43653 Vietnam Development Report 2008 Social Protection Joint Donor Report to the Vietnam Consultative Group Meeting Hanoi, December 6-7, 2007 CURRENCY EQUIVALENTS CURRENCY UNIT = DONG US$ = 16,113 GOVERNMENT FISCAL YEAR January 1 to December 31 ACRONYMS AND ABBREVIATIONS ADB Asian Development Bank ASEAN Association of South East Asian Nations CHS Commune Health Station CIEM Central Institute for Economic Management CIT Corporate Income Tax CHS Commune Health Station DFID Department for International Development DRG Diagnostic Related Group EC European Commission GDC German Development Cooperation GDP Gross Domestic Product GSO General Statistics Office HCMC Ho Chi Minh City HCFP Health Care Fund for the Poor HEPR Hunger Eradication and Poverty Reduction Program HPAI Highly Pathogenic Avian Influenza ICA Investment Climate Assessment ILSSA Institute for Labor Studies and Social Affairs IMF International Monetary Fund LUC Land-Use Right Certificate MDG Millennium Development Goal MOET Ministry of Education and Training MOF Ministry of Finance MOH Ministry of Health MOLISA Ministry of Labor, Invalids and Social Affairs MONRE Ministry of Natural Resources and the Environment MPI Ministry of Planning and Investment NGO Non-Governmental Organization NME Non-Market Economy NTPH National Target Health Program PCI Provincial Competitiveness Index PLWD People Living with Disabilities SBV State Bank of Vietnam SEDP Socio-Economic Development Plan SFE State Forestry Enterprise SHI Social Health Insurance SOE State-Owned Enterprise TRIPs Trade Related Intellectual Property Rights VASS Vietnam Academy of Social Sciences VBARD Vietnam Bank for Agriculture and Rural Development VBSP Vietnam Bank for Social Policies VCCI Vietnam Chamber of Commerce and Industry VGCL Vietnam General Confederation of Labor VHLSS Vietnam Household Living Standards Survey VHW Village Health Worker VSS Vietnam Social Security WHO World Health Organization WTO World Trade Organization ACKNOWLEDGMENTS This report was prepared in partnership by the Asian Development Bank (ADB), the Department for International Development (DFID) of the United Kingdom, the European Commission (EC), the German Development Cooperation (GDC), and the World Bank. A major contribution by all these donors was to finance or conduct studies in areas related to the content of the report over the months and years preceding its writing. Such studies, listed in the bibliography section, were conducted as part of each donor's own country program. All these donors also contributed staff time to the preparation of the report, through their active engagement in workshops and discussions. The donors involved in this partnership also provided inputs and guidance to the overall effort through a Steering Committee comprising Kanokpan Lao-Araya (ADB), Kirsty Mason (DFID), Leo Faber (on behalf of the European working group on health), Maike Lerch (GDC), and Martin Rama (World Bank). 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 Ha Bui (Ministry of Planning and Investment - MPI), Son Kim Dang (Institute of Policy and Strategy for Agriculture and Rural Development - IPSARD), Dung Tran Nghiem (Ministry of Health - MOH), Cuong Manh Nguyen (Ministry of Labor, Invalids and Social Affairs - MOLISA), Huong Lan Nguyen (Institute of Labor Studies and Social Affairs - ILSSA), Huu Hai Nguyen (MOLISA), Long Hoang Nguyen (MOH), Thang Nguyen (Vietnam Academy of Social Sciences - VASS), Son Luong Pham (Vietnam Social Security - VSS) and Nga Thuy Tran (MOLISA). The team in charge of formally writing the report is affiliated with the World Bank. It was led by Martin Rama and included Noritaka Akamatsu (on chapter 11), Quang Hong Doan (6), Nga Nguyet Nguyen (4, 8, 9 and 10), Duc Minh Pham (5) and Carolyn Turk (1, 2, 3 and 7). Viet Tuan Dinh, Quy-Toan Do, Maria Delfina Alcaide Garrido, Severin Kodderitzsch, Samuel S. Lieberman, Ornsaran Manuamorn, Dzung The Nguyen, Minh Van Nguyen and Zen Delica Willson also provided specific inputs for various chapters of the report. The writing team benefited from a range of substantive contributions by colleagues in the broader development community. Special thanks go to Bob Baulch (Institute of Development Studies, Sussex), Paulette Castel (consultant), Ian Fox (consultant), Paul Glewwe (University of Minnesota), Ross Hughes (consultant), Dung Kim Le (Oxfam GB), Trung Dang Le (VASS), Adam McCarty (Mekong Economics), Bruno Moser (consultant), Tuan Anh Do Nguyen (IPSARD), Steve Price-Thomas (Oxfam GB), Mireille Razafindrakoto (Institut de Recherche pour le Développement- Développement, Institutions et Analyse de Long terme - IRD-DIAL), Francois Roubaud (IRD-DIAL), Jan Jung-Min Sunoo (International Labour Organisation), Huong Than Thi Thien (DFID), and Nicolaus Tideman (Virginia Tech University). Assistance with the analysis of data and documentation was provided by Cuong Viet Nguyen and Trung Dang Le (National Economics University). Nha Thi Vu (Vietnam Development Information Center - VDIC) was in charge of bibliographical research and referencing. The processing of the document was done by a World Bank team including Dung Thi Ngoc Tran on editing, Chi Kim Tran on publishing, and Phuong Minh Le and Ha Thi Thu Le on administrative support. Overall guidance was provided by Vikram Nehru and Ajay Chhibber from the World Bank. Robin Mearns, Ana Revenga and Adam Wagstaff, also from the World Bank, were the peer reviewers. The report was discussed with the International Monetary Fund (IMF). Comments and suggestions by numerous colleagues are gratefully acknowledged. TABLE OF CONTENTS Acknowledgements Executive Summary PART I: SOURCES OF VULNERABILITY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1 1. A Poverty Update . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3 2. Ethnic Minorities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13 3. Shifts and Shocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .22 PART II: POLICIES AND PROGRAMS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .33 4. Labor Markets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .35 5. Taxes and Fees . