesses and small-scale enterprises can rely on personal relationships and trust. At the other end, SOEs and foreign companies (through their business associations and embassies) have privileged access to government, in case of disputes. Butthese options are not available for middle-sized enterprises. 136 11. THE LOCAL DIMENSION The attitude of public officials towards businesses, and how they choose to exercise their official discretion as a result, makes an enormous difference inthe provincial investment climate. The decentralization process and the increased delegation of decision-making power to local authorities have ledto vast gaps inbusiness buoyancy, and more generally ineconomic and social development, across Vietnam. Contiguous provinces, with similar endowments but different commitment o f the local authorities to a multi-stakeholder economy, are often worlds apart in terms o f enterprise registration, investment by foreign companies and the availability o f wage employment. Gaps can be found in a range of apparently small decisions which, compounded, make all the difference. Registering an enterprise, obtaining a license, having access to land, securing a credit, dealingwith tax officials, relying on good infrastructure or being protected from corruption, all these key aspects of the business activity are shaped by the attitude of provincial authorities. Rigorously measuring the quality o f governance (either locally or at the national level) i s not an easy task, and any specific indicator is bound to be questioned on methodological grounds. Identifying which aspects of local governance have a stronger impact on business performance, from access to land to quality of infrastructure to extent o f corruption, i s not easy either. In spite of these caveats, benchmarking provinces on key aspects potentially affecting business activity is a step in the right direction. The modernization o f provincial planning processes, along the lines observed at the national level, relying on consultation with the local business community, should encourage further improvements. Combined with appropriate budget allocations for provinces with stronger needs, the promotion of local leaders who succeed infostering growth and reducingpovertywould provide apowerful incentive for change. Wide provincialvariation Striking contrasts in overall performance can be found between otherwise similar provinces across all of Vietnam, from North to South. It is difficult to account for those contrasts on the basis of history, location, topography, population size, or conditions at the beginningo f the reform process. On these bases, it would have been reasonable to expect a convergence in trajectories. Instead, divergence i s often observed, and it seems to a large extent due to the quality o f provincial leadership. Consider the North first. On the surface, neighboring provinces, H a Tay and Vinh Phuc, have a number o f similarities (Table 11.1). For instance, by 2004 the population o f these two provinces had roughly the same access to roads and electricity, and their poverty rates were almost identical. In some respects, Ha Tay does better than Vinh Phuc. At 51 percent, the fraction of households with access to safe water was much lower inthe latter than inthe former, where itreached 72 percent. Yet, Vinh Phuc has been much more dynamic than Ha Tay in recent years. Based on enterprises censuses, between 2001 and 2003 the profitability o f registered firms was much higher. Also, the share of formal employment relative to population grew twice as fast in Vinh Phuc, and the growth o f gon-state industrial output was much higher as well. The reported similarity in poverty rates as o f 2004 i s in fact a stunning accomplishment: by 2002 the poverty rate of Vinh Phuc was 39 percent, compared to 26 percent for H a Tay. These figures need to be interpreted with caution, as they are computed on samples o f less than a thousand households per province. But the contrast is suggestive o f a substantially different performance between the two provinces. Part o f the gap in performance could be related to the higher levels o f public expenditure enjoyed by Vinh Phuc, although this higher level does not translate into large investment projects. But a large part seems to be due to the quality o f local leadership and its ability to innovate and create a climate for vibrant enterprise development. Table 11.1: Striking Contrasts: from Northto South North Center 1 I South I Vinh Long Binh Indicators Ha Tay Phuc Hue I Danang An Duong GDP per capita (current dollars, as of 2003) 275 365 771 Poverty rate (percent of population, in2004) 16 17 2 Employment inregisteredenterprises (per thousandinhabitants in2003) 25 27 379 ~ ~~ Registeredenterprises inoperation (per million inhabitants in2003) 409 467 750 2202 680 2306 Average capital per enterprise (inbilliondong in2003) 9.0 16.7 7.7 12.1 10.1 27.6 Accumulated FDI (indollars per inhabitant) 117 385 127 364 233 1556 Government promotes private sector (from 1to 10) 1.2 8.1 5.1 7.2 5.9 9.3 Speedto register, get landand licenses (from 1to 10) 4.3 7.5 6.3 8.8 7.2 7.7 Communication on plans andpolicies (from 1to 10) 3.8 5.4 4.5 6.7 3.5 6.1 Source: Based on data from GSO and VNCI (2005b). The last three indicators are compiled out of responses by firms operating ineach ofthe provinces. Inthe central region o f Vietnam, a clear example o f diverging trajectories is provided by the case o f two contiguous provinces: Hue and Danang. Both are on the coast and both have remarkable assets to attract tourism. But Danang does better on most counts. Its poverty rate i s one fifth o f Hue's. Its populationhas much better access to roads, safe water and education. And the ratio o f its employment inregistered enterprises to the total population is four times higher. It could be objected that most o f these differences are due to the fact that Danang is more urban than Hue. A more defensible comparison needs to focus on the growth rates of key indicators, as opposed to their levels. But Danang does better on this count too. Based on data from enterprise censuses, between 2001 and 2003 it added twice as many people (relative to the population) to the ranks o f the formally employed as did Hue. 138 The attitude of local authorities is not as different across these two central provinces as it i s between H a Tay and Vinh Phuc. Admittedly, doing business i s easier in Danang than in Hue. For instance, the time neededto register an enterprise, acquire land and receive licenses (if at all needed) is much shorter. Danang has also being more successful at attracting FDI. By now its accumulated investment by foreign companies is three times higher than in Hue. But this i s not necessarily due to different attitudes of the two provincial governments towards the involvement o f the state ineconomic activity. Infact, the industrial output from provincially managed SOEs i s seven times higher inDanang than inHue. The main difference between the two provinces seems to be related to their investment in infrastructure. In 2003, Hue's spending per capita in large investment projects was about 60 percent below that o f Danang. The result has been a vast disparity betweenthe two provinces inthe availability of landwith proper infrastructure services. In the south, finally, an interesting contrast is offered by Long An and Binh Duong. Strictly speaking, these two provinces are not neighbors, but they do occupy similar positions on either side o f HCMC, Vietnam's largest urban and economic centre. Though both provinces do remarkably well compared with national averages, Long An has a higher proportion o f the population living in poverty than Binh Duong. Part of the gap can be attributed to a much stronger economy in BinhDuong. Its number of registered enterprises, relative to its population, i s three times higher than in Long An. Nearly nine times as many people (again relative to the population) are employed by employed by registered enterprises. Strikingly, Binh Duong has attracted nearly seven times as much foreign investment per capita than has LongAn (and, by the way, almost four times as much as Danang). In this case, the difference cannot be attributed to large-scale public investment. Admittedly, access to roads is poorer inLongAn, due to the seasonalflooding that takes place in some parts o f the province. But in 2003, allocations from the Public Investment Program were almost identical, inper capita terms, inthe two provinces. (Interestingly, they were clearly below those of Danang and Hue at the time). The two southern provinces also have a similar industrial output produced by SOEs. But Binh Duong has a non-state industrial output which i s about 16 times higher than that of LongAn. As in the case of Vinh Phuc and Ha Tay, an important part of the gap between the two provinces seems to be related to the difference in the attitude of provincial governments towards the private sector. Based on data from a provincial competitiveness initiative supported by USAID, enterprises inBinhDuongrate the pro-activity of local leadership at over nine, on a one- to-ten scale. This is by far the highest score in the country. Enterprises in BinhDuong also give higher ratings to their government than those of LongAn on issues such as access to planning and legal documents, communication of new laws and regulations, and predictable implementation of announced policies. Governancequality The contrast between similar and even some times contiguous provinces, in the North, Center and South of Vietnam, is suggestive of the diversity of the mechanisms through which attitudes and decisions by local governments can influence performance. A range o f studies and anecdotal evidence can be usedto shed more light on those mechanisms. What those studies and evidence reveal i s the considerable lee way given by the decentralization process to local authorities in the implementation, and even inthe interpretation, of supposedly common national policies. Opportunities for diversity are further amplified by the inconsistencies and overlaps o f a still emerging legal framework. As a result of such inconsistencies and overlaps, which law or 139 BUSINESS regulation is valid in each particular case, and how it should be applied, i s not always clear. Increased decentralization and still considerable ambiguity make the quality o f provincial governance central to the success o f economic reforms. From a business perspective, differences in the implementation o f the Enterprise L a w warrant close examination. In general, enterprise feedback gathered by the I C s by the World Bank is quite positive about the attitude o f the Business Registration and LicensingAgency under the DPIs. Efforts to facilitate the entry o f new firms, through the simplification o f the administration process, and the enhancement o f capacity at registration offices, seem to be widespread. Yet, a small-scale survey o f enterprises, conducted by the World Bank too, revealed some striking differences between locations. In Hung Yen province, for example, all of the registered companies sampled had succeeded in completing all administrative procedures and were operational, compared to just 44 percent in Bac Ninh and none in H a Tay. The sample o f this survey was too small to make any strong assertion out o f these results, but they do suggest that differences exist in the way provincial DPI offices register and monitor private companies under their remit. A study jointly conducted by CIEM and GTZ yields similar results. Based on the field survey underlying this study, a number o f provinces failed to meet the 15-day deadline to issue business registration certificates. In Hung Yen, 41 percent o f the enterprises surveyed reported receiving their certificates past the official deadline. They were 48 percent in Dak Lak and H a Nam, and 67 percent in H a Tay. This compares with only 26 percent in An Giang, 17 percent in Lao Cai and only 7 percent in Quang Nam. It appears then that despite the simplifications brought about by the Enterprise Law, a number o f provinces are finding it difficult to comply. Even generally successful administrative tools, like the OSS model, can prove to be double-edged swords. The OSS model provides business people with a means to submit and apply for business registration, a tax code and a seal simultaneously, dealing with a single government outfit. The study by C I E M and GTZ found that this "three-in-one" system worked well in Lao Cai. But it had effectively added one more step to business registrationprocedures in Dak Lak, where applicants were forced to use the OSS to simply pass the documents on to the business registrar and back to the business. Licensing is another area deserving attention. Despite the intention o f the Enterprise L a w to ban provincial governments from issuing licenses, some forms o f regulation o f local activity are bound to be required. The example, provided before, o f access by concrete mixer trucks to city streets during day hours i s illustrative inthis respect. Based on a survey finalized in 2002 by VCCI, only 23 percent o f business licenses are imposed by the central government, whereas 55 percent are issued by provincial or district authorities. However, these local licenses can be more or less burdensome depending on the inclination o f provincial and district authorities towards facilitation, as opposed to control, inthe pursuit o f the common good. The very interpretation o f the common good may also vary from place to place, as revealed by business activities which might be considered akin to "social evils". In some provinces, for instance, karaoke bars, accommodation rental, massage services, nightclubs and coffee shops are regarded as vulnerable and, therefore, explicitly or implicitly prohibited from doing business. Local governments also have a central to play in relation to land. In addition to deciding on zoning and land conversion, provincial governments are in charge o f setting the administrative price for land. In recent times, press articles have described a whole series o f scandals involving abuse o f local government authority in the allocation o f land. Irregularities were uncovered in Bac Lieu and Kien Giang provinces, as well as in Do Son town (Haiphong province), Phu Tho town (Phu Tho) and Nha Trang town (Khanh Hoa). But the actual list i s presumably longer. On the other hand, in Nghe An province the allocation o f new residential land to households was 140 LOCALDIMENSION conducted fairly, with priority for vulnerable groups and war veterans' households. Nghe An and QuangNam provinces have also made the most progress interms of recovering land from SFEs. Hueand BinhDinhhave made considerable progress as well. Investment incentives are another area in which local governments have considerable discretion. M P I increased the investment threshold below which investment projects can be approved by Provincial People's Committees from 5 to 20 million dollars. In addition to approving, provincial governments can also add several types of incentives. Industrial zones and clusters have been used for this purpose. But their operation varies considerably across provinces. In some of them, private sector enterprises, including joint ventures, lease the land from the government, develop the relevant infrastructure and then sub-lease to enterprises; an example is Hanh Phuc company, in charge of