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .48 6. Budget Allocations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .58 7 Targeted Programs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .67 8. Health Care . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .74 9. Health Insurance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .84 10. Old-age Pensions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .96 11. Insurance Schemes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .105 PART III: THE WAY FORWARD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .113 12. An Integrated Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .115 Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .124 Statistical Appendix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .143 Boxes Box 1.1: Top Down, Bottom Up and in between . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .6 Box 1.2: Pockets of Severe Poverty in Rural Areas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .9 Box 1.3: What is Happening at the Upper End? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .12 Box 3.1: A Vastly Under-Reported Internal Migration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .24 Box 3.2: New Diseases: Who Is at Risk? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .28 Box 3.3: Anti-Dumping and Catfish Farmers in the Mekong Delta . . . . . . . . . . . . . . . . . . . . . . . . . . .32 Box 4.1: Towards a Better Monitoring of the Labor Market . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .37 Box 4.2: Trade Unions in Vietnam and in China . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .46 Box 5.1: How Costly Is It to Comply with Tax Regulations? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .53 Box 5.2: Attitudes towards Tax Evasion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .56 Box 6.1: Regional Plans, Infrastructure Master Plans and Public Projects . . . . . . . . . . . . . . . . . . . . . .65 Box 7.1: A Variety of Targeting Approaches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .68 Box 7.2: Linking Targeting Programs and Mainstream Social Insurance . . . . . . . . . . . . . . . . . . . . . . .70 Box 7.3: The Impact of Rural Roads . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .72 Box 8.1: Health Care at the Grassroots Level . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .78 Box 8.2: Drug Production, Commercialization and Pricing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .82 Box 9.1: Alternative Provider Payment Mechanisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .95 Box 10.1: Other Benefits in the Compulsory Program . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .98 Box 10.2: Participation in Voluntary Pension Programs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .100 Box 11.1: Private Pension Programs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .108 Box 12.1: The Limits of Mandatory Social Health Insurance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .119 Box 12.2: Managing Public Pension Reserves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .122 Tables Table 1.1: Poverty Rates and the Poverty Gap . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .4 Table 1.2: Poverty Rate across Regions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .7 Table 1.3: What Lies behind Changes in Urban Poverty? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11 Table 1.4: Share of Expenditures by Population Quintile . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .12 Table 2.1: Public Programs Are Reaching Out . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .17 Table 2.2: Size of Land Used by Ethnicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .17 Table 2.3: Access to Health Care by Ethnicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .19 Table 2.4: Access to Credit by Ethnicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .20 Table 3.1: The Economic Consequences of Health Shocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .29 Table 3.2: Who Is more Vulnerable? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .31 Table 4.1: The Structure of the Labor Force . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .36 Table 4.2: The Safety Net for Redundant SOE Workers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .41 Table 4.3: Minimum Wages and Average Wages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .43 Table 5.1: How Progressive Are Taxes and Fees? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .51 Table 5.2: Revenue Assignment by Level of Government . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .54 Table 6.1: Budget Allocation Norms for Provinces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .61 Table 7.1: A Growing Number of Beneficiaries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .69 Table 7.2: Coverage and Leakage of Targeted Programs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .71 Table 9.1: Average Spending by Type of Insured . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .92 Table 10:1: Social Insurance Revenue and Spending . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .103 Figures Figure 1.1: Consistent Decline but Different Levels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .5 Figure 1.3: Poverty Maps at the Province Level . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .8 Figure 1.2: Accelerated Regional Convergence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .8 Figure 2.1: Household Expenditures: A Two-wave Story . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .14 Figure 2.2: Diversity among Ethnic Groups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .15 Figure 2.3: Same Remote Areas, Different Histories . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .16 Figure 2.4: Educational Attainment by Ethnicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .18 Figure 3.1: A Rapidly Ageing Population . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .23 Figure 3.2: Disaster-Prone Areas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .25 Figure 3.3: Going under Water? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .26 Figure 3.4: The Evolving Causes of Mortality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .27 Figure 4.1: Determinants of Labor Earnings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .39 Figure 4.2: Who Are the Unemployed? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .40 Figure 4.3: Number of Strikes by Enterprise Ownership . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .45 Figure 5.1: Who Pays Taxes in Vietnam? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .49 Figure 5.2: Tax Revenue by Source . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .50 Figure 5.3: The Tax Burden on Enterprises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .52 Figure 5.4: Negotiating with Tax Officials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .57 Figure 6.1: A Breakdown of Budget Expenditures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .59 Figure 6.2: Not Overburdening Future Generations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .60 Figure 6.3: How Large are Equalization Grants? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .62 Figure 6.4: Net Transfers and Poverty Rates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .64 Figure 8.1: Child Mortality in International Perspective . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .75 Figure 8.2: Infant Mortality across Regions in Vietnam . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .76 Figure 8.3: Quality of Care to Children with Diarrhea . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .77 Figure 8.4: Health Expenditures in International Perspective . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .79 Figure 8.5: Mostly from Out-of-Pocket Payments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .80 Figure 9.1: A Varying Degree of Enterprise Compliance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .85 Figure 9.2: A Rapid Expansion in Coverage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .86 Figure 9.3: Targeting and Leakage in Decision 139 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .87 Figure 9.4: Vast Disparities across Regions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .88 Figure 9.5: Health Care Contacts by Group of Insured . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .90 Figure 9.6: Costs and Revenues of the Health Insurance Program . . . . . . . . . . . . . . . . . . . . . . . . . . . .92 Figure 9.7: Sources of Hospital Income . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .93 Figure 10.1: Working Less, Getting Sick more Often . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .97 Figure 10.2: Coverage of the Compulsory Pension Program . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .99 Figure 10.3: Households Receiving Pension Income . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .101 Figure 10.4: Pension Benefits and the Actuarial Benchmark . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .102 Figure 10.5: The Reserves of the Pension Program . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .104 Figure 11.1: A Steady-Growing Sector . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .106 Figure 12.1: The Cost of Subsidizing Participation in the Pension Program . . . . . . . . . . . . . . . . . . .116 Figure 12.1: A Fragmented Approach to Information Management . . . . . . . . . . . . . . . . . . . . . . . . . .121 The popularity of the term "social protection" is evidenced in the recent literature and policy debate. [...S]ocial protection has however fairly recently come into extensive use in developing countries. The concept remains confusing mainly due to the range of existing definitions and the variety of ways in which it is interpreted. Since the late 1990s, a growing concern with the global dimensions of social policy has been apparent. A majority of agencies take an action approach to social protection policies, seeing it as a collection of measures to manage risks and improve or protect livelihoods. [...T]here is an outer circle of promotional measures, which includes mainly macro policies, development strategies and institutional measures. They help enhance the social status and empower the rights of the marginalized and the disadvantaged. Preventive measures, which make up the middle circle, are generally made up of direct measures for averting deprivation. This typically refers to a range of social insurance and other services which protect people from falling into crisis and requiring public assistance. Lastly, the inner circle of protective measures consists of narrowly targeted safety nets which aim to provide the vulnerable with in-kind and cash donations, or other short-term public assistance and relief from shocks (e.g. natural disasters, animal disease, family illness). [...] The three components can overlap and interact, but are not identical in practice. [...T]hrough the discussion held so far on the topic and the Government's activities in this area, both social protection researchers and practitioners seem tentatively not to object to define social protection as (i) direct assistance to poor and vulnerable household, (ii) social insurance and (iii) other activities aimed at reducing the vulnerability caused by such risks as unemployment, aging and disabilities. Le, Bach Duong, and others. 2005. Social Protection for the Most Needy in Vietnam. The Gioi Publishers, Hanoi. EXECUTIVE SUMMARY Vietnam will soon join the group of middle- establish mechanisms to preserve the inclusive income countries. In a simple, mechanical nature of its development so far, at a time way, this means that its income per capita will when rapid economic growth could easily cross some more or less arbitrary threshold. result in increased hardship and inequality. At a deeper level, becoming a middle-income country entails harnessing new, more complex Vietnam is at the dawn of unprecedented policy challenges. Two decades of Doi Moi economic and social transformation. The have sustained a spectacular transformation of average citizen is a mostly healthy 26 years Vietnam, from being a poor agrarian economy old who lives in a rural area, belongs to a to becoming a globally integrated player, from multi-generation household including grand- being run through command and planning to parents and works in agriculture or does