Duc Hoa 1 Industrial Zone in Long An province. In other cases, non-profit state enterprises have been established to carry out a similar function; IZ Infrastructure Development company, in Phu Tai Industrial Zone, Binh Dinh province is an example o f this arrangement. In many provinces in the North, industrial zones are managed directly by state Management Boards. Provincial authorities frequently use low land rents as an additional incentive to draw private investmentinto their realm. Long An province, for example, exempts all new investment projects from land rental fees throughout the term of their licence. This runs counter to declared government efforts to reduce subsidies within the land market. The time needed to get an industrial land lease matters as well. According to an ADB study, enterprises in Long An province reported the completion of leasing procedures and the issuance o f LUCs within 15 days of approaching the companies in charge of industrial zones. Among the seven provinces surveyed by CIEM and GTZ, the shortest time period for granting or leasing land was in Dak Lak, followed by HungYen, QuangNam, HaNam, Lao Cai, HaTay, and finally An Giang. Interactions between businesses and government officials on issues such as tax payments are especially important, because they are a potential source of corruption. How tax payments are determined could in principle be influenced by enterprise characteristics such as size and turnover, but it should not be influenced by the province the enterprise i s based on. Yet, the ICs by the World Bank reveals considerable differences in approach across regions. In the Mekong River delta about 85 percent of respondents say that government officials accept their tax calculation without requiringany changes. But only 40 percent of respondents in the Red River delta and 37 percent of respondents inthe Southeast say so. More broadly, corruption is by no means uniform throughout Vietnam. Nationwide, there are clear differences across government agencies; in some, it is infrequent or even rare for public officials to expect a payment. But some ofthese differences are exacerbated at provincial levels. For instance, accordingto the ICs ofthe World Bank, it is uncommon for officials to expect or requesta gift when a fm appliesfor a licenseinthe SouthCentral Coast; butthis is not unusual inthe RedRiver Delta. The Mekong Delta region stands out, because entrepreneurs spend less time dealing with government officials, and the official payments they make are only a fraction of what is paid in other regions. Perhaps one of the areas in which provinces differ most i s in their handling of infrastructure development needs. Some o f the most dynamic ones have created LDIFs, to mobilize resources from a variety of sources, get into joint ventures with the private sector, implementinvestment projects and, in some cases, divest the state share in those joint ventures. But a recent analysis of the performance of several LDIFs shows that gaps across provinces can be to a large extent attributedto differences inthe commitment of local governments. Out o f the 13 LDIFs established so far, HIFU leads the field. A second tier, in terms o f performance, includes Binh Duong, DongNai and Hanoi (in alphabetic order). The remaining nine LDIFs are 141 far behind. Provincial commitment will also be fundamental to mitigate the risks created by arrangements o f this sort. Where LDIFs are very effective at raising funds and establishing risk- sharing arrangements with the private sector, their operation raises substantial governance issues. The creation of new enterprises using state capital, the lack of unifiedaccounting and reporting standards, the possible conflict of interest between investment and regulation, and the prospect of contingent liabilities buildingup are among the main risks posed by LDIFs. Whether provincial governments will manage to establish an arms-length relationship with their LDIFs remains to be seen, and the answer, again, i s likely to vary from province to province. Can it be measured? One practical way to monitor what local governments do i s to ask entrepreneurs to indicate what they perceive as severe or major constraints for the development o f their business. This is the approach underlying investment climate assessments. Unfortunately, the ICs for Vietnam, by the World Bank, does not have a sample size allowing a rigorous analysis of constraints at the provincial level. And this i s despite Vietnam's ICs having the fourth largest sample, worldwide. But moving up to one level it is possible to count with a couple o f hundred observations for each of five regions: the Red River Delta, the North Central Coast, the South Central Coast, the Southeast, and the Mekong River Delta. This i s enough to uncover some telling differences in the nature of the constraints faced by enterprises (Figure 11.1). Given that even contiguous provinces within one region can differ considerably in the quality o f their governance, these regional estimates most certainly underestimate the gap between best and worst performers. Based on this crude regional comparison, access to finance and access to land are among the biggest obstacles to business development inthe North of Vietnam. More than 70 percent of respondents in the North Central Coast and almost 50 percent in the Red River Delta complain about the difficulties to get credit; more than 40 percent of respondents in both regions also complain about limited access to land. Except for the Southeast, where land i s an issue, problems appear to be much less severe inthe south of Vietnam. It is interestingto note that accessto land is not at all perceived as a constraint in the Mekong River Delta, which i s also the region where the land market is most developed. On the other hand, the Center seems to do much better than other regions in terms o f corruption. Less than 5 percent of respondents in the North Central and South Central Coasts report it as a major or a severe obstacle to business development, compared to almost 15 percent in the Red River Delta and even more in the Southeast. However, this finding should not be interpretedas a wholesale clearing of the central provinces of Vietnam. For instance, almost 30 percent of respondents in the North Central Coast see customs and trade regulations as severe or major obstaclesto business development. Another approach to benchmark provincial performance i s to rate a range o f government processes which are potentially relevant for business performance. This i s what the recently developed Provincial Competitiveness Index (PCI) does. This index, developed as part of a research project supported by USAID, assesses and ranks provinces by their regulatory environments for private sector development (Box 11.1). Based on its results, Binh Duong and Danang are the best performers across the 42 provinces considered, whereas Ha Tay and Ninh Binhare the worst. 142 Figure 11.1: Binding Constraints acrossRegions 60 50 40 ~ 30 ~ 20 k c e s s to finance Pccess to land '"" 25 20 I 10 l5 10 5 ~~ 0 I Customs and trade regulations Corruption Source: Basedon the ICs surveyby the World Bank. Much the same as with cross-country indicators of competitiveness, its release has attracted an enormous amount of attention by the media, and prompted a variety o f responses by provincial governments. Some provinces doing well in terms of economic growth and poverty reduction have complained about their low ratings in the PCI; others have seized the opportunity to try to improve theirs. While there is room to strengthenthe methodology underlyingthe PCI, the systematic rating o f local government processes and the enthusiastic response by public opinion at large highlight the potential of this kind o f benchmarking as a tool to motivate improvements in governance at the provincial level. The impact of local policies The comparison between similar and sometimes even contiguous provinces in the north, center and south of Vietnam had hinted at the possibility o f different drivers o f success in each case, Economic analysis can certainly help identify areas where better policies and more 143 effective implementation should help businesses thrive. But much the same as inthe assessment of competitiveness rankings across countries, it is important to focus on actual outcomes as well. Entrepreneurial capacity includes the ability to prosper in less-than-perfect regulatory environments, circumventing obstacles that may seem at first insurmountable. Identify which obstacles can be bypassed and which ones can undermine business development may thus help focus the efforts to improve local governance. Box 11.1: The ProvincialCompetitivenessIndex Buildingon previous work by the Asia Foundation and VCCI, the PCI represents an effort to quantify the business environment in which private sector f m s operate. The indicators chosen are selected based on a review o f the literature on competitiveness in general, and o f factors highlighted by researchers and practitioners in Vietnam. Specifically, nine sub-indexes are constructed. They deal with entry costs, access to land, transparency and access to information, time costs and regulatory compliance, informal charges, implementation o f policies, state sector favoritism, pro-activity o f provincial leadership, andprivate sector development policies. Three main sources o f information are usedto compute the sub-indices. The first is a mail-out survey o f 16,200 enterprises in 42 provinces, directly asking entrepreneursto report their perceptions. A second source o f information includes available statistics on issues such as the state share o f output, or the variation inthe number o f provincial SOEs over time. Finally, a third source is made o f a set o f subjective assessments by the research team, for instance regarding the overall quality o f provincial governments' websites. These three sources were combined through a two-stage process. In the first one, more than 40 detailed indicators were constructed; each o f them was standardized to a ten-point scale, and related indicators were averaged to become the nine sub-indexes. In the second stage, different weights were applied to each sub-index in order to construct the overall PCI. The weights were derived from a statistical analysis o f the importance o f the sub-indexes as determinants o f private sector investment per capita. Looking forward, there are two main areas in which this methodology could be improved. One o f them refers to the survey o f perceptions by entrepreneurs. The mail-out approach resulted in a low response rate; at only 13 percent, the risk o f dealing with a non-representative group o frespondents is quite high. Moreover, the sample size is such that barely more than 25 observations are available for some provinces. The other area for improvement concerns the list o f indicators considered. At present, several o f them reflect outcomes, rather than government processes. For instance, the state share o f output depends on how vibrant the private sector is, so that usingit to explain private sector investment per capita involves some circularity. Improvements inthese two areas may lead to some re-ranking o f provinces, but are unlikely to alter much the overall picture emerging from the PCI. If anything, such improvements would make it an even more effective tool to benchmark the policies and attitudes o f provincial government towards businessdevelopment. Source: Based on VNCI (2005). Overall, there is little doubt that indicators such as those considered by the PCI do have an influence on business performance. The higher this index is, the larger is the number of registered enterprises in operation in a province, relative to its population (Figure 11.2). This pattern is not affected if other indicators o f provincial governance are used instead. This is true, inparticular, ifonly a few variables which are totally incontrol o f provincialauthorities are used 144 to measure local governance. A simplified index, involving only a dozen selected indicators out of the more than 40 that underlie the PCI, and averaging without using any weights, yielded basically the same result. The pattern i s such that moving for the bottom to the top o f the local governance league would increase business activity by more than 25 percent. This should be seen as evidence that local governance matters. Figure 11-2: ProvincialGovernance and Enterprise Registration 9 - .-C0m -- Hanoi $ 2 m 9 8 HCMC Danang .fgcu; Binh Duong 8 4 zg $ :'E 6 z Q 5 ' Source: Own estimates based on data from GSO and VNCI (2005b). At the same time, it is clear that the quality of local governance does not explain all the variation in performance across provinces. Some do substantially worse than could be expected, given their attitudes towards business activity, and some do quite better. BinhDuong, Danang, Hanoi and HCMC fall in this latter group. If they were removed from the ana1ysis;the relationshipbetweenthe PCI and business activity would become flatter, and the predicted impact of moving from the bottom to the top of the league would be much attenuated. This calls for a closer look at the role o f specific governance indicators. The research supported by USAID tried to disentangle the contribution o f each of the sub-indexes of the PCI to private sector investment per capita. It found that entry costs and transparency have the highestimpact; on the other hand, access to land and informal charges have a smaller impact. It also hinted at the possible interaction between sub-indexes. For instance, informal charges appear to matter in transparent environments, but not to do too much harm to business development in non-transparent ones. However, these results should be seen as preliminary findings only. Their meaning is somewhat blurred by the classification o f specific indicators under each sub-index. For instance, the sub-index for the allegedly important entry costs includes an indicator for the waiting time for land. It is thus difficult to tell whether access to landhas a highor a low impact on business. Detailed analyses using enterprise data bear more promise. In Vietnam, such analyses have beenconducted on the survival rate of enterprises and on the location o f foreign companies. Enterprise survival and growth rates were examined using the three-round survey o f SMEs 145 conducted by ILSSA, as part o f a research project supported by DANIDA. It appeared that the probability o f survival in Ha Tay and Long An was higher than in HCMC. This result was interpreted as evidence that competition in the former provinces i s less pronounced than in the latter, probably reflecting administrative and structural barriers to entry. Meanwhile, a couple of studies have focused on location decisions by foreign investors. One o f them, slightly dated by now, considers the period 1988-1998. It concludes that infrastructure, the quality o f labor force and the size o f the local market are the most important factors influencing the allocation of FDIflows. Government tax incentives, on the other hand, do not appear to be effective at attractingforeign companies to poor and remote provinces. Another, more recent study in the same spirit focuses on the cumulative number o f FDI projects up to the year 2000, as well as on the number of new FDIprojects inthat particular year. The findings indicate that the availability of land in industrial