thriving on market mechanisms. The main informal activities, such as being a casual structural reforms have been adopted, or are in laborer or helping in a household business. the process of being implemented. Markets Because of a rapid demographic transition, in have been opened for competition, or created a few decades population will be much older altogether. A hard budget constraint and on average. And because of rapid rural-urban private initiative have been brought into migration, it will have moved too. By then public enterprises. The banking sector is this average citizen may have health concerns, being reorganized and capital markets are will most probably have migrated to a city, being developed to support an efficient could belong to a nuclear household including mobilization of resources. While much no grand-children (and perhaps no children), remains to be done in order to consolidate will most probably be part of the formal these transformations, the general sense of economy as a wage employee or a direction is clear. But a second generation of businessperson, and will not be so far from reforms is increasingly needed. retirement. As Vietnam enters the ranks of middle-income How well will the institutional arrangements countries, it needs to layout the foundations of of Vietnam support this transformation? How a more sophisticated economy. It has to can they contribute to making the personal sustain productivity growth in the long term, journeys of millions of citizens become moving up towards more skill-intensive, stories of prosperity and accomplishment, not knowledge-based activities. It should avoid of hardship and uncertainty? Jobs will be draining down its natural resources and needed for people to move out of agriculture damaging its environment in ways that could and informal sector activities. Local be irreversible or too costly for future development will be required to boost the generations to redress. It needs to increase living standards of those who do not migrate. transparency and accountability in public Social services should be accessible for those administration, avoiding the waste of who do, and may need them away from home. corruption and its potential damage to public Health care should be affordable and of good confidence and morale. And it needs to quality. Instruments should be available for i SOCIAL PROTECTION people to save towards their old age. And groups, as they experience hardship even in protection mechanisms should exist to cope good times. with multiple risks along the way, from natural disasters to anti-dumping actions by This report aims to contribute to the Vietnam's trade partners to catastrophic health discussion of appropriate social protection shocks. strategies for Vietnam. It does so in the understanding that this is a new area, closer to Targeted programs aimed at the most the second-generation reforms faced by vulnerable will not be enough. Targeted middle-income countries than to the structural programs have not been a critical component reforms Vietnam has successfully gone of the government strategy anyway. They through over the last two decades. It also does have aimed at helping the neediest households so in the understanding that a common vision to cope with the costs of social services and to exists among policy makers, academics and gain access to credit. They have also the population at large. But at this early stage supported the most disadvantaged communes the specifics remain unclear, often resulting in to integrate with the market economy. But frustration in the daily implementation of they have never represented a major budgetary social policies and programs. outlay, and only reach a fraction of the poor. The strategy of the government has relied The vision is one of shared and inclusive more on economic growth, the redistribution growth, in the socialist spirit of Vietnam. of land and reasonably good health and Based on the success of Doi Moi, it involves education. And when it comes to social relying on market mechanisms to the extent policies, the stated ambition of the possible, to ensure an efficient allocation of government is to build universal programs. resources in the short term and sustain From this perspective, targeted programs can innovation and growth in the longer term. But become important in supporting the it also entails mobilizing government resources participation of the poor and the vulnerable in to finance the participation of the poor and mainstream social programs. But mere social other vulnerable groups. And it requires assistance will not do. market regulation and policy stewardship, to ensure that everybody has access to services However, the policy agenda of a country at and quality standards are met. Vietnam's development level cannot be to mimic the social protection systems of Because the vision is clear, the report focuses industrial countries either. Poverty remains a on the specifics. While its scope is wide, the serious issue. It continues to decline rapidly topics covered can be regrouped into three in rural areas, but for the first time it seems to main areas: opportunity to thrive, support for have stagnated, perhaps even increased, in those left behind and insurance to cope with urban areas. The higher cost of living in shocks. booming economic hubs may start to take a Opportunity. Job creation has been one of the toll on those whose means are barely above main avenues out of poverty in Vietnam. A the poverty line. In rural areas, the plight of flexible labor market has allowed absorbing ethnic minorities deserves special attention. almost one and a half million new entrants to Many in this population group are bound to the labor force, year after year. Wage "cross" the poverty line in the coming years, employment is particularly important in this which will result in continued poverty respect. Regardless of whether this is ideal or reduction overall. But they will still be barely not, access to social programs, from health above the line for the foreseeable future. And insurance to old-age pensions, is easier for meanwhile, many among the ethnic minorities those with formal sector jobs. Sustaining will still be suffering from deep poverty. A rapid job creation, especially in the formal policy agenda focused on managing risks and sector, is crucial to keep growth inclusive. shocks only would fail to deliver to these But success should not be taken for granted. ii EXECUTIVE SUMMARY The implementation of commitments made to redistribute resources through Personal accede to the World Trade Organization Income Tax (PIT), they run the risk of (WTO) will require that the minimum wage penalizing the formalization of economic faced by domestic private enterprises be raised activity, without mobilizing much revenue. rapidly over the next few years. The Taxes on valuable urban land bear a much temptation to balance the books of the social higher potential to redress inequality and to insurance agency through higher social raise resources for local infrastructure security contributions will only add to the development, without distorting incentives burden faced by employers. Meanwhile, the much. But they would require a substantial industrial relations system in force is not modernization of both land and tax conducive to decentralized wage bargaining at administration. the enterprise level and exacerbates labor conflicts. Moving out of such an outdated Insurance. Households face a range of system could also help Vietnam exit the Non- shocks, some of which may have devastating Market Economy (NME) status it was granted effects on their livelihoods. Natural disasters, when entering the WTO. This, in turn, would crop failure, serious health conditions, the help confront protectionist anti-dumping death of a breadwinner or longevity beyond actions by Vietnam's trade partners, which are resources are among them. Insurance markets another major threat to jobs. have developed considerably in Vietnam, and may soon extend their reach to areas such as Support. Opportunities do not arise health, pensions and agriculture. However, spontaneously, and not everybody has access for their coverage to be substantial and for the to them. Deliberate efforts are needed to risk of default to be contained, the government connect communities to markets and to needs to provide adequate regulation. In some develop local infrastructure, so that businesses cases, like flood-based insurance, it may need can thrive. In a country as decentralized as to lay the foundations of a market through the Vietnam, the amount of budgetary resources establishment of verifiable indicators and available to a province has a decisive hazard maps. In other cases, regulation will influence on its capacity to deliver social just not be enough. Health insurance faces services, implement social policies and well-known adverse selection problems, finance infrastructure development. A set of resulting in the exclusion of high-risk transparent budget allocation norms has been populations. Old-age pensions suffer from a set up to transfer resources from richer to free rider problem, whereby people may not poorer provinces, on a scale that dwarfs save enough in the expectation that similar efforts in much richer regions of the government will provide support when the world. But there has been less progress in the time comes. The organization of the social allocation of provincial resources to districts, insurance system tries to address these communes and below. As a result, there are problems, but in doing so it also introduces marked gaps between funding and needs. other distortions. In the end, implementation Programs targeted to specific households and details can make all the difference. Voluntary communes are of mixed quality, with some health insurance or pension programs may not having a considerable impact and others being be sufficient to address the corresponding clearly less effective. Redistribution of market failures and substantially expand resources also happens through taxation, an coverage. Provider payment mechanisms or area currently undergoing a dramatic benefit packages may not elicit a cost- overhaul. Overall, the taxes paid by effective delivery of services. The regulation households are progressive in Vietnam, but of providers may result in some population this is not true of local fees. And taxes paid by groups (the poor in particular) being denied enterprises seem to put a heavy burden on service, or at least quality service. The smaller units, while entailing high compliance complexity of these problems is what makes costs. As for the current attempts to the social protection agenda part of a second, iii SOCIAL PROTECTION technically more demanding generation of improving information and managing policy reforms. resources. In reviewing the situation and challenges in Extending coverage. While the vision relation to opportunity, support and insurance underlying the social protection agenda is one the current report tries to provide analytical of shared growth and inclusion, Vietnam is input for important policy debates ahead. quite far from anything resembling universal During the remaining of this five-year period, coverage for its social programs. In the case the National Assembly will most probably of health insurance, the initial target was to pass a Law on the Minimum Wage and revise get there by 2010. By now close to half of the the Labor Code, especially on industrial population is covered, which is certainly an relations. A Law on Social Assistance may accomplishment for a l?