zones is a highly significant determinant o f location, and actually has the highest explanatory power. On the other hand, SOEs do not significantly inhibitthe inflow of FDI. A possible interpretation is that provincial SOEs may not perceive foreign entrants as a threat to their market share, or might be able to protect their interests in different ways. Where the strength o f SOE does matter i s inthe selection of the FDImodality. Greenfieldinvestments are more common inprovinces with a limited SOE presence, whereas joint ventures dominate when the provincial state sector i s large. Other variables having a significant impact on FDI inflows are the provincial population size, its transport infrastructure, its education level and its GDP growth rate. Incentives for change This review of provincial business performance and its determinants confirms the general perception that the quality of local leadership is decisive. It also shows that local governance may impact business activity through a variety of mechanisms. Measuring the quality o f local governance regarding each of those mechanisms is not an easy task, and identifyingwhich ones have the highest impact on the investment climate in a specific context i s not straightforward either. But the review suggests several ways to support business development, and more generally economic growth and poverty reduction, at provincial levels. A first way to encourage improvements is to systematically benchmark provincial policies, across the broad range o f areas that could affect business decisions. The success o f the PCI in capturingthe attention of the media and local authorities shows the potential impact o f this approach. If anything, the recommendation would be to simplify and strengthen the indicators used in the benchmarking process. Not mixing up indicators o f different nature and not using complex procedures to aggregate them into broader indexes would make the meaning o f the benchmarks even more transparent to the public at large. Using larger samples and securing higher response rates would reduce the risk of biased responses, and bring the benchmarking process closer to more standard report cards approaches. Regardless o f the specifics, this i s a line to pursue, and the remarkable take up of the PCI i s very promising. A second, related change is to work with provincial governments to modernize their planning process, in the spirit of the changes that have occurred at the national level. The modernization of planning involves relying on development objectives as motivators for public actions, while downplaying quantitative targets for material production. It also implies identifying the most appropriate policies to attain the selected development objectives, and it would be surprising if many o f these policies were not related, in one way or another, to the improvement of local governance and the development of a vibrant investmentclimate. Last but 146 LOCALDIMENSION not least, modern planning cannot be dissociated from consultation. Establishing mechanisms to seize the pulse of the provincial business community, much the same as the VBF does at the national level, would be a direct way to get inputs from local business associations and the community more generally. Last but not least, the incentives faced by local authorities should be reconsidered. Laggard provinces, with the highest incidence of poverty, need more resources to make investments in infrastructure and deliver social services. Without such resources, their ability to catch up could be hampered. And current budget allocation mechanisms already do some o f this, by transferring considerable amounts of resources from richer to poorer provinces. But in the absence of other, parallel incentives, just giving more resourcesto laggard provinces would create a perverse incentive for change. An indispensable complement to budget-based redistribution is to hold provincial authorities accountable for nurturing business development, promoting economic growth and achieving poverty reduction. 147 12. SOCIAL IMPACTS Further integrating into the world economy, completing the first generation o f reforms, embracing a second, more ambitious one, and strengthening governance at local levels, are all bound to foster business activity and increase economy efficiency. In a way, these developments hold the key for Vietnam to become a middle-income country. But will there be a social or an environmental price to pay for increased efficiency? Increased reliance on market mechanisms will unleash the potential of the most productive members of society, but it could also harm some of the weakest ones. For instance, in a more developed land market, poor farmers could more easily lose their land in the event of a bad shock. Similarly, in a more open economy, the least competitive sectors could be forced to shed workers in large numbers. And even a perfected market mechanism could be unable to address environmental externalities, so that rapid growth would lead to increasing pollution and degradation. Based on a systematic review o f the available evidence,the foreseeable adverse effects do not appearto be large; except maybe for the environment. This calls for a more systematic use of enforceable environmental standards. But potentially larger social impacts might simply be unforeseeable with the relatively rudimentary analytical tools at hand. It i s therefore important to put in place effective mechanisms to rapidly cope with the unforeseen. Assistance to ailing firms should not be one o f them. Notjust because it is restricted under WTO rules, but also because it could lead to rent seeking and disguised protectionism. Assistance to workers who lose their jobs bears more potential, as shown by the successful safety net for redundant SOE workers. But large-scale mitigation policies will require budget transfers, from the provinces benefiting the most from economic reforms to those adversely hit. Transfers of this sort would not damage efficiency, and could do much to ensure that inthe quest for prosperity no one i s left behind. Landlessness The initial allocation o f agricultural land to rural households, in the late 1980s and early 1990s, was remarkably egalitarian. A privatization process of such a massive scale could have been vulnerable to capture by local authorities, especially given that land was Vietnam's most valuable asset at the time. And yet, there is clear evidence that the annual land allocated by communes was distributed evenly across households. Even in the case o f residential land, allocations were conducted quite fairly, on the basis of need. The priority given to vulnerable groups and war veterans probably helped in this respect. Only in the case o f perennial and forestry landwas the initial allocation unequal. Distributingland on an egalitarian basis is not necessarily good for efficiency. Ifmore of it had gone to the most productive (presumably wealthier) households, the overall productivity of agriculture would have been higher. By one estimate, total household incomes ended up being 5 percent lower than an efficiency-oriented allocation of land would have allowed. But market forces could gradually make the initial egalitarian allocation drift inthe direction o f the efficient one. As land transactions become easier, less productive (presumably poorer) households could gradually be forced to sell some of the land they received, while more productive households SOCIAL !MPACTS would stand a chance to acquire more land. This prospect i s a source of concern, and warrants a careful examination o f landmarket developments. A land market has indeed emerged in Vietnam. Initially, few transactions were registered. An analysis of 34 informal land transfers for ADB concluded that deals were conducted within one or two days, without any transaction costs. The deals were supported through a variety o f mechanisms, including verbal and handwritten contracts, involvement o f witnesses, and informal notification to commune authorities. The reduction o f taxes on LUC transfers, in 2000, prompted a substantial formalization o f transactions. At that time, the tax rate was set at 2 percent for agricultural land, and 4 percent for residential land. The new Land Law of 2003 further promoted market development, by stipulating that allocations of land subject to land-use levies must be conducted through tenders. Evidence also suggests that rental markets have been able to develop efficiently, even without formal registration. In spite of these developments, the total volume of land transactions remains relatively small, Based on household survey data, only 8 percent of households have bought land during the last decade, and a meager 2 percent sold any. Given the limited amount o f land changing hands, there has not been a widening of consumption inequality across households, or a surge in poverty associated with landlessness. Admittedly, the same period, the Gini coefficient for land inequality among rural households rose from 0.49 to 0.64 between 1993 and 2004. But the correlation between rural land inequality and income inequality turns out to be weak inVietnam. However, there could still be localized impacts deservingcloser examination. A household can be considered landless if it has no land, other than residential. The share o f the population living in landless households inrural Vietnam increased from 8 percent in 1993 to 12 percent in 2004. This modest increase hides important regional differences, however. The south is characterized by levels o f landlessness which are higher than the national average, and also increase faster than the average. For instance, between 1993 and 2004, the fraction of landless rural households in the Mekong River Delta rose by 9 percentage points, from 14 to 23 percent. The rise was even steeper inthe South East. Yet landlessness is not necessarily associated with poverty. In Vietnam it is rather the opposite: the poor are the least likely to be landless, and households at middle levels o f consumption are the most likely. This i s because many families moving out o f poverty also moved out o f farming. It i s estimated that roughly three quarters o f the overall increase in landlessnesswas due to falling poverty. Ifanything, the trend is becoming stronger over time. But there are also notable regional differences, with the Mekong River Delta and the Southeast standing out as the exceptions to the pattern elsewhere. There, signs o f polarization are emerging, with higher landlessnessbeingmore common among the poor. The Central Highlands display a different pattern too. This is mainly due to the policies adopted after reunification, to facilitate the massive migration of people of the Kinh ethnic majority into the region and promote the establishment o f SFEs. These policies were successful at raising agricultural output. By now the region has become the main producer of coffee, rubber, cashew, tea and cotton, among others. But in the process, the native hill-tribe people became a minority group. They gradually lost their land and moved deeper into the forest. Land issues had led to social tensions inthe Central Highlands, including demonstrations and riots by the hill-tribe people in2001. Inresponseto these tensions, the government vowed to allocate residential land to all ethnic minority households in the region. This was part o f a broader program to improve their production and living conditions. In principle, every ethnic minority family in the region should be provided with 400 square meters o f land for accommodation, and one hectare of mountain fields or half a hectare o f plain fields for 149 production. The government also pledged to provide 5 million dongs in aid to each family to buildor buytheir houses. So far, slightly less than two thirdsofthe target has been attained. Environmental degradation The ambitious reform agenda of Vietnam could raise two types on environmental issues. First is the emergence of previously hidden environmental liabilities, particularly in the state sector. As SOEs are gradually equitized, restructured, or managed more efficiently, their older ways o f handling pollution and natural resource degradation are bound to become more "visible", and increasingly more unacceptable. Who will bear responsibility to control such pollution and redress the inherited misdoings is an open question. The second important issue i s relatedto the growth pattern of an economy that becomes more globally integrated. The comparative advantages o f Vietnam could favor the growth of sectors which are more intensive in water or toxic pollution. And low environmental standards could also attract foreign investors to those sectors, inthe expectation of operating at lower costs than elsewhere. The sector with potentially largest environmental liabilities is coal. An operational review o f four SOEs affiliated with Vinacoal, by Ernst and Young, is revealing in this respect. Poor standards in areas such as mine water discharge, dust emissions, and waste dump rehabilitation in open-cut mines are common. Two mines discharge water that i s clearly in the acid range. And their waste dump areas, some o f them directly facing Halong bay, have been subject to little or no rehabilitationwork. Their height, drainage control and overall stability are a matter for concern. The quality of dust management is uneven too. Near major buildings, around the coal preparationplants and on major internal roads, the level of housekeeping is good, with these areas generally being clear of dust and rubbish. Trucks and fixed sprays are used to apply water to internal roadways and stockpiles, thus limiting the amount of dust. But a build-up o f fine coal, which can be a significant source of dust, is noticeable around train tracks and roads. And the inside of two preparation plants is very dirty and untidy, with a large build-upo f fines, mud and coal spillages. This may partly be a result o f the age and condition of the equipment. However, there are also indications of good practice. In one mine, a wastewater management system was introduced in 2000, with support from Japan. It includes 16 settling ponds, where water is treated to remove all solids, to be then returnedwith make-up water for re- use in the coal preparation facility. Thanks to this system, water i s now re-circulated in a closed circuit and no longer discharged to sea. As an additional safeguard, the company has prepared a dam to catch drainage water. The dam is cleaned out by dredge every month. The surface drainage system is routinely dredged, and annually cleared out, to ensure sufficient holding capacity inthe rainy season and minimize the risk of slurry being discharged to sea. Environmental monitoring is conducted at 27 nominated sites. Weather, dust levels and noise are recorded at most locations; vibration, water quality and hazardous gases are checked only at selected ones. Monitoring i s conducted four times per year, as a spot check, by a Hanoi- based private company. Current best practice for dust fallout is the continuous collection o f samples, with analysis at least monthly; for water quality, it involves weekly or monthly sampling. Thus, while the selection o f sites and the type of sampling and testing seem appropriate, the frequency of monitoring i s not. As for the environmental impacts o f global integration, and o f economic reforms more broadly, they can be divided into