-income country. But lead to the revamping of targeted programs. subsequent gains could prove increasingly Tax policies and tax administration will be difficult. Coverage is even more limited in subject to a major overhaul, with new laws relation to pensions, raising the prospect of passed for almost every major tax instrument, large liabilities when the young of today including property tax. A Law on Health reach retirement age, if budgetary resources Insurance will be discussed in 2008, while a need to be mobilized to pay for non- new voluntary pension program will be contributory pensions. Targeted programs do implemented from 2009 onwards. A decree on not reach all of their intended beneficiaries private pensions will also be passed as part of either; far from that. the agenda for capital markets development. Meanwhile, the adoption of a Natural Disaster Implementing the vision will require bolder Management Strategy raises the prospect of steps. The full subsidization of participation developing new market-based insurance by the poor in mainstream programs, already instruments. adopted in the case of health insurance, opens a promising avenue. Something similar could The report is organized under the form of be done in the case of pensions, where a "vertical", self-standing chapters dealing with minimum contribution towards retirement and each of the key issues to be addressed by the some basic benefits could be paid by National Assembly and by policy makers in government to the social insurance agency on government during the next few years. These behalf of each poor household head. A chapters can be seen as brief policy notes, necessary complement to this approach is the trying to provide an analytical structure for the partial subsidization of participation by the discussion. Because of their brevity, they near-poor. Without it, the incentive for cannot go in depth into each of the aspects households to escape poverty could be covered, and can only sketch the findings of reduced and enrollment could decline among more rigorous and detailed studies. However, those who do escape. There is also a need to for those interested in digging deeper, the delimit the extent of voluntariness. Group report includes an extensive bibliography, insurance (involving all household members mainly made of references produced by local or a minimum fraction of all people in a thon) researchers and think tanks. can help address adverse selection in the case of health. Mass organizations could be The report also tries to extract a few major mobilized to support enrollment into the policy recommendations, cutting across all the voluntary pension program. "vertical" chapters. These recommendations amount to basic principles which could help One important question is whether a strategy build a more integrated social protection involving the subsidization of a large fraction system in Vietnam. They can be summarized of the population can be afforded. The answer under the form of four policy priorities: is probably yes. To begin with, part of the extending coverage, strengthening incentives, subsidization simply entails shifting current iv EXECUTIVE SUMMARY government funding from providers (say, actuarial benchmark are important priorities in public hospitals) to purchasers of services (the relation to social insurance. social security agency, or a dedicated fund for chronic illnesses and life-threatening Some of the most complex incentive issues diseases). In the case of pensions, the might actually be faced in rural areas. alternative to the subsidy is not savings: it is a Sustaining the emergence of flood insurance contingent liability, as something will need to and other market-based tools to deal with be done sooner or later to support old people natural disasters will not be easy. Without a without income. The truly additional proper institutional set up, involving cadastral spending can be contained as long as policies maps, hazard modeling and verifiable are in place to support a rapid reduction in information, insurers may simply not be poverty. interested. In other areas, such as agricultural extension, finding the right incentives may Strengthening incentives. Increased reliance involve a much deeper understanding of on market mechanisms, one of the foundations cultural differences between ethnic minorities of Doi Moi, can certainly lead to large and the majority group. The current offer of efficiency gains. The improvement in the services is based on models which have quality of care provided by hospitals operating worked well in the lowlands but may be of in an autonomous manner is a case in point. little relevance for swidden cultivation on However, in sectors which are plagued by sloping plots of land. market imperfections, undesirable outcomes are common as well. Hospitals may become Improving information. The vision of an good at catering to those who can pay, but they integrated approach to social protection is at can also neglect the mandatory provision of odds with the current fragmentation of free services to the poor. Improving the information systems. Targeted programs use monitoring of health care providers is thus an the classification of households by poverty important priority. status, conducted by local officials of MOLISA. Health insurance and old-pension The mechanisms used by the government to programs run on separate information regulate these sectors can be a source of systems, focused on individuals rather than distortions as well. The payment schedules households. All of these programs rely on the used by the social insurance agency to place where the household or the person lives, reimburse health care providers can encourage or is supposed to live, through a system of excessive diagnosis and treatment, resulting in residential registration which often restricts an unwarranted escalation of health care costs. access to benefits in other localities. Such The benefit formulas used for the old-age system seems ill-adapted to a country where a pensions program affect the incentives to save large fraction of the population will be moving and to participate in the system. Pensions are from rural to urban areas in the coming years. quite far from what an actuarially fair system would provide. Government employees and The fragmentation of information systems is women get more, out of the same stream of such that the ministries in charge of contributions, than private sector workers and formulating policies cannot effectively men do. After 20 years of contributions, the monitor the behavior of the populations they incentive to remain enrolled starts declining are supposed to serve. MOH has difficulty and it is in the interest of participants to under- figuring out how health care services are used declare their earnings. The emergence of by different participants in the health private pension funds will require that insurance program. MOLISA knows little attention be paid to the portability of benefits about the typical jobs and earnings transitions across programs. Introducing better-designed of contributors towards old-age pensions. provider payment mechanisms in health and Because of this fragmentation, participants bringing the pension program closer to the can easily be denied service. Current, v SOCIAL PROTECTION scattered attempts to introduce modern communes do receive the resources they are information technology, through the use of entitled to and manage them properly. It will bank cards for pensions or chip cards for be necessary to monitor that those who are health, may add layers of complexity to an classified as poor get their social security already difficult problem. contributions totally or partially paid for out of public resources. The purchasing decisions The main recommendation in this area is to of the social insurance agency will involve introduce a single social insurance number, to hundreds of thousands of providers, including be given to every person at birth regardless of public hospitals, private clinics, facilities of coverage. Unlike current registration cards, non-governmental organizations (NGOs) and such number allows the full portability of certified doctors. Scrutiny will be required to benefits throughout the country. The poverty avoid abuse or favoritism, as in any other status of the household the person belongs to public procurement decision. can be updated on an annual basis. And the number can serve to document personal An area deserving special attention is the histories in connection to employment, management of the reserves of the old-age earnings and health care use, among others. pension program. Those reserves already With adequate information management exceed the equivalent of 4 billion dollars. But systems, these personal histories can help they are placed in bonds and liquid assets design better provider payment mechanisms in yielding low returns. There is a need for health, and eventually support the introduction strategic decisions as to what the authorized of individual accounts in the case of pensions. investments are, and how portfolio A single social insurance number could also reallocation decisions are going to be made. limit the denial of service often observed at Transparent mechanisms could also be present. considered to report to the National Assembly on the financial performance of social Managing resources. Delivering on the social protection programs, including the protection agenda also requires solid financial management of reserves, perhaps on an management. Targeted programs involve annual basis. transferring resources to thousands of communities and millions of households. Some may question whether these four Extending the coverage of social programs policy recommendations address the most will require some form of subsidization for the important policy priorities faced by Vietnam. poor and perhaps the near poor, with the It can be argued that the strengthening of budget covering the contributions of a monetary and financial policies, the significant portion of the population, or development of infrastructure and the supporting provinces to do so. The modernization of tertiary education are more development of the health insurance program urgent at this point. And it is true that less amounts to shifting an increasingly large share hinges on effective social protection and of the funding for health care from public inclusive growth in the short term. providers to wholesale purchasers of health However, this agenda may still be of critical care services. The pension program, in turn, is importance for the medium term. While bound to accumulate a massive amount of many developing countries have made it to reserves during several decades. This is middle-income level, very few have because the population is young and the labor market is formalizing, so that dependency continued to become industrial nations. ratios will remain low. Handling the social protection agenda well could hold the key for a successful transition Transparent financial management will be to the next level a few decades down the crucially important in all these areas. There road. To get there, it is better to make the will be a need to make sure that targeted right policy choices now. vi PART I: SOURCES OF VULNERABILITY 1. A POVERTY UPDATE Household survey data from 2006 confirms the low for an economy growing as fast as Vietnam, continued reduction of poverty in Vietnam, with and might even have declined slightly. However, the fraction of households living below the household expenditure surveys may not be well- poverty line attaining 16 percent, compared to suited to capture a rapid increase in wealth at the 28.9 percent in 2002, and 58.1 percent in 1993. top of the distribution. Other estimates, using different methodologies, yield an equally encouraging picture. But Main trends progress is uneven. Poverty remains much higher among the ethnic minorities than among Poverty rates continue to fall in Vietnam. One the Kinh and Chinese majority. Most of the poor way to assess the speed of this decline is based live in rural areas, and it is encouraging that rural on data on household expenditures from five poverty continues to decline, albeit at a slower surveys on living standards conducted by the pace than before. On the other hand, urban General Statistics Office (GSO) in 1993, 1998, poverty rates appear to have stagnated, and could 2002, 2004 and 2006. One international even be for the first time on an upward trend. methodology to produce a number of indicators The combination of these diverging urban and that are consistent over time is to compare rural trends is probably behind the convergence household expenditures per capita with a poverty in poverty rates observed across regions and line. The latter is defined as the cost of a food provinces. The Northwest, the Central and non-food consumption basket allowing a Highlands and the North Central Coast remain healthy life. In the case of Vietnam, it has been much poorer than the rest of the country. But customary to consider the poverty line as the cost even in these regions some provinces and groups of a basket allowing a daily