three main components. The first one refers to the way trade 150 patterns affect the overall scale of the economy; the second one, to the way they influence the techniques of production; and the third one, to their impact on the sectoral composition of output. Experience elsewhere suggests that the third effect i s in practice the most important one. Integration into the world economy can substantially modify the specialization pattern o f a country. Specializing in the production o f more pollution-intensive goods leads to dirtier environments. This may be Vietnam's case at present (Box 12.1). Box 12.1: Becominga PollutionHeaven? The sectoral composition of output in Vietnam has change in parallel with the gradual opening up ofthe economy. Manufacturing output from sectors with ahighintensity ofwater pollution has grown faster than the output of less pollution-intensive sectors. There is also evidence indicating that exports from sectors with a high intensity of toxic pollution have increased significantly, and investments by foreign companies have been particularly important in those sectors. This trend is not surprising, as it is not uncommon among developing countries. However, Vietnam is an increasingly large exporter of garments, which have a low intensity of pollution, whereas it imports large volumes o f textiles, which have a high intensity. This makes it anet importer ofpollution-intensive goods. The sectors which have displayed the highest increase in exports are textiles, leather and rubber products. O f these, the textile industry i s also a major consumer of industrial chemicals. Therefore, increasedexports oftextiles have fueled a simultaneous increase inthe production of industrial chemicals. Similarly, export revenues from craft villages have expanded significantly in recent years, and are now close to half a billion dollars. But craft villages are a major source of toxic and water pollution. It i s thus not surprising to note that exports are getting "dirtier" over time. This i s the result of a change in overall trade pattern of Vietnam. The export specialization has drifted away from oil and primary commodities towards manufacturing, especially leather and textiles. These findings suggest that rapid integration with the world economy, while boosting economic growth, might be having adverse environmental consequences. However, they should be interpreted with caution, as the input-output dataunderlyingthe results do not allow a clear differentiationbetween highly polluting textiles and relatively clean garments. More generally, the data on which they are grounded does not distinguish betweenvalue addedand output value, which questionsthe precision ofthe results. Beyond the precise estimates, the possibility of adverse environmental impacts calls for further improvements in environmental standards, especially in the fast-growing industries with the highest intensity in water and toxic pollution. Policies to protect natural assets and public health are needed as well. While some of them may result in an additional cost for businesses, they should reassure those foreign investorsconcernedabout corporateresponsibility. Source: Basedon Muthukurama Mani and Shreyasi Jha (2005). Job losses Changes in the sectoral composition o f output, as Vietnam further integrates with the world economy, could also have adverse effects on employment. At an aggregate level, global integration should lead to faster economic growth and an overall increase in labor demand. But 151 the process may entail bothjob creation in the sectors expanding andjob destruction inthe ones contracting. A flexible labor market, like that of Vietnam, should reduce the social cost o f this kindofjob churning. Indeed, workers who are retrenchedby one employer can hope to get work with a different one withina short period oftime. Butjobs are not all equal; for instance interms of their access to social insurance. Churning might thus be associated with some workers losing "good" formal sectorjobs and ending up inthe informal sector. For them, knowing that overall wage employment i s expandingmay not be a consolation. Unfortunately, knowing in advance which sectors may contract i s not easy. Consider accession to the WTO. The commitments made throughout the process involve a myriad changes in the level of trade barriers, the extent of competition in services, and the overall operation of markets. Moreover, the government could react to these changes invarious ways. A fall intrade- related revenue, for instance, could be compensated through higher tax rates elsewhere, or through lower public spending, or through higher levels of public debt. All this would be on top of other changes, such as those relatedto AFTA commitments. In summary, assessingthe sectoral implications of further global integration might sound like a straightforward concept, but there i s actually a range of mechanisms through which business activity could be affected, and it i s virtually impossible to factor them in all in a single analysis. Different studies on the consequencesof further global integration emphasize different mechanisms to construct integration "scenarios". And they all tend to be quite plausible. So, rather than choosing a preferred scenario to infer possible sectoral impacts of global intergration, there might be something to be learned from their combination. A recent assessment of the potential impacts of WTO accession pooled the findings of two dozen studies recently conducted on Vietnam's integration process to try to infer some common patterns. Such studies were of two sorts. One group, quite popular among economists, relies on so-called computable general equilibrium (CGE) models to simulate the effects of changes in tariffs and subsidies, all the way to wage levels, returns to capital and land rent. The core of these models i s the combination of an input-output table, summarizing the technology o f each sector, and simulated markets for goods and factors of production, mimicking the equalization of supply and demand in each o f them. Another group o f studies ignores links between markets and activities, relying instead on an in-depthknowledge o f specific sectors, and the economic actors who participate in them, to predict how their production and employment levels could react as the economy opens up. The dozen CGE studies considered in this broader assessment differ in the way they model the economy. The number of sectors included inthe analysis ranges from 9 to 33, and the number o f factors of production, from two (just capital and labor) to 14 (including a range o f labor skills, for instance). Put together, this dozen CGE studies includes 36 scenarios for Vietnam's global integration. Almost all o f them involve changes in Vietnamese trade barriers. But 11also consider changes intrade barriers abroad, two assume changes inthe extent o fmarket competition, and three in technology. Two thirds o f the scenarios introduce adjustments in tax rates to compensatefor the lost intrade-related government revenue. Some common patterns emerge from the comparison of the 36, pooled scenarios. If the findings o f these studies are to be believed, global integration could lead to a strong expansion in textile and garments, intrade and freight, in petroleum and lubricants, and inpaper products. But it could also lead to a substantial contraction in equipment and machinery, electric equipment, personal services, and transport vehicles (Figure 12.1). The picture i s more blurred for other sectors, which appear to expand or contract depending on the scenario considered. The dozen sectoral studies display an enormous diversity in their methodology. Partial equilibrium models are used in eight o f them, household surveys in five and focus group 152 discussions in two. Some o f the studies use more than one methodology. There i s also a considerable diversity intheir coverage. Six o f the studies refer to rice, three to textiles, three to sugar, two to maize, two to coffee and two to tea. For some sectors, such as fisheries, only one study could be identified. But some studies cover several sectors at once. Figure 12.1: Changes in Sectoral Output due to Global Integration Trade and freight I 9 8 u) $ 7 5 6 P 5 cO A b 6 f 3 $ 4 $ 2 21 0 0 Equipment and machinery I 6 , ............................................................................................................................................................................ Electric and electronic 7-1............................ ............................................................................................................................................. .: , I Change inoutput Change in output Source: Based on Martin Rama and Sa Kim L e (2005). Figures are constructed out o f a pool o f 36 global integration scenarios from CGE models for Vietnam. The height of the bars indicates how many scenarios fall in each category. Information on all sectors was not available for all scenarios, implying that the sum ofthe bars is generally less than 36. Because o f their narrower focus, and their direct consultation o f key stakeholders, sectoral studies tend to provide richer detail on the possible impacts o f further global integration. Compared to CGE models, where changes in employment are linked to changes in output in a rather mechanical way (through their input-output table), some o f the sectoral studies allow disentangling economic and social impacts. This is because o f their ability to differentiate impacts across enterprises o f different sizes, or operating with different technologies, or with different degrees o f formality. However, some o f the sectoral studies also reflect a high level o f concern for adverse social impacts, and a few are actually close to advocacy pieces. N o t surprisingly, their assessment o fthe consequences o f further global integration is somewhat more negative than the one emerging from CGE studies (Table 12.1). 153 BUSINESS Povertyand inequality Through the simulation of the effects of trade integration on wage levels, returns to capital and land rent, CGE studies also allow identifying possible impacts on poverty or inequality. To do this, they rely on a typology of households with different endowments o f factors of production. For instance, poorer households usually have little more than unskilled labor and some land; on the other hand, richer households typically own capital and their working-age members are skilled. These typologies are constructed usinghousehold survey data. Inthe case of Vietnam, the studies reviewedinclude a range of breakdowns, from just two types o f households at one end (urban and rural) to 20 at the other (by income deciles, in both urban and rural areas). Table 12.1: Economic and Social Impacts in Selected Sectors Sector Economic Impact Social Impact Livestock Positive Unknown Fisheries Positive Unknown Rice Positive Positive Maize Unknown Negative Sugar Positive Negative Coffee Positive Unknown Tea Positive Unknown Textiles Negative Unknown Source: Based on Martin Rama and Sa Kim Le (2005). The changes in factor prices induced by further global integration can then be used to estimate the change intotal income by household type. For instance, if returns to capital were to increase by a smaller percentage than wages for unskilledlabor, richer households would less (in relative terms) than poorer ones, and inequality would decrease. If,on the other hand, the wages for unskilledlabor were to decline, poverty would increase. Each global integration scenario can then be associatedwith a change inthe overall income distribution, and inthe poverty rate. Inthe studies reviewedfor Vietnam, however, results tend to be quite ambiguous. This is because all factor prices tend to change by a similar percentage (Figure 12.2). Across all the integration scenarios for which data could be gathered, the prices o f factors o f production tend to vary by a similar amount, thus leading to very minor changes in either poverty or inequality. On the surface, the predicted increase inreturns to capital appears to be slightly more robust than the predicted increase inwage levels. For the latter, there are a few scenarios predicting no increase at all. However, the difference is probably too small to be o f any significance. Overall, then, the available CGE studies for Vietnam do not allow foreseeing any major change in poverty or inequality stemming from further global integration. 154 Do We Really Know? These analyses convey a reassuring sense of technical rigor, but it would be safer to treat their results as broad indications of the possible impacts only, not as precise estimates. One limitation o f studies of this sort concerns their inability to mimic the effects of increased competition on capital accumulation and on productivity at the firm level. Accession to the WTO should bring more foreign companies into play in Vietnam, domestic enterprises could be tempted to imitate their more performing processes, and stronger exposure to imported products can be expected to lead to quality upgrades. None of these potentially important developments is captured by the studies reviewed. Their underlyingCGE models are basically static, inthe sense that they take the capital endowment and production technology o f each sector as given, or modify them on the basis of ad hoc assumptions. Figure 12.2: Changes inReturnsto Capital and Wages Capital I Chanae in return i Sklled labor Unskilled labor 5 25 z 0 Changeinwage Change inwages Source: Basedon Martin Ramaand Sa Kim L e (2005). See Figure 12.1 for details. Careful comparisons between expected and actual outcomes, after a major episode of trade integration such as the North American Free Trade Agreement, show that CGE models do a poor job at predicting the surge in exports occurring in sectors that initially did not trade much with the rest of the world. Butthere are additional reasons to be prudent in a transition economy like Vietnam. One o f the most important impacts of WTO accession i s to foster competition behind the border, thus helping to perfect market mechanisms "behind the border". But CGE models assume that domestic markets are perfectly competitive to begin with. The local dimension of business development inVietnam makes this assumption all the more questionable. Assuming that the impact of global integration on, say the wages of unskilled workers will be the same across all regions does not sound realistic. 155 Moreover, the studies reviewed focus mainly on the impact o f changes in trade barriers such as tariff cuts or the suppression o f quotas, but these changes are simply be dwarfed by changes in international prices. From rice, to coffee to oil, worldwide instability in commodity markets has been a more important source o f economic fluctuations for Vietnam than trade liberalization (Figure 12.3). Because global integration makes the domestic economy more integrated with world markets, these economic fluctuations should also be considered as social impacts o f the reform agenda. And they are of course much more unpredictable than tariff cuts or the suppression ofquotas. Figure 12.3: Reform Impactsversus External Shocks Percent change between 2002 and 2005 ~ ~ ,; .-a, 80 ~ p 1 aconsensedmilk S t ~ ~ -8.00 -6.00 -4.00 -2.00 0.00 2.00 1 I Change in domestic price implied by change in tariffs I I Source: Own calculations, basedon data from GSO and GDC. The price changes in bothaxes are measuredinpercent, over the period2002-2005. Addressing the potentially adverse effects of further integration with the world economy requires that mechanisms to deal with the unforeseen be put in place. The unforeseen could be job losses in a sector that is suddenly facing a tougher competition than expected; or it could be an increase in poverty in a region specializing in some crop whose price suddenly declines. Neither the workers nor the regions affected can be identified in advance with great confidence. This makes the problem faced by policy makers similar in nature to the problem faced by insurers. Rather than trying to guess where the next hazard will occur, the task i s to estimate the damage accurately if and when it occurs, and to swiftly provide commensurate assistance. Towards rapid response Several options exist for the government to provide assistance in the event o f adverse social impacts from economic reforms, or simply from external shocks. Those options can infact be seen as potential complements of the global integration agenda. However, not all o f them are bound to be equally effective, so that it i s worth discussing the way they would operate inpractice and what their implications would be. 156 SOCIAL~MPACTS A first option is to focus on trade policy instruments. Inpractice, this option amounts to introducing safeguards in the commitments for WTO accession. For instance, by negotiating higher tariff rates for agricultural products which are deemed vulnerable to foreign competition, and whose production is fundamentalto sustainthe livelihoods of the poor. Or to try to maintain tariff rate quotas, triggering higher duties when imports exceed some threshold. While technically feasible, this option could be excessively costly to Vietnam. Not knowing exactly which sectors could need safeguards, it would be necessary to introduce a large number of them. And this in turn would require substantial concessions to negotiating partners in other sectors, if their support to WTO accession is to be secured. Moreover, trade policy instruments are not particularly efficient redistribution tools. Therefore, this first option is not particularly appealing. A second option is to focus on enterprises, by relying on industrial policy tools. For example, by using government transfers to support enterprises or sectors which find it difficult to cope with a surge in imports, or which face more efficient suppliers in foreign markets. Unlike the previous one, this option might simply not be feasible. WTO rules regarding subsidies basically tie the hands of the government, preventing it from making transfers that violate the nationaltreatment principle. The third option is to focus on those who lose their jobs. This is what the social safety net for redundant SOE workers does. The fund has been inoperation for three years now, and has assisted more than 100,000 people (Table 12.2). The severance pay package made available to those who lose their jobs is roughly commensuratewith the loss in earning and wellbeing they experience. This has resulted inrelatively high levels of satisfactionwith the assistance received, as revealedby two tracer surveys ofretrenchedworkers. Table 12.2: The SafetyNet for Redundant SOEworkers I 2002 I 2003 I 2004 1 2005 I Total Total spending(billion dong) 29 533 1,299 1,466 3,327 Workers assisted (persons) 1,147 18,445 43,659 46,815 110,066 Participant SOEs (number) 34 453 873 868 2,228 Assistance (million dongper worker) 25.5 28.9 29.8 31.3 31.3 30.2 Workers assisted per SOE (number) 33.7 40.7 50.0 53.9 49.4 Source: Based on data from MOF. Figures for 2005 are for the first nine months o f the year. Reportedassistance figures are averages across retrenchedworkers. The very existenceof this safety net fund means that the third option is already effective in Vietnam. The question is whether more is needed. Given that the most protected industries, hence the most vulnerable to global integration, are those with a dominant state sector, the answer could be negative. Moreover, scaling up the safety net so as to cover private sector workers as well could be prohibitively expensive, because of the generous assistance provided. Private sector workers with formal jobs are entitled to severance pay, to be provided by their employers, in the event ofjob loss. This is a relatively effective way to provide income support to those becoming unemployedin a country at Vietnam's development level. It could then be arguedthat not muchmore shouldbe done onthis front. Finally, the fourth option is to focus assistance on provinces, if they either suffer disproportionately from adverse shocks, or simply if they lag behind while other parts of the 157 BUSINESS country thrive. Budget transfers targeting geographical areas (as opposed to enterprises) are compatible with WTO rules. Moreover, due to the geographical concentrationof several crops, it i s likely that adverse impacts would be felt more severely in some provinces than in others. And Vietnam already has in place an effective mechanism to transfer budget resources from richer provinces to poorer one. This mechanism combines the fraction o f tax revenue that each province can retain, the equalization grants provided by the National Assembly, spending in large-scale infrastructure, and transfers from national targeted programs. The result has been a massive redistribution of resources across provinces in recent years. And this redistribution has in turn proved effective at reducingpoverty (Figure 12.4). Figure 12.4: BudgetTransfers to Provinces and Poverty Reduction 0 , ?-I- a, ~ 4a,- 2 - 2 x '*' -3 n - ' -6 - Initial poverty rate (percent) Absolute - - - Relative 1 I Source: Own estimates based on data from GSO, MOF and MPI. The lines indicatethe reductioninpoverty associatedwith a net budget transfer of one million dong per person, depending on the initial poverty rate of the province. The effects were estimated for the changes in poverty between 2002 and 2004, using data on transfers around 2003 and controlling for other province characteristics. The initialpovertyrate refers to 1998. The size of the budget transfer to each province varies from year to year. Some o f its components, such as large-scale infrastructure investments, are difficult to modulate on a short notice, But equalization grants are determined on an annual basis, as part of the budget. Several budget allocation norms guidethose grants. Now that Vietnam i s moving towards a more reliable way to measure poverty, these norms could be revised so as to explicitly include poverty rates among their parameters. This would ensure that provinces which are not doing well can receive more assistance (although in that case it would be important not to reward provincial authorities, to avoid introducing perverse incentives). However, there could be cases in which assistance cannot wait for an entire budget cycle. In preparationfor those cases, a contingent fund could be set aside in the budget, in the spirit of the contingent resources which are allocated on an emergencybasis to provinces hit bytyphoons. 158 BIBLIOGRAPHY The word "processed" describes informally reproduced works that may not be commonly available through libraries. ADB (Asian Development Bank). 2004a. Adding Value to Vietnam's Rice Industry and Improvingthe Incomes o f the Poor. Markets and DevelopmentBulletin No. I. Central Institutefor Economic Management, Hanoi. ___------2004b. The Impact o f Land Market Processes on the Poor: "Implementing de Soto". . Making Markets Work Betterfor the Poor, Discussion Paper No. 3. Asian Development Bank, Hanoi. -____----2004c. The Value Chain for Tea in Vietnam: Prospects for Participation of the Poor. . Making Markets Work Betterfor the Poor, Discussion Paper No. 1. Asian Development Bank, Hanoi. ---___---2005a. Agricultural Commercialisation, Value Chains and Poverty Reduction. Making . Markets Work Better for the Poor, Discussion Paper No. 7. Asian Development Bank, Hanoi. --___----2005b. The Conversion of Agricultural Landfor Industrial Use. Hanoi. Processed. . -----____2005c. Labor Market Segmentation and Poverty: A Reportfor ADB M W 4 P Project. . Asian Development Bank, Instituteof World Economics and Politics, Central Institute of Economic Management, and Instituteof Labor and Social Affair Studies, Hanoi. ---______ . 2005d. Vietnam:Private SectorAssessment. Asian DevelopmentBank, Hanoi. Agrifood Consulting International. 2002. Rice ValueChain Study: Vietnam. Report Prepared for the World Bank. Hanoi. Processed. Arulpragasam, Jehan, Francesco Goletti, Tamar Manuelyan Atinc and Vera Songwe. 2004. Trade in Sectors Important to the Poor: Rice in Cambodia and Vietnam and Cashmere in Mongolia. In Kathie Krumm and Homi Kharas (eds.), East Asia Integrates: A Trade Policy Agendafor Shared Growth (pp. 149-169). World Bank, Washington, D.C. Athukorala, Prema-chandra. Forthcoming. Trade Policy Reforms and the Structure o f Protection inVietnam. WorldEconomy. Auffret, Philippe. 2003. Trade Reform in Vietnam: Opportunities with Emerging Challenges. WorldBank Policy Research WorkingPaper 3076. World Bank, Washington, D.C. Bosworth, Malcolm and Trong Khanh Do. 2005. Vietnam's WTOAccession: Tar18 Industry and Subsidy Analysis. Book 2: Report on Subsidies and WTO Accession: Compliance and Policy Implications for Vietnam. Viet Nam: Small Scale Technical Assistance for Capacity Building o f Ministry of Finance to Support Tariff, Industry and Subsidy Analysis for the WTO Accession. Ministry o f Finance, Asian Development Bank and Finance PublishingHouse, Hanoi. Brasdt, Loren. 2005. Prelimnary ObservationsRegarding Land. Processed. Bui, Sy Loi. 2004. Labour Inspection and Corporate Social Responsibility. Paper Presented at the Worshop "Corporate Social Responsibility in the International Economic Integration Process of Vietnam", Ho Chi MinhCity, February 2004. Carlier, Amanda S. and Thanh Son Tran. 2004a. Firm Dynamism: Beyond Registration, How are Vietnam New Domestic Private Firms Faring? Vietnam Private Sector Development Policy Note. The World Bank inVietnam, Hanoi. -----_---. 2004b. Private Sector and Land: Availability, Affordibility, and Administration. VietnamPrivate Sector DevelopmentPolicy Note. The World Bank inVietnam, Hanoi. --------_. 2005. Promoting Business to Business Commercial Contracts in Vietnam: Contractual Linkages between Vietnam's Large and Small-scaled Firms: Dispute Resolution and Contract Enforcement. Vietnam Private Sector Development Policy Note. The World Bank inVietnam, Hanoi. Castel, Paulette and Martin Rama. 2005. Comments on the Draft of the New Social Insurance Law. The World Bank inVietnam, Hanoi. Processed. Chan, Anita and Hongzen Wang. 2004. The Impact of the State on Workers' Conditions: Comparing Taiwanese Factories in China and Vietnam. Pacijk Affairs, 77(4):629-646. Cheshier, Scott, Jag0 Penrose and Jonathan Pincus. 2005. Flying Geese or Cooked Goose? Viet Nam's Insertion into Regional and Global TradePatterns. UNDP, Hanoi. Chu, Quang Khoi. 2004. Determinants of Total Factor Productivity Growth in Vietnam, 1986- 2002. Paper Presented at the Conference of Economic Studies, Hanoi National Economics University, March 2004. CIEM (Central Institutefor Economic Management) and World Bank. 2005. Report of the Study on Post-Equitization of Stated-OwnedEnterprises. Hanoi: Central Institute for Economic .Management and World Bank. Processed. Clarke, Simon. 2004. The Changing Character of Strikes in Vietnam. Processed. Communist Party o f Vietnam. 2005. Preliminary Report Findings )om Diagnostic Survey on Anti-Corruption in Vietnam. Hanoi. Processed. Daley-Harris, Sam. 2004. State of the Microcredit Summit Campaign Report 2004. Microcredit Summit Campaign, Washington, D.C. Dalton, RussellJ. and Nhu-Ngoc T. Ong. 2001. The VietnamesePublic in Transition: The 2001 World Values Survey. Center for the Study of Democracy, University of California, Irvin, Irvin, CA. Dang, NhuVan. 2005. Liberalization of Trade in Services: Policy Implications and Institutional Impacts. Hanoi. Processed. Davidsen, Soren and Carrie Turk. Forthcoming. The Citizens' Report Card in Vietnam: Approach, Methodology and Process. The World Bank inVietnam, Hanoi. DFC. 2005. Assessing Key Issues in the Financial Sector of Vietnam: Draft Final Report. I September 2005. DFC, Barcelona. Processed. Doan, Quang, Lan Huong Nguyen and Thanh Cong Gian. 2004. Reform-Induced Wage Inequality in Vietnam: The Impact of Trade Liberalization and Educational Attainment. Processed. Dong, Hieu. 2005. Industrial Production: SOEs Lose Position, FIEs on the Rise. VietNamNet, October 13. 161 Dufhues, Thomas, Franz Heidhues and Gertrud Buchenrieder. 2004. Participatory Product Designby Using Conjoint Analysis inthe Rural Financial Market of Northern Vietnam. Asian Economic Journal, 18(1):81-113. Ernst & Young. 2004. Vinacoal Operational Review (Funded by Japan Policy and Human ResourceDevelopment (PHRD) grant). Hanoi. Processed. Freeman, Nick J. 2004. Expanding Horizons: Equity Finance in Vietnam. Mekong Project Development Facility Private Sector DiscussionsNo. 16. MPDF, Hanoi. Freeman, Nick, Van LanNguyen and HanhNamNguyen. 2005. Beyondthe Headline Numbers: Business Registration and Startup in Vietnam. Mekong Project Development Facility private sector discussionsNo. 20. MPDF, Hanoi. Greig, Froniga and ThiNguyen. 2005. Voicesof WomenEntrepreneurs in Vietnam. Processed. GSO (General Statistical Office). 2004. Results of the Survey on Households Living Standards (VHLSS) 2002. Statistical PublishingHouse, Hanoi. GTZ (German Agency for Technical Cooperation). 2005a. Business Licensing: Current Status and Ways Forwards (draft). Paper presented at the workshop "Business Licensing in Vietnam: Current Status and the Ways Forward", Hanoi, 18 October 2005. ------_-_. 2005b. Monitoring of Business Licenses and Business Conditions in Vietnam. Paper Presented at the Workshop "Business Licensing in Vietnam: Current Status and the Way Forwards", Ho Chi MinhCity, 14 October 2005. GTZ (German Agency for Technical Cooperation) and CIEM (Central Institute for Economic Management). 2005. From Business Idea to Reality: Still a Long and Costly Journey. Business Issues No. 3. Hanoi. Hansen, Henrik, John Rand and Finn Tarp. 2004. SME Growth and Survival in Vietnam: Did Direct Government Support Matter? Institute of Economics, University of Copenhagen DiscussionPapers04-13. Copenhagen. Huang, Haimin and Thanh Van Thai. 2005. Roundup: Vietnam Strives to Develop Stock Market. People'sDaily Online,August 21. ILONietnamIndustrial Relations Project. 2005. The User'sGuide to the Labour Law, Hanoi. ILSSA (Institute of Labour Science and Social Affairs). 2004. Study on Corporate Social Responsibility Labour-Related Practices. Hanoi. Processed. IMF (International Monetary Fund). 2005. Vietnam - Staff Report for the 2005 Article I V Consultation. Prepared by the Staff Representatives for the 2005 Consultation with Vietnam. Processed. JBIC (Japan Bank for International Cooperation). Various years. Survey Report on Overseas Business Operations by Japanese Manufacturing Companies. Results of JBIC FY2003 Survey: Outlook for Japanese Foreign Direct Investment (15th Annual Survey). JBIC Institute,Tokyo. JETRO (Japan External Trade Organization). 2005. The 15thSuwey of Investment-RelatedCost Comparison in Major Cities and Regions in Asia. JETRO, Tokyo. Kazmin, Amy. 2005. Make or Break Time for Vietnam's IT Entrepreneurs. Financial Times Online,October 11. 162 KfW (German Reconstruction Bank). 2005. Vietnam Country Pilot of Harmonization in Assistance Procedures and Practices - Project Preparation: Final Report. Colenco Power EngineeringLtd., Hanoi. Processed. Kim, Annette M. 2004. A Market without the 'Right' Property Rights: Ho Chi Minh City, Vietnam's Newly-Emerged Private Real Estate Market. Economics of Transition, 12(2): 275-305. Kim, Jong Seok. 2005. Regulatory Reform for a More Competitive Business and Investment Climate. Paper Presented at the Workshop "Business Licensing in Vietnam: Current Status and the Way Forwards", Ho Chi MinhCity, 14 October 2005. Kokko, Ari, Katarina Kotoglou and Anna Krohwinkel-Karlsson. 2003. The Implementation o f FDI in Viet Nam: An Analysis of the Characteristics of Failed Projects. Transnational Corporations, 12(3):40-76. Kokko, Ari and Fredrik Sjoholm. 2004. The Internationalization of VietnameseSMEs. Draft. Stockholm School of Economics. Processed. Mai, Pham Hoang. 2002. Regional Economic Development and Foreign Direct Investment Flows in Vietnam, 1988-98. Journal of theAsia PaciJic Economy, 7(2):182-202. Malesky, Edmund J. and Markus Taussig. 2005. Where is Credit Due? Companies, Banks, and Locally Diferentiated Investment Growth in Vietnam. Paper Presented at the International Convention o fAsian Scholars, Shanghai, China, August 19-23, 2005. Mani, Muthukumara and Shreyasi Jha. 2005. Trade Liberalization and the Environment in Vietnam. Processed. Marsh, Sally P., T. Gordon MacAulay and HuuAnh Le. 2004. Credit Use by Farm Households in Vietnam: Implications for Rural Credit Policy. Paper Presented at the 48 Annual Conference of the Australian Agricultural and Resource Economics Society, Sheraton Towers, Melbourne, 11-13 February 2004. Mekong Economics. 2004. Gender in the Factory: A Study of Emerging Gender Issues in the Garment and Footwear Industries in Vietnam GforNCFAW-W D P - W E Project VIE-01- 015). Mekong Economics, Hanoi. Meyer, Klaus E. and HungV o Nguyen. 2005. Foreign InvestmentStrategies and Sub-national Institutions in Emerging Markets: Evidence from Vietnam. TheJournal of Management Studies,42( 1):63-93. MPDF (Mekong Private Sector Development Facility) and IFC (International Finance Corporation). 2005. A National Survey of Women Business Owners in Vietnam: An ExecutiveReport. Hanoi. Processed. M P I (Ministry of Planning and Investment)and JICA (Japan International Cooperation Agency). 2004. Joint Work between Vietnam and Japan to Strengthen the Competitiveness of Vietnamese Industries in the Context of Formulation of Socio-economic Development Plan 2006-2010: Main Report. Hanoi. Nguyen, Duy Vy. 2004. Role of the Vietnam Confederation of Labour in Promoting Decent Work in Vietnam. Paper Presented at the ASEM Informal Seminar "The Future of Employment in ASESM: The Decent Work Agenda in a Globalised Economy", Hanoi, April 2004. Nguyen, Ha Thanh and Klaus E. Meyer. 2004. Managing Partnerships with State-Owned Joint Venture Companies: Experiences from Vietnam. Business Strategy Review, 15(1):39-50. 163 Nguyen, Hoa and Ulrike Grote. 2004. Agricultural Policies in Vietnam: Producer Support Estimates, 1986-2002. Centerfor Development Research (ZEF) Discussion Paper No. 93. International Food Policy ResearchInstitute,Washington, D.C. Nguyen, Khac Minh, Thanh Long Giang and Ngoc Thang Bach. 2005. Technical EfJiciency in Vietnamese Small Manufacturing Firms: Scholastic Frontier Production Function and DEA Approaches. First draft. Hanoi. Processed. Nguyen, Phuong Quynh Trang, and others. 2001. Doing Business under the New Enterprise Law: A Survey of Newly RegisteredCompanies. Mekong Project Development Facility Private Sector DiscussionsNo. 12. MPDF, Hanoi. Nguyen, Phuong Quynh Trang and Jonathan A. Stromseth. 2002. Business Associations in Vietnam: Status, Roles and Performance. Mekong Project Development Facility Private Sector DiscussionsNo. 13.MPDF, Hanoi. Nguyen, Thang. Forthcoming. Policies and Rural Poverty: A Vietnam Case Study on Using Poverty Predictor Model for Participatory Policy Evaluation. Vietnam Academy of Social Sciences, Hanoi. Nguyen,Thi ThuTrang. 2005. Overview of Business Licenses in Vietnam(WorkingPaper no. 1. Licensing Component.ADB-TA 4418-VIE). Paper Presented at the Workshop "Business Licensing inVietnam: Current Status and the Ways Forward", Hanoi, 18 October 2005. Nguyen, Thi Tue Anh, Xuan NguyetHong Vu, Toan Thang Tran and Manh HaiNguyen. 2005. Impacts of Foreign Direct Investment on the Economic Growth in Vietnam: Research Report of the Capacity Building Project for Policy Research to Implement Vietnam's Socio-economic Development Strategy in the Period 2001-2010. Central Institute for Economic Mangement and SIDA, Hanoi. Nguyen, Viet Cuong. 2005. Impact of Micro-Credit on the Poor and Vulnerable in Vietnam: Empirical Results +om Matching Evaluation Method. Paper Submitted for Vietnam Economic ResearchConsortium (VERCON), May 24,2005. Hanoi. Processed. Oostendorp, Remco. 2004. Migration and Firm-Level Segmentation in the Textileand Garments Sector in Vietnam. ResearchNote. Processed. Oxfam. 2004. Extortion at the Gate: Will Viet Nam Join the WTO on Pro-Development Terms? Oxfam Briejing Paper No. 67. Oxfam International, Washington, D.C. ___----. 2005. Do as ISay, not as IDo: The Unfair Terms for Viet Nam's Entry to the WTO. - Oxfam Briejing Paper No. 74. Oxfam International, Washington, D.C. Pham, Minh Hac and Thanh Nghi Pham. 2004. Public Attitudes toward a Market Economy in Vietnam. Paper Presented at the Conference on "Citizens, Markets and Democracy around the Pacific Rim" March 19-20, 2004, Center for the Study o f Democracy, U C Irvine, The University o f Missouri. Pham, Thi Thu Hang. 2005. Business Women in VietNarn. Processed. Rama, Martin. 2005. Assessing and Projecting the Opportunities and Challenges Faced by Vietnamin the WorldEconomic Context after its Accession to the WTO. Paper Presented at the Workshop on Opportunities and Challenges for Vietnam after Its Accession to the World Trade Organization, MOFA, Hanoi, October 6,2005. Rama, Martin and Sa Kim Le. 2005. Impacts of WTOAccession: Can They be Predicted? What to Do about Them? Paper Presented at the Workshop on "Growth and Social Impacts of WTO Accession", Vietnamese Academy of Social Sciences, Hanoi, November 2,2005. 164 Rand, John. 2005. Credit Constraints and Determinants of the Cost of Capital in Vietnamese Manufacturing. University of Copenhagen. Department of Economics. Discussion Paper. Ravallion, Martin and Dominique van de Walle. 2005. Land and Poverty in Vietnam'sAgrarian Transition. World Bank, Washington, D.C. Roa, Jorge Velazquez. 2005. International Experience and Best Practices in Licenses Reform in the Light of Vietnam's Licenses System (Working paper no. 3. Licensing Component. ADB-TA 4418-VIE). Paper Presented at the Workshop "Business Licensing in Vietnam: Current Status and the Ways Forward", Hanoi, 18 October 2005. Ronnas, Per and Bhargavi Ramamurthy (eds.). 2001. Entrepreneurship in Vietnam: Transformation and Dynamics. Nordic Institute of Asian Studies and Institute o f Southeast Asian Studies, Singapore. Seward, James. 2004. Financial Sector Policy Issues Note: VietnamBank for Social Policies. Draft. World Bank, Hanoi. Processed. Sikor, Thomas. 2004. Conflicting Concepts: Contested Land Relations in North-western Vietnam. Conservation& Society,2(1):75-95. Smith, William. 2005. Strengthening Environmental Management and Land Administration (SEMLA): Short-term Adviser Report for Missions Conductedfiom April 1lth-29th h August 8th-September 1st 2005. MONRE (Ministry of Natural Resources and Environment) and SIDA, Hanoi. Socialist Republic of Vietnam. 2002. Law on Amendments of and Additions to a Number of Articles of the Labour Code. National Assembly Legislature X, 1lth Session. Hanoi, March-April 2002.,Hanoi. STAR-Vietnam. 2003. An Assessment of the Economic Impact of the United States - Vietnam Bilateral TradeAgreement. The National PoliticalPublishingHouse, Hanoi. ______--- . 2005. TheImpact of the US.-Vietnam Bilateral TradeAgreement on Overall and US. Foreign Direct Investmentin Vietnam. Hanoi. Processed. Sunoo, Jan,Jung-Min. 2005. Understanding and Preventing Strikes in Vietnam. ILONietnam Industrial Relations Project, Hanoi. Tarp, Finn, John Rand, HuuDzung Nguyen and Quang Vinh Dao. 2003. Documentation of the Small and Medium Scale Enterprise (SME) Survey in Vietnam for the Year 2002. Department of Economics, University of Copenhagen, Copenhagen. Tran, Quoc Trung, and others. 2005. The Development of Non-Farm Household Enterprises in Vietnam. Vietnam Economic Research Network (VERN) Revised Research Paper No. 1. Institute of Economics, Hanoi. Tran, Tuan. 2005. Life Insurers Reinvest VND15000 Billion Premiums in Vietnam Economy. Vanguard,,September 19. Transparency International. 2005. Corruption Perceptions Index 2005. Transparency International, London. UNCTAD (United Nations Conference on Trade and Development). 2005, World Investment Report 2005: Transnational Corporations and the Internationalization of RhD. United Nations, New York. 165 UNDP (United Nations Development Programme) and MPI (Ministry of Planning and Investment). 2005. Study on CompetitivenessandImpacts of Liberalisation of Financial Services: The Case of Banking Services: Final Draft Report. Ministry of Planning and Investment, United Nations Development Programme - VIE/02/009. Prepared by M C G Management Consulting Ltd.Hanoi. Processed. UNICEF (United Nations Children's Fund). 2005, Survey Assessment of Vietnamese Youth. Ministryof Health, General Statistics Office, with Technical and Financial Support from the World HealthOrganizationand UnitedNations Children's Fund, Hanoi. VCCI (Vietnam Chamber o f Commercce and Industry) and The Asia Foundation. 2005. Replacing Business Licenses by a Regulatory OversightSystem. VCCI, Hanoi. Vietnam Invests More Abroad. 2005. Angolapress Online, September 20. Vijverberg, Wim. 2005. Non-Farm Household Enterprises: Comparison of VHLSS and AHBS Concepts. Draft. Processed. VNCI (Vietnam Competitiveness Initiative). 2005a. Fiscal Incentives for Domestic Investment in Vietnam: Are They Effective? VNCIPolicy Paper No.2. VNCI, Hanoi. VNCI (Vietnam Competitiveness Initiative). 2005b. The Provincial Competitiveness Indexon the Business Environment inVietnam: Summary Report. VNCI, Hanoi. Vu, Quoc Ngu. 2002. The State-Owned Enterprise Reform in Vietnam: Process and Achievements. Visiting Researchers Series No. 4 (2002). Institute of Southeast Asian Studies, Singapore. ___------2003. Technical Efficiency of Industrial State-owned Enterprises in Vietnam. Asian . Economic Journal, 17(1):87-101. World Bank. 2003. Strengthening Developing Country Governments' Engagement with Corporate Social Responsibility: Conclusions and Recommendations from Technical Assistance in Vietnam. Processed. __-_-----2005a. Local DevelopmentInvestmentFunds in Vietnam:Draft Briefing Paper. World . Bank, East Asia and Pacific Region InfrastructureDepartment. Processed. ----__-_-2005b. Overview of the Capital Markets in the Socialist Republic of Vietnam, and . Recommendationsfor Their DevelopmentStrategy. Draft. Hanoi. Processed. _--------. 2005c. Project Appraisal Document on a Proposed (Loadcredit) in the Amount of SDR 45.5 Million (USD 65.9Million Equivalent) to the Socialist Republic of Vietnamfor a Customs Modernization Project. The World Bank inVietnam, Hanoi. Processed. ------_-_. 2005d. Vietnam:Food Safety andAgricultural HealthAction Plan. Hanoi. Processed. --------- . 2005e. WorldDevelopmentIndicators 2005. World Bank, Washington, D.C. ---------. Forthcoming. VietnamInfrastructure Strategy. The World Bank inVietnam, Hanoi World Bank, Netherlands Development Organization and MARD (Ministry o f Agriculture & Rural Development). 2005. State Forest Enterprises Reform: Review of Policy and Implementation Frameworkfor Decree 200. Hanoi. Processed. World Economic Forum. 2005. Growth Competitiveness Index. In August0 Lopez-Claros, Michael E. Porter and Klaus Schwab (eds.), Global Competitiveness Report 2005-2006. Palgrave Macmillan, NewYork. 166 STATISTICAL APPENDIX Population and Employment Table 1.1 Population Table 1.2 Population by locality Table 1.3 Total employment by sectors National Account Table 2.1 GDP by industrial origin and by economic sector in current prices Table 2.2A GDP by industrialorigin and by economic sector in constant prices Table 2.2 B GDP by industrialorigin -- growth rate Table 2.3 A GDP deflator Table 2.3 B Change in GDP deflator Table 2.4 National accounts: sources and uses Balance of Payments Table 3.1 Balance of payments Table 3.2 Merchandiseexports by commodities Table 3.3 Merchandise imports by commodities Monetary Survey Table 4.1 Monetary survey Budget Table 5.1 State budget revenues Table 5.2 State budget revenues: percentage share of GDP Table 5.3 State budget expenditures Table 5.4 State budget expenditures: percentage share of GDP Table 5.5 Public Debt Prices Table 6.1 A Monthly change in consumer retail prices Table 6.1 B Monthly comsumer price index Table 6.2A Price index by commodity groups Table 6.2 B Price index by commodity groups: monthly growth rates Agriculture Table 7.1 Agriculture production in current price Table 7.2 Agriculture production in constant price Table 7.3 Industrial crop production and yields Industry Table 8.1 Industrial productionoutput Table 8.2 Major industrial products Table 1.I POPULATION : (thousand persons) Population Growth By Sex By Area Year (mid-year) Rate Male Female Urban Rural 1976 49,160 2.35 23,597 25,563 10,127 39,033 1977 50,237 2.19 24,197 26,039 10,116 40,114 1978 51,337 2.19 24,813 26,524 10,105 41,226 1979 52,462 2.19 25,444 27,018 10,094 42,368 1980 53,630 2.23 26,047 27,583 10,295 43,335 1981 54,824 2.23 26,665 28,159 10,499 44,324 1982 56,045 2.23 27,297 28,747 10,708 45,336 1983 57,292 2.23 27,944 29,348 10,921 46,371 1984 58,568 2.23 28,607 29,961 11,138 47,429 1985 59,872 2.23 29,285 30,587 11,360 48,512 1986 61,109 2.07 29,912 31,197 11,817 49,292 1987 62,452 2.20 30,611 31,841 12,271 50,181 1988 63,727 2.04 31,450 32,277 12,662 51,065 1989 64,774 1.64 31,589 33,185 12,919 50,801 1990 66,017 1.92 32,203 33,814 12,880 53,136 1991 67,242 1.86 32,814 34,428 13,228 54,015 1992 68,450 1.80 33,242 35,208 13,588 54,863 1993 69,645 1.74 34,028 35,616 13,961 55,683 1994 70,825 1.69 34,633 36,191 14,426 56,399 1995 71,996 1.65 35,237 36,758 16,938 55,057 1996 73,157 1.61 35,857 37,299 15,420 57,737 1997 74,037 1.20 36,473 37,564 16,835 57,202 1998 75,456 1.92 37,090 38,367 17,465 57,992 1999 76,597 1.51 37,662 38,935 18,082 58,515 2000 77,635 1.36 38,166 39,469 18,805 58,830 2001 78,686 1.35 38,684 40,002 19,481 59,205 2002 79,727 1.32 39,197 40,530 20,004 59,723 2003 80,902 1.47 39,755 41,147 20,870 60,033 2004 82,032 1.40 40,318 41,714 21,591 60,441 Note: Population by sex and by area may not add to the total due to the possible exclusion of the armed force and migrant worker Source: GSO, Statistical Yearbook 2004. Table 1.2: POPULATION BY SEX AND LOCALITY IN 2004 (thousand persons) Total By sex By iocality ProvincesICities Male Female Urban Rural Ha Noi 3,083 1,543 1,540 2,000 1,083 Hai Phong 1,771 897 874 708 1,063 Ha Giang 661 327 334 73 588 Tuyen Quang 718 355 363 70 648 Cao Bang 508 249 259 68 440 Lang Son 722 363 359 146 575 Lai Chau 308 155 153 26 282 Dien bien 405 221 183 71 334 Lao Cai 586 282 284 104 462 Yen Bai 724 359 365 143 581 Bac Can 296 148 148 45 251 Thai nguyen 1,095 539 557 218 877 Son La 973 488 485 106 867 Hoa Binh 803 399 404 123 681 Vinh Phuc 1,155 558 597 160 995 Phu Tho 1,315 647 668 196 1,119 Bac ninh 987 480 508 130 858 Bac Giang 1,564 773 791 143 1,421 Quang Ninh 1,087 543 524 506 561 Ha Tay 2,500 1,213 1,288 225 2,275 Hai Duong 1,698 822 877 263 1,436 Hung Yen 1,120 543 577 123 997 Thai Binh 1,843 884 958 133 1,710 Nam Dinh 1,947 948 999 302 1,645 Ha Nam 820 400 421 79 741 Ninh Binh 912 443 468 124 788 Thanh Hoa 3,647 1,785 1,862 354 3,293 Nghe An 3,003 1,475 1,529 317 2,686 Ha Tinh 1,287 636 651 141 1,146 Quang Binh 832 411 420 115 717 Quang Tri 617 305 312 150 466 Thua Thien - Hue 1,120 549 571 350 770 Quang Nam 1,145 703 442 248 898 Da Nang 765 376 389 610 155 Quang Ngai 1,259 611 648 180 1,079 Binh Dinh 1,545 751 795 387 1,158 Phu Yen 849 421 428 177 672 Khanh Hoa 1,111 551 561 441 671 Ninh Thuan 555 274 281 179 375 Binh Thuan 1,136 566 570 394 742 Gia Lai 1,096 545 551 290 806 Kon Turn 336 184 153 127 209 Dac Lac 1,688 864 824 375 1,313 Dac Nong 386 198 188 50 336 Lam Dong 1,139 570 569 428 711 Ho Chi Minh City 5,731 2,774 2,957 4,888 842 Binh Duong 883 419 464 258 625 Tay Ninh 1,030 505 525 174 856 Binh Phuoc 784 398 386 119 665 Dong Nai 2,175 1,077 1,098 702 1,473 Baria- Vung Tau 898 449 449 398 500 Long An 1,401 687 713 230 1,170 Dong Thap 1,639 799 841 246 1,393 An Giang 2,170 1,065 1,105 560 1,610 Tien Giang 1,682 815 867 249 1,433 Ben Tre 1,346 653 693 131 1,215 Vinh Long 1,045 507 538 155 890 Tra Vinh 1,016 492 524 145 870 Can Tho 1,123 551 572 560 563 Hau giang 791 383 408 119 672 SOCTrang 1,527 613 915 232 1,296 Kien Giang 1,630 805 825 389 1,241 Bac Lieu 786 385 401 206 580 Ca Mau 1,201 592 609 238 962 Note: Population by sex and by area may not add to the total due to the possible exclusion of the armed force and migrant workers. Source:GSO. Statistical Yearbook 2004. Table 1.3: TOTAL EMPLOYMENT BY SECTOR (thousandof persons) Rev Prel. 1999 2000 2001 2002 2003 2004 Total Employed Labor Force 35,976 37,610 38,563 39,508 40,574 41,586 State 3,433 3,501 3,604 3,751 4,035 4,142 Non-state 32,543 34,109 34,959 35,757 36,538 37,445 State Sector Employment 3,433 3,501 3,604 3,751 4,035 4,142 Central 1,422 1,442 1,499 1,569 1,628 1,641 Local 2,011 2,059 2,105 2,181 2,407 2,501 Employmentby Sector Agriculture, forestry & fisheries 24,792 23,492 23,386 23,174 23,117 23,026 Industryand Construction 4,300 4,930 5,552 6,085 6,671 7,217 Services 6,884 9,188 9,625 10,249 10,786 11,344 Note: Figures are rounded. Source: GSO, Statistical Yearbook 2004. Table 2.1: GDP BY INDUSTRIAL ORIGINAND BY ECONOMICSECTOR (VND billion,current price) Rev Prel 1999 2000 2001 2002 2003 2004 Total 399,942 441,646 481,295 535,762 613,533 713,071 State 154,927 170,141 184,836 205,652 239,736 279,704 Non-State 245,015 271,505 296,459 330,110 373,797 433,367 Agri, Forestry and Fishery 101,723 108,356 111,858 123,383 138,375 155,144 Agriculture 83,335 87,537 87,861 96,543 106,385 118,258 Forestry 5,737 5,913 6,093 6,500 7,775 9,412 Fisheries 12,651 14,906 17,904 20,340 24,215 27,474 Industry and Construction 137,959 162,220 163,515 206,197 242,126 285,864 Mining 33,703 42,606 44,345 46,153 57,326 72,492 Manufacturing 70,767 61,979 95,211 110,285 125,476 144,924 Electricity and Water 11,725 13,993 16,028 18,201 22,224 23,890 Construction 21,764 23,642 27,931 31,558 37,100 44,558 Services 160,260 171,070 185,922 206,182 233,032 272,063 Trade 59,384 62,636 67,788 75,617 83,297 97,508 Hotel and Restaurant 13,412 14,343 15,412 17,154 18,472 22,381 Transportation and Communication 15,546 17,341 19,431 21,095 24,725 30,402 Finance, Bankingand Insurance 7,488 8,148 6,762 9,763 10,858 12,737 Science and Technology 1,902 2,345 2,646 3,009 3,694 4,315 Real Estate and Renting 18,260 19,173 21,589 24,452 27,287 31,304 Public Administration 11,683 12,066 12,784 13,816 16,676 19,061 Education and Training 14,004 14,841 16,245 18,071 21,403 23,335 Healthcare and socialwelfare 5,401 5,999 6,417 7,057 8,865 10,851 Culture and Recreation 2,376 2,558 2,800 2,987 3,376 3,693 Party and Association 584 614 651 712 774 885 Community and Social Service 9,323 9,853 10,412 11,412 12,497 14,354 Private HouseholdEmployment 895 953 985 1,037 1,108 1,237 Source: GSO, StatisticalYearbook 2004. Tabla 2.2: GDP BY INDUSTRIALORIGINAND BY ECONOMICSECTOR (VND billion.constant 1994price) Prel Prel 1999 2000 2001 2002 2003 2004 Total 256,272 273,666 292,535 313,247 336,242 362,092 State 103,531 111,522 119,824 128,343 138,160 148,865 Non-State 152,741 162,144 172,711 184,904 198,082 213,227 Agri, Forestryand Fishery 60,895 63,717 65,618 68,352 70,827 73,309 Agriculture 52,372 54,493 55,613 57,912 59,761 61,499 Forestry 2,535 2,544 2,556 2,568 2,589 2,610 Fisheries 5,988 6,680 7,449 7,872 8,477 9,200 Industryand Construction 88,047 96,913 106,986 117,126 129,399 142,601 Mining 17,200 18,430 19,185 19,396 20,611 22,948 Manufacturing 46,105 51,492 57,335 63,983 71,363 78,585 Electricityand Water 5,531 6,337 7,173 7,992 8,944 10,015 construction 19,211 20,654 23,293 25,755 28,481 31,053 Services 107,330 113,036 119,931 127,769 136,016 146,182 Trade 41,994 44,644 47,779 51,245 54,747 59,338 Hotel and Restaurant 8,517 8,863 9,458 10,125 10,646 11,432 Transportationand Telecom 10,141 10,729 11,441 12,252 12,925 13,975 Finance, Bankingand Insurance 5,327 5,650 6,005 6,424 6,935 7,495 Scienceand Technology 1,267 1,571 1,749 1,909 2,044 2,196 Real Estateand Renting 11,926 12,231 12,631 13,106 13,796 14,396 PublicAdministration 7,723 8,021 8,439 8,768 9,228 9,773 Educationand Training 8,809 9,162 9,687 10,475 11,260 12,125 Healthcareand socialwelfare 3,707 3,946 4,151 4,464 4,853 5,234 Culture and Recreation 1,505 1,601 1,648 1,706 1,857 1,997 Party and Association 300 317 334 353 372 395 Communityand Social Service 5,564 5,734 6,026 6,353 6,743 7,194 Private HouseholdEmployment 550 567 583 589 610 632 Source : GSO, StatisticalYearbook 2004. Table 2.26: GDP BY INDUSTRIAL ORIGIN --GROWTH RATE (Billionsof Dong in Constant 1994 Prices) Prel Prel 1999 2000 2001 2002 2003 2004 Total 4.8 6.8 6.9 7.1 7.3 7.7 State 2.6 7.7 7.4 7.1 7.6 7.7 Non-State 6.3 6.2 6.5 7.1 7.1 7.6 Agri, Forestry and Fishery 5.2 4.6 3.0 4.2 3.6 3.5 Agriculture 5.5 4.0 2.1 4.1 3.2 2.9 Forestry 3.1 0.4 0.5 0.5 0.8 0.8 Fisheries 3.8 11.6 11.5 5.7 7.7 8.5 Industry and Construction 7.7 10.1 10.4 9.5 10.5 10.2 Mining 13.4 7.2 4.1 1.1 6.3 11.3 Manufacturing 8.0 11.7 11.3 11.6 11.5 10.1 Electricity and Water 7.7 14.6 13.2 11.4 11.9 12.0 Construction 2.4 7.5 12.8 10.6 10.6 9.0 Services 2.3 5.3 8.1 6.5 6.5 7.5 Trade 2.0 6.3 7.0 7.3 6.8 8.4 Hoteland Restaurant 2.5 4.1 6.7 7.1 5.1 7.4 Transportation and Telecom 6.3 5.8 6.6 7.1 5.5 8.1 Finance, Bankingand Insurance 10.0 6.1 6.3 7.0 8.0 8.1 Science and Technology -9.0 24.0 11.3 9.1 7.1 7.4 Real Estate and Renting 2.1 2.6 3.3 3.8 5.3 4.3 Public Administration -5.5 3.9 5.2 3.9 5.2 5.9 Education and Training 2.3 4.0 5.7 8.1 7.5 7.7 Healthcare and socialwelfare 4.0 6.4 5.2 7.5 8.7 7.9 Culture and Recreation 6.6 6.4 2.9 3.5 8.9 7.5 Party and Association 1.o 5.7 5.4 5.7 5.4 6.2 Community and Social Service 2.4 3.1 5.1 5.4 6.1 6.7 Private HouseholdEmployment 1.5 3.1 2.8 1.o 3.6 3.6 Source: GSO, StatisticalYearbook 2004. Table 2.3A: GDP DEFLATOR Prel Prel 1999 2000 2001 2002 2003 2004 Total 156.1 161.4 164.5 171.0 182.5 196.9 State 149.6 152.6 154.3 160.2 173.5 187.9 Non-State 160.4 167.4 171.7 178.5 188.7 203.2 Agri, Forestry and Fishery 167.0 170.1 170.5 180.5 195.4 211.6 Agriculture 159.1 160.6 158.0 166.7 178.0 192.3 Forestry 226.3 232.4 238.4 253.1 300.3 360.6 Fisheries 211.3 223.1 240.4 258.4 285.7 298.6 Industry and Construction 156.7 167.4 171.5 176.0 187.1 200.5 Mining 195.9 231.2 231.1 238.0 278.1 315.9 Manufacturing 153.5 159.2 166.1 172.4 175.8 184.4 Electricity and Water 212.0 220.8 223.4 227.7 248.5 238.5 Construction 113.3 114.5 119.9 122.5 130.3 143.5 Services 149.3 151.3 155.0 161.4 171.3 186.1 Trade 141.4 140.7 141.9 147.6 152.1 164.3 Hotel and Restaurant 157.5 161.8 163.0 169.4 173.5 195.8 Transportation and Telecom 153.3 161.6 169.8 172.2 191.3 217.5 Finance, Bankingand insurance 140.6 144.2 145.9 152.0 156.6 169.9 Science and Technology 150.1 149.3 151.3 157.6 180.7 196.5 Real Estateand Renting 153.1 156.8 170.9 186.6 197.8 217.4 Public Administration 151.3 150.4 151.5 157.6 180.7 195.0 Education and Training 159.0 162.0 167.7 172.5 190.1 192.5 Healthcare and social welfare 145.7 152.0 154.6 158.1 182.7 207.3 Culture and Recreation 156.0 159.8 169.9 175.1 181.8 184.9 Party and Association 194.7 193.7 194.9 201.7 208.1 224.1 Community and SocialService 167.6 171.8 172.8 179.6 185.3 199.5 Private HouseholdEmployment 162.7 168.1 169.0 176.1 181.6 195.7 Source: GSO, StatisticalYearbook2004. Table 2.38: CHANGE IN GDP DEFLATOR (1995=100) Prel Prei 1999 2000 2001 2002 2003 2004 Total 156.1 161.4 164.5 171.0 182.5 196.9 State 127.3 129.8 131.3 136.4 147.7 159.9 Non-State 176.5 184.2 188.8 196.4 207.6 223.6 Agri, Forestry and Fishery 137.8 140.3 140.6 148.9 161.1 174.6 Agriculture 131.8 133.0 130.8 138.1 147.4 159.3 Forestry 191.0 196.2 201.2 213.7 253.5 304.4 Fisheries industry and Construction 139.4 148.9 152.6 156.6 166.4 178.3 Mining 184.1 217.2 217.2 223.6 261.4 296.8 Manufacturing 135.2 140.2 146.3 151.8 154.9 162.5 Electricity and Water 152.6 159.0 160.8 163.9 178.9 171.7 Construction 104.7 105.8 110.8 113.2 120.3 132.6 Services 126.9 128.7 131.8 137.2 145.7 158.2 Trade 126.7 126.1 127.1 132.2 136.3 147.2 Hotel and Restaurant 123.1 126.5 127.4 132.4 135.6 153.0 Transportation and Teiecom 132.0 139.2 146.3 148.3 164.7 187.3 Finance, Bankingand insurance 120.3 123.4 124.9 130.1 134.0 145.4 Science and Technology 127.3 126.5 128.2 133.6 153.2 166.6 Real Estate and Renting 120.3 123.2 134.3 146.6 155.4 170.9 PublicAdministration 129.1 128.4 129.3 134.4 154.2 166.4 Educationand Training 133.6 136.1 140.9 145.0 159.7 161.7 Healthcareand social welfare 120.4 125.6 127.7 130.6 150.9 171.3 Culture and Recreation 138.1 139.6 148.4 153.0 158.8 161.6 Party and Association 153.6 152.9 153.8 159.2 164.2 176.8 Communityand Social Service 131.2 134.6 135.3 140.7 145.1 156.2 Private HouseholdEmployment 140.9 145.6 146.3 152.5 157.3 169.5 Source: GSO, StatisticalYearbook 2004. Table 2.4la: NATIONAL ACCOUNTS: SOURCES AND USES (VND billion, current price) Rev Prel 1999 2000 2001 2002 2003 2004 Sources 411,360 452,524 492,277 563,446 664,671 767,071 GDP 399,942 441,646 481,295 535,762 613,443 713,071 Trade Balance 11,418 10,878 10,982 27,684 51,228 54,000 Uses 411,360 452,524 492,277 563,446 664,731 767,071 Total Consumption 301,690 321,853 342,607 382,137 445,221 511,221 Gross Capital Formation 110,503 130,771 150,033 177,983 217,434 253,686 Statistical Discrepancy -833 -100 -363 3,326 2,076 2,164 Source : GSO, StatisticalYearbook 2004. Table 2.4lb: NATIONAL ACCOUNTS: SOURCES AND USES (VND billion, constant 1994 price) Rev Prel 1999 2000 2001 2002 2003 2004 Sources 269,429 283,751 304,230 334,640 357,691 392,216 GDP 256,272 273,666 292,535 313,247 336,243 362,093 Trade Balance 13,157 10,085 11,695 21,393 21,448 30,123 Uses 269,429 283,751 304,232 334,640 367,691 392,216 Total Consumption 194,350 200,665 210,029 225,610 243,515 260,940 Gross Capital Formation 75,830 83,496 92,487 104,256 116,623 128,916 Statistical Discrepancy -751 -410 1,716 4,774 7,553 2,360 Source: GSO, Statistical Yearbook 2004. Table 3.1: BALANCE OF PAYMENTS (US$ million, unless otherwise indicated) Rev Prel 1999 2000 2001 2002 2003 2004 Exports (fob) 11,540 14,448 15,029 16,706 20,149 26,503 Imports (fob) 10,463 14,071 14,399 17,620 22,745 28,860 Trade Balance 1,077 377 630 -914 -2,596 -2,357 Non-Factor Services -548 -615 -573 -640 -775 -880 Services (net) -427 -597 -635 -795 -815 -950 Transfers (net) - 1,180 1,476 1,250 1,931 2,240 2,485 Official Transfers 130 136 150 131 140 175 - Private Transfers (net) 1,050 1,340 1,100 1,800 2,100 2,310 Current Account Balance (incl. Grants) 1,153 641 672 -418 -1,946 -1,702 Capital Account -323 -527 -479 865 4,092 2,375 Mediumand Long-Term -423 729 623 526 1,047 1,443 - Disbursements 944 1,411 988 1,100 1,540 1,920 ODA Loans 787 1,361 958 1,075 1,260 1,394 - Amortizations Non-concessional LT Borrowing 0 50 30 25 280 526 1,367 682 365 574 493 655 Short-Term -600 -1,755 -1,535 -1,270 1,745 -290 Direct Investment 700 1,100 1,252 2,023 1,900 1,950 FDIloanrepayments 603 601 819 414 600 550 Overall Balance 0.8 114 193 447 2,146 673 Financing 0.8 -114 -193 -447 -2,146 -673 Source: SBV, IMF andWB estimates. Table 3.2: MAJOR EXPORTS BY COMMODITY (US$ million) Rev Pre 1999 2000 2001 2002 2003 2004 Total Exports 11,540 14,448 15,027 16,706 20,176 26,503 Rice 1,025 667 625 726 721 950 Quantity (000 tons) 4,508 3,477 3,729 3,241 3,813 4,060 Average Unit Value (US$/ton) 227 192 168 224 189 234 Crude oil 2,092 3,503 3,126 3,270 3,812 5,671 Quantity (000 tons) 14,882 15,424 16,732 16,879 17,143 19,501 Average Unit Value (US$/ton) 141 227 187 194 222 291 Coal 96 94 113 156 184 355 Quantity (000 tons) 3,260 3,251 4,290 6,049 7 246 I 11,624 Average Unit Value (US$/ton) 29 29 26 26 25 31 Rubber 147 166 166 268 378 641 Quantity (000 tons) 265 273 308 449 433 975 Average Unit Value (US$/ton) 555 607 539 597 872 658 Tea 45 70 78 83 60 96 Quantity (000 tons) 36 56 68 75 60 99 Average Unit Value (US$/ton) 1,250 1,250 1,150 1,103 1,002 961 Coffee 585 501 391 322 505 641 Quantity (000 tons) 482 734 931 719 749 975 Average Unit Value (US$/ton) 1,214 683 420 449 674 658 Cashew Nut 110 167 152 209 284 436 Quantity (000 tons) 18 34 44 62 84 105 Average Unit Value (US$/ton) 5,978 4,892 3,474 3,358 3,390 4,150 Black Pepper 137 146 91 107 105 152 Quantity (000 tons) 35 37 57 77 74 112 Average Unit Value (US$/ton) 3,914 3,943 1,601 1,399 1,416 1,362 Marine Products 974 1,479 1,778 2,023 2,200 2,401 Vegetable & Fruits 107 214 330 201 151 179 Textiles and Garments 1,747 1,892 1,975 2,752 3,687 4,386 Footwear 1,387 1,465 1,559 1,867 2,268 2,692 Handicraft 168 237 235 331 367 426 Wood products 221 294 335 435 567 1,139 Electronic and Computer parts 585 7,884 709 605 855 1,075 Electric cables and wires 154 186 263 389 Plastic products 76 100 134 153 154 261 Source: GSO, Statistical Yearbook 2004. Table 3.3: MAJOR IMPORTS BY COMMODITY (US$ millions) Rev Pre 1999 2000 2001 2002 2003 2004 Total Imports 11,742 15,637 16,162 19,733 25,227 31,954 Petroleum products 1,054 2,058 1,828 2,017 2,433 3,574 Quantity (000 tons) 7,403 8,777 8,998 9,966 9,995 11,050 Average Unit Value (US$/ton) 142 234 203 202 243 323 Fertilizers 464 509 404 477 628 824 Quantity (000 tons) 3,782 3,973 3,189 3 a24 4,119 4,079 ~ Average Unit Value (US$/ton) 123 128 127 125 152 202 Steel and Irons 587 812 965 1,334 1,657 2,573 Quantity (000 tons) 2,264 2,868 3,938 4,951 4,574 5,186 Average Unit Value (US$/ton) 259 283 245 269 362 496 Others Machineryand Equipment 2,005 2,571 2,741 3,793 5,359 5,249 Textile fiber and yarn 194 231 247 314 298 339 Cotton 91 101 132 97 106 190 Materialfor garment & footwear 1,096 1,422 1,590 1,711 2,034 2,253 Motorvehicles 130 238 433 604 834 904 Motorbikes 399 787 670 422 329 452 Pharmaceutical material 57 62 69 83 76 100 Medicine 263 325 296 320 374 410 Paper of all kinds 102 142 159 193 230 248 Chemicals 258 307 352 406 510 683 Chemical products 227 304 361 482 582 706 Plastic materials 346 480 495 617 785 1,191 Electroniccomponents 518 748 503 431 526 430 Computer and component 112 134 163 233 449 912 Wood - sawn and log 91 158 163 179 274 539 Milk and dairy products 101 141 247 122 164 206 Animal feed and materials 115 159 179 234 421 475 Source : GSO, Statistical Yearbook 2004. Table 4.1: MONETARY SURVEY ACCOUNT 1999 2000 2001 2002 2003 2004e (in trillion dong, end of period) Net ForeignAssets 61.2 95.7 117.6 117.4 131.4 145.8 Foreign assets 77.7 112.7 135.9 135.9 150.5 172.3 Foreign liabilities -16.5 -17.0 -18.3 -18.4 -19.1 -26.4 Net Domestic Assets 99.2 127.2 162.2 211.7 279.8 390.3 Domestic credit 115.7 155.2 191.2 239.9 316.9 435.2 Net claims on government 3.0 -0.5 2.1 8.8 20.1 14.9 Credit to the economy 112.7 155.7 189.1 231.1 296.7 420.3 Claims on state enterprises 54.3 69.9 79.7 89.5 105.4 142.9 Claims on other sectors 58.4 85.8 109.4 141.6 191.3 277.4 Other items, net -16.5 -28.0 -29.0 -28.2 -37.0 -44.9 Broad money (M2) 160.5 222.9 279.8 329.1 411.2 536.2 of which: total deposit 119.1 170.7 213.5 254.9 320.6 427.1 Dong liquidity 116.7 152.5 191.1 235.5 314.1 408.1 Currency outside banks 41.4 52.2 66.3 74.3 90.6 109.1 Deposits 75.3 100.3 124.8 161.2 223.6 299.0 Foreign currency deposits 43.8 70.4 88.7 93.6 97.1 128.1 (Annual change in percent) Net ForeignAssets 96.2 56.4 22.9 -0.2 11.9 11.0 Net Domestic Assets 39.1 28.2 27.5 30.5 32.2 39.5 Domestic credit 42.7 34.1 23.2 25.5 32.1 37.4 Credit to the economy 55.0 38.2 21.5 22.2 28.4 41.7 Claims on state enterprises 42.5 28.7 14.0 12.3 17.8 35.6 Claims on other sectors 68.8 46.9 27.5 29.4 35.1 45.0 Total liquidity 56.6 38.9 25.5 . 17.6 25.0 30.4 of which: total deposit 57.7 43.3 25.1 19.4 25.8 33.2 Dong liquidity 48.9 30.7 25.3 23.2 33.4 29.9 Currency outside banks 53.3 26.1 27.0 12.0 22.0 20.4 Deposits 46.5 33.2 24.4 29.2 38.7 33.7 Foreign currency deposits 81.7 60.7 26.0 5.6 3.7 32.0 Note: Datafrom 1999 onwards comprise the SBV, six SOCBs and 83 non-state banks. Source : SBV and IMF. Table 5.1: STATE BUDGET REVENUES (VN dong billion) 1999 2000 2001 2002 2003 2004 final account est A Total revenues and grants 78,489 90,794 103,888 121,716 157,056 180,397 I Current revenues 75,357 87,928 100,918 118,346 144,822 163,884 II Taxes 63,942 79,497 91,688 106,154 126,947 153,432 1 Corporate income tax 17,419 28,950 33,298 36,826 47,410 55,290 2 Individual income tax 1,856 1,831 2,058 2,338 2,951 3,521 3 Land and housing tax 342 366 330 336 359 438 4 License tax 361 381 400 407 778 629 5 Tax on the transfer of properties 1,046 934 1,191 1,332 1,817 2,610 6 Tax on land use right transfer 347 213 298 327 408 640 7 Value added tax 17,239 17,072 19,327 25,916 33,130 41,394 8 Special consumption tax 4,475 5,250 6,229 7,272 8,850 14,212 9 Natural resources tax 4,552 7,487 8,416 8,543 9,719 13,026 10 Agricultural tax 1,973 1,776 814 772 151 130 11 Export & import tax 14,441 13,437 17,458 21,915 21,374 21,542 12 Other taxes 1,236 158 158 170 111 Fees, charges and non-tax 11,415 8,431 9,230 12,192 17,875 10,452 13 Revenue from discrepancy of import prices 950 131 116 168 133 40 14 Fees and charges 3,558 4,950 5,120 6,016 6,483 6,841 15 Rental of land 574 568 570 459 513 1,053 16 Others 3,446 2,782 3,424 5,549 10,746 2,518 IV Capital revenues 771 838 959 1,120 9,265 14,199 Vlll Grants 2,361 2,028 2,011 2,250 2,969 2,314 B Carry-over 3,400 2,145 19,353 16,390 Source MOF Table 5.2:STATE BUDGET REVENUES (share of GDP) 1999 2000 2001 2002 2003 2004 final account est A Total revenues and grants 19.6 20.6 21.6 22.7 25.6 25.3 I Current revenues 18.8 19.9 21.0 22.1 23.6 23.0 1.1 Taxes 16.0 18.0 19.1 19.8 20.7 21.5 1 Corporate income tax 4.4 6.6 6.9 6.9 7.7 7.8 2 Individual income tax 0.5 0.4 0.4 0.4 0.5 0.5 3 Land and housing tax 0.1 0.1 0.1 0.1 0.1 0.1 4 License tax 0.1 0.1 0.1 0.1 0.1 0.1 5 Tax on the transfer of properties 0.3 0.2 0.2 0.2 0.3 0.4 6 Tax on land use right transfer 0.1 0.0 0.1 0.1 0.1 0.1 7 Value added tax 4.3 3.9 4.0 4.8 5.4 5.8 8 Special consumption tax 1.1 1.2 1.3 1.4 1.4 2.0 9 Natural resources tax 1.1 1.7 1.7 1.6 1.6 1.8 10 Agricultural tax 0.5 0.4 0.2 0.1 0.0 0.0 11 Export & import tax 3.6 3.0 3.6 4.1 3.5 3.0 12 Other taxes 0.3 0.0 0.0 0.0 0.0 0.0 1.2 Fees, charges and non-tax 2.9 1.9 1.9 2.3 2.9 1.5 13 Revenue from discrepancy of import prices 0.2 0.0 0.0 0.0 0.0 0.0 14 Fees and charges 0.9 1.1 1.1 1.1 1.1 1.0 15 Rental of land 0.1 0.1 0.1 0.1 0.1 0.1 16 Others 0.9 0.6 0.7 1.o 1.8 0.4 II Capital revenues 0.2 0.2 0.2 0.2 1.5 2.0 111 Grants 0.6 0.5 0.4 0.4 0.5 0.3 B Carry-over 0.0 0.0 0.7 0.4 3.2 2.3 Source: MOF. Table 5.3:STATE BUDGET EXPENDITURES (VN dong billion) 1999 2000 2001 2002 2003 2004 final account est A Total expenditures 81,817 99,751 117,285 129,434 162,150 186,299 I Current expenditures 55,120 70,127 77,049 84,216 102,521 123,425 1 Administration expenditure 6,793 8,089 8,734 8,599 11,359 14,710 2 Expenditure on economic affairs & services 4,772 5,796 6,288 7,987 8,164 11,423 3 Social expenditures 25,576 30,694 37,369 40,747 50,185 59,750 3.1 Education 7,994 9,910 12,006 13,758 17,390 21,337 3.2 Training 2,341 2,767 3,426 4,086 5,491 6,265 3.3 Health 3,117 3,453 4,211 4,656 5,372 6,976 3.4 Science, technology & environment 859 1,243 1,625 1,852 1,853 2,411 3.5 Culture 713 919 921 1,066 1,258 1,490 3.6 Radio and television 682 717 838 681 1,056 1,121 3.7 sports 321 387 483 586 648 643 3.8 Population and family planning 547 559 434 841 666 473 3.9 Social subsidies 9,002 10,739 13,425 13,221 16,451 19,034 4 Interest payment 2,327 3,514 4,485 5,330 6,395 6,450 5 Defence 13,058 14,968 6 Public security 5,745 7,054 7 Others 15,652 22,034 20,173 21,553 7,615 9,070 II Investment expenditure 26,697 29,624 40,236 45,218 59,629 62,874 1 Capital expenditure 24,684 26,211 36,139 40,740 54,430 58,125 2 Others 2,013 3,413 4,097 4,478 5,199 4,749 B Carry-over 3,400 2,145 4,443 16,390 22,513 Source : MOF. Table 5.4: STATE BUDGET EXPENDITURES (share of GDP) 1999 2000 2001 2002 2003 2004 final account est A Total expenditures 20.5 22.6 24.4 24.2 26.4 26.1 I Current expenditures 13.8 15.9 16.0 15.7 16.7 17.3 1 Administration expenditure 1.7 1.8 1.8 1.6 1.9 2.1 2 Expenditure on economic affairs & services 1.2 1.3 1.3 1.5 1.3 1.6 3 Social expenditures 6.4 6.9 7.8 7.6 8.2 8.4 3.1 Education 2.0 2.2 2.5 2.6 2.8 3.0 3.2 Training 0.6 0.6 0.7 0.8 0.9 0.9 3.3 Health 0.8 0.8 0.9 .0.9 0.9 1.o 3.4 Science, technology & environment 0.2 0.3 0.3 0.3 0.3 0.3 3.5 Culture 0.2 0.2 0.2 0.2 0.2 0.2 3.6 Radio and television 0.2 0.2 0.2 0.1 0.2 0.2 3.7 sports 0.1 0.1 0.1 0.1 0.1 0.1 3.8 Population and family planning 0.1 0.1 0.1 0.2 0.1 0.1 3.9 Social subsidies 2.3 2.4 2.8 2.5 2.7 2.7 4 Interest payment 0.6 0.8 0.9 1.o 1.o 0.9 5 Defence 2.1 2.1 6 Public security 0.9 1.0 7 Others 3.9 5.0 4.2 4.0 1.2 1.3 II Investment expenditure 6.7 6.7 8.4 8.4 9.7 8.8 1 Capital expenditure 6.2 5.9 7.5 7.6 8.9 8.2 2 Others 0.5 0.8 0.9 0.8 0.8 0.7 B Carry-over 0.8 0.4 0.8 2.7 3.2 Source : MOF. Table 5.5. PUBLIC DEBT (US$ million, unless otherwise indicated) ~ 1999 2000 2001 2002 2003 rev DEBT OUTSTANDING (LDOD) 20,479 11,581 11,427 12,160 14,189 Public and publicly guaranteed 20,479 11,581 11,427 12,160 14,189 Official creditors 16,817 8,677 9,165 10,508 12,745 Multilateral 1,606 1,895 2,210 2,868 3,560 Concessional 1,584 1,846 2,156 2,794 3,852 Bilateral (1) 15,211 6,782 6,955 7,639 8,819 Concessional 3,975 6,015 6,193 6,850 7,828 Private creditors 3,662 2,904 2,262 1,652 1,444 Bonds 560 560 560 560 560 Commercial banks 2,322 1,685 1,101 552 406 Other private 781 659 602 540 464 Memorandum item IDA 989 1,113 1,344 1,715 2,472 Debt services public - in - percent of XGS 10.0 7.5 6.7 6.0 3.4 - in percent of GDP 4.9 4.2 3.7 3.4 2.0 Note: Figures are rounded. (1) From 2000 onwards, data reflects the reschedulingof non-convertibleRussian debt. Source : World Bank, Global DevelopmentFinance 2005. Table 6.1A: MONTHLY CHANGE IN CONSUMER RETAIL PRICES MonthNear 1999 2000 2001 2002 2003 2004 January 1.7 0.4 0.3 1.1 0.9 1.1 February 1.9 1.6 0.4 2.2 2.2 3.0 March -0.7 -1.1 -0.7 -0.8 -0.6 0.8 April -0.6 -0.7 -0.5 0.0 0.0 0.5 May -0.4 -0.6 -0.2 0.3 -0.1 0.9 June -0.3 -0.5 0.0 0.1 -0.3 0.8 July -0.4 -0.6 -0.2 -0.1 -0.3 0.5 August -0.4 0.1 0.0 0.0 -0.1 0.6 September -0.6 -0.2 0.5 0.2 0.1 0.3 October -1.o 0.1 0.0 0.3 -0.2 0.0 November 0.4 0.9 0.2 0.3 0.6 0.2 December 0.5 0.1 1.o 0.3 0.8 0.6 Source : GSO, Statistical Yearbooks 2004. Table 6.1B: MONTHLY CONSUMER RETAIL PRICE INDEX (Jan 1995=100) MonthNear 1999 2000 2001 2002 2003 2004 January 130.7 129.1 128.3 130.4 135.3 139.5 February 133.2 131.2 128.8 133.2 138.2 143.6 March 132.3 129.8 127.9 132.2 137.4 144.8 April 131.5 128.9 127.3 132.2 137.4 145.6 May 131.0 128.1 127.0 132.6 137.2 146.9 June 130.6 127.4 127.0 132.7 136.8 148.1 July 130.1 126.7 126.8 132.6 136.4 148.8 August 129.5 126.8 126.8 132.6 136.3 149.7 September 128.8 126.6 127.4 132.8 136.4 150.1 October 127.5 126.7 127.4 133.2 136.1 150.1 November 128.0 127.8 127.7 133.7 137.0 150.3 December 128.6 127.9 129.0 134.1 138.0 151.2 Annual Index: 130.21 128.1 127.6 132.7 136.9 147.4 (Jan 1995=100) I Annual Growth Rate 4.31 -1.6 -0.4 4.0 3.2 7.7 Dec/Dec Growth Rate 0.1 -0.5 0.8 4.0 3.0 9.5 Source: GSO. Statistical Yearbooks 2004. .? 0 0 8 I1 0 hl C -J I 23 s (3 5 0 a 0 a 0 L3n z 0W K n 2 EQ F Table 7.1: AGRICULTURAL PRODUCTION (VND billion, current prices) Gross Output 128,416 129,141 130,178 145,021 153,955 172,696 Crop Cultivation 101,648 101,044 101,403 111,172 116,066 131,754 Livestock 23,773 24,960 25,501 30,575 34,457 37,344 Services 2,995 3,137 3,273 3,275 3,433 3,598 Source: GSO. Statistical Yearbooks 2004. Table 7.2: AGRICULTURAL PRODUCTION (VND billion, constant 1994 prices) Rev Prel 1999 2000 2001 2002 2003 2004 Gross Output 106,368 112,112 118,990 122,150 127,628 133,046 Crop Cultivation 86,309 90,858 92,907 98,061 101,763 106,581 Food Crops 52,720 55,163 55,066 59,619 61,029 63,538 Industrial Crops 19,906 21,782 23,109 22,247 24,175 25,351 Livestock 17,337 18,505 19,283 21,200 22,907 23,439 Services 2,650 2,748 2,800 2,890 2,958 3,026 Memorandum Items: Paddy Ouput (000 tons) 31,394 32,530 32,108 34,447 34,569 35,868 Cultivated Area (000 ha) 7,654 7,666 7,493 7,504 7,452 7,444 Yield (ton/ha) 4.10 4.24 4.29 4.59 4.64 4.82 Source: GSO. Statistical Yearbooks 2004. Table 7.3: INDUSTRIAL CROP PRODUCTIONAND YIELDS ~ Rev Prel 1999 2000 2001 2002 2003 2004 Production (000 metric tons) Cotton 22 19 34 40 35 30 Jute 9 11 15 20 12 14 Sedge 73 61 65 88 96 89 Sugar cane 17,760 15,044 14,657 17,120 16,855 15,880 Peanut 318 355 363 400 406 451 Soya-beans 147 149 174 206 220 242 Tobacco 36 27 32 33 32 28 Tea - raw and fresh 317 315 340 424 449 488 Coffee 553 803 841 700 794 835 Rubber 249 291 313 298 364 400 Black pepper 31 39 44 47 69 74 Coconut 1,104 885 892 915 893 931 Area Cultivated (000 ha) Cotton 21 19 28 34 28 27 Jute 4 6 8 10 5 5 Sedge 11 9 10 12 14 13 Sugar cane 344 302 291 320 313 287 Peanut 248 245 245 247 244 259 Soya-beans 129 124 140 159 166 183 Tobacco 33 24 24 27 23 19 Tea 85 88 98 109 116 119 Coffee 478 562 565 522 510 503 Rubber 395 412 416 429 441 451 Black pepper 18 28 36 48 51 51 Coconut 164 161 156 140 134 133 Average Yield (metric ton/ha) Cotton 1.o 1.o 1.2 1.2 1.3 1.1 Jute 2.3 2.1 1.9 2.1 2.6 3.0 Sedge 6.7 6.6 6.6 7.2 6.8 7.0 Sugar cane 51.6 49.8 50.4 53.5 53.8 55.3 Peanut 1.3 1.5 1.5 1.6 1.7 1.7 Soya-beans 1.1 1.2 1.2 1.3 1.3 1.3 Tobacco 1.1 1.I 1.3 1.2 1.4 1.5 Tea raw and fresh - 3.7 3.6 3.5 3.9 3.9 4.1 Coffee 1.2 1.4 1.5 1.3 1.6 1.7 Rubber 0.6 0.7 0.8 0.7 0.8 0.9 Black pepper 1.8 1.4 1.2 1.o 1.4 1.4 Coconut 6.8 5.5 5.7 6.5 6.7 7.0 Source : GSO, Statistical Yearbook 2004. Table 8.1: INDUSTRIAL PRODUCTIONOUTPUT (VND billion, constant 1994 price) Rev Prel. 1999 2000 2001 2002 2003 2004 Gross Industrial Output 168,749 198,326 227,342 261,092 305,080 354,030 State sector 73,208 82,897 93,434 105,119 117,637 131,570 Central 48,395 54,962 62,119 69,640 80,917 92,653 Local 24,813 27,935 31,316 35,479 36,720 38,917 Non-state sector 37,027 44,144 53,647 63,474 78,292 96,150 Collectives 1,076 1,334 1,575 1,668 1,770 1,912 Private, households and mixed 35,951 42,810 52,072 61,807 76,522 94,238 Foreign-invested sector 58,515 71,285 80,261 92,499 109,152 126,310 Key Industries Coal 2,048 2,366 2,695 3,189 3,689 4,910 Oil and gas 20,582 22,746 23,766 23,817 25,132 28,648 Mining and metal ores 191 209 239 281 244 439 Stones and other mining 1,759 2,015 2,398 3,039 3,597 4,306 Food and beverage 37,744 43,634 50,373 56,061 64,585 73,636 Cigarettes and tobacco 4,796 5,744 6,690 7,658 9,189 10,037 Textile products 8,388 10,046 10,641 12,338 14,214 16,415 Garment apparel - 5,218 6,042 6,862 8,182 10,466 12,709 Leather tanning and processing 7,725 8,851 9,529 11,096 13,535 15,976 Wood and wood products 3,180 3,598 3,903 4,488 5,485 6,646 Paper and paper products 3,470 3,930 4,562 4,877 5,655 6,697 Printing and publishing 2,012 2,274 2,453 2,876 3,515 4,746 Chemicals 9,682 11,123 12,852 14,714 16,323 18,578 Rubber products and plastic 5,427 6,456 8,128 9,706 11,291 13,260 Non-metallic products 14,785 18,259 21,625 25,913 29,855 34,709 Metalic production 5,000 5,914 6,842 8,516 10,430 12,063 Metallic products 5,036 5,768 7,063 8,506 10,646 13,274 Machinery and equipment 2,163 2,761 3,421 3,711 4,612 5,601 Computer and office equipment 1,703 1,295 977 1,003 1,538 1,918 Electric and electronic equipments 2,944 3,622 5,172 6,520 7,462 8,704 Radio, W and telecom 3,993 4,395 5,407 6,169 7,162 8,230 Production & repairing motor vehicles 1,846 3,232 4,265 5,774 8,306 8,282 Production & repairing other transport means 4,574 6,414 7,090 8,534 9,676 10,911 Furnitures 3,395 3,931 4,759 6,057 7 846 9,977 ~ Recycles products 127 150 151 174 204 241 Electricity and gas 9,496 11,828 13,551 15,741 18,071 20,656 Water supply 971 1,066 1,152 1,328 1,361 1,451 Source: GSO, Statistical Yearbook 2004. Table 8.2: MAJOR INDUSTRIAL PRODUCTS (VND billion, constant 1994 price) Rev Prel Product Unit 1999 2000 2001 2002 2003 2004 Assembled automobiles unit nla 13,547 20,526 29,536 47,701 42,651 Assembled motorbikes 000' unit 242 643 610 1,052 1,180 1,569 Assembled tivi sets 1,000 903 1,013 1,126 1,597 2,188 2,479 Beverage mil. liters 690 779 971 940 1,119 1,166 Bicycle tires 000 Pieces 18,326 20,675 21,656 22,778 26,686 27,000 Bicycle tubes 000 pieces 21,544 21,917 22,997 24,032 36,083 37,600 Bricks mil. pieces 7,831 9,087 9,811 11,365 12,810 14,501 Cement 000 tons 10,489 13,298 16,073 21,121 24,127 25,329 Chemical fertilizers 000 tons 1,143 1,210 1,270 1,158 1,294 1,453 Chromium ore 000 tons 59 76 70 66 91 95 Cigarettes mil. packs 2,147 2,836 3,075 3,375 3,871 4,065 Coal mil. tons 10 12 13 16 19 26 Crude oil mil. tons 15 16 17 17 18 20 Diesel engines Pieces 15,347 15,623 18,721 32,570 184,418 192,838 Electric engines Pieces 38,091 45,855 53,442 64,085 95,779 100,208 Electricity mil. kWh. 23,599 26,682 30,673 35,888 40,546 46,048 Fabrics of all kinds mil. meters 322 356 410 470 496 518 Fish sauce mil. liters 174 167 162 176 191 202 Glass products 000 tons 106 113 115 115 147 152 Insecticides 000 tons 21.9 20.1 20.0 20.7 40.9 44.3 Paper and paper products 000 tons 349 408 445 490 687 781 Porcelain mil. pieces 220 247 314 284 524 530 Rice mill equipment pieces 12,136 12,484 18,298 13,433 10,112 10,200 Salt 000 tons 653 590 699 974 909 880 Sawn wood 000 m3 1,466 1,744 2 036 2,667 3,291 ~ 4,115 Soap and detergent 000 tons 214 247 326 361 377 459 Steel 000 tons 1,375 1,583 1,914 2,503 2,954 2,929 Sugar 000 tons 947 1,209 1,067 1,069 1,360 1,371 Tea 000 tons 64 70 82 100 85 88 Textile fibers 000 tons 79 130 162 227 235 239 Tin (sticks) Tons 1,693 1,803 1,728 1,565 1,915 2,000 Transformers pieces 10,264 13,535 15,664 18,633 33,364 35,634 Water pumps for agri pieces 3,031 3,496 4,238 3,578 7,787 7,975 Source: GSO. Statistical Yearbooks 2004.