intake of 2,100 are witnessing important declines in their calories per person per day. Based on this poverty rates. Understanding why poverty measure, it appears that poverty has fallen remains high in specific areas requires a closer dramatically between 1993 and 2006, both in analysis of local circumstances, as shown by the rural and in urban areas and for both Kinh and for diversity of experiences across poor rural ethnic minority groups (Table 1.1). According to communities. As for urban poverty, the rate these estimates, over a 13 year period Vietnam could be somewhat inflated due to statistical has seen 42 percent of the population (equivalent biases, such as the better sampling of migrants in to about 35 million people) move above the recent surveys and the re-classification of wards poverty line. In 2006, 16 percent of the at the fringe of cities as urban. However, these population remained below the poverty line. biases do not explain everything. It is likely that rapidly growing prices of goods and services, Rural areas still account for the bulk of the poor possibly associated with much higher land in Vietnam, but it is the rapid decline in rural prices, are pushing households that were barely poverty that is driving poverty reduction at making it back into poverty. As for inequality, national levels. Between 2004 and 2006, the standard indicators suggest that it remains quite rural poverty rate fell by 2.3 percentage points 3 SOCIAL PROTECTION Table 1.1: Poverty Rates and the Poverty Gap 1993 1998 2002 2004 2006 Poverty rate 58.1 37.4 28.9 19.5 16.0 Urban 25.1 9.2 6.6 3.6 3.9 Rural 66.4 45.5 35.6 25.0 20.4 Kinh and Chinese 53.9 31.1 23.1 13.5 10.3 Ethnic minorities 86.4 75.2 69.3 60.7 52.3 Food poverty 24.9 15.0 10.9 7.4 6.7 Urban 7.9 2.5 1.9 0.8 1.2 Rural 29.1 18.6 13.6 9.7 8.7 Kinh and Chinese 20.8 10.6 6.5 3.5 3.2 Ethnic minorities 52.0 41.8 41.5 34.2 29.2 Poverty gap 18.5 9.5 6.9 4.7 3.8 Urban 6.4 1.7 1.3 0.7 0.7 Rural 21.5 11.8 8.7 6.1 4.9 Kinh and Chinese 16.0 7.1 4.7 2.6 2.0 Ethnic minorities 34.7 24.2 22.8 19.2 15.4 Source: Based on preliminary GSO data. Estimates for 2006 are unofficial. per year, compared to 3.5 points between 1993 expenditure level associated with the poverty and 2004. While this reflects a slowdown, the line. This measure is used to describe whether overall trend remains strong. By contrast, the poverty is shallow, with most poor people being urban poverty rate appears to have stabilized, quite close to the poverty line, or deep. or to have even increased slightly. Confidence intervals are such that no clear trend can be While falling rapidly, food poverty still affects inferred from the data in this case. And to nearly nine percent of rural households, and an some extent, poverty rates are bound to astounding 29 percent of ethnic minority stabilize as they get closer to zero. However, households. The good news is that poor people the slight slowdown in poverty reduction in across all population groups are moving closer rural areas and its stagnation in urban areas to the poverty line, with the poverty gap falling deserve closer analysis. from 6.9 percentage points in 2002 to 3.8 percent in 2006. Even among ethnic minorities, Other commonly used indicators refer to food poverty and the poverty gap. A household is the poverty gap in 2006 was comparable to that said to be food poor when its expenditures are of the Kinh and Chinese majority in 1993. so low that even if they were entirely allocated to buying food that would not be enough to Other metrics attain 2,100 calories per person per day. It is clear that even the poorest households spend on As for all issues related to measurement, non-food items. Therefore, a food poor methodology matters. In Vietnam's case, at household can be expected to suffer from least two other ways of measuring poverty hunger, at least during some periods of the year. receive considerable attention. One of them The poverty gap is the average "distance" considers a poverty line set at one dollar per between the expenditures of poor people and the person per day. The main point of using such 4 A POVERTY UPDATE poverty line is to facilitate comparisons with questioned. More importantly, PPP factors are other countries. However, prices also differ as difficult to estimate as household considerably across countries, so that the expenditures. Because they are subject to purchasing power of the dollar needs to be occasional revisions, it is not uncommon to adjusted as well. This is usually done through see unexpected jumps in poverty rates, in one a correction factor called Purchasing Power direction or another. Parity (PPP), which seeks to find the equivalent of one dollar in the US. As of Other commonly reported poverty figures for 2006, the PPP factor for Vietnam was around Vietnam are produced by MOLISA, using an 3.36, which means that 30 cents can buy alternative methodology. Again, the underlying roughly the same as one dollar in the US. principle is to compare some indicator of household living standards with some minimum On the one-dollar-a-day metric, poverty in acceptable benchmark. But both household Vietnam has declined at the same rapid pace living standards and the poverty line are suggested by the 2,100-calories metric (Figure measured in ways that differ from the 1.1). As of 2006, only 4.9 percent of the international methodology. population was below this poverty line. This can be interpreted as meaning that only 4.9 The poverty line used by MOLISA was percent of the population was poorer than originally conceived as a certain amount of rice. someone living on one dollar per day in the Back in 1993, people were considered poor if US. Beyond supporting international they could not afford 20 kilos of rice per person comparisons, the usefulness of this per month in urban areas, and 15 kilos in rural information to local policy makers can be areas. These amounts were increased to 25 and Figure 1.1: Consistent Decline but Different Levels Source: Based on data from GSO, MOLISA and World Bank. 5 SOCIAL PROTECTION Box 1.1: Top Down, Bottom Up and in between The poverty lines estimated by GSO rely on a painstaking measurement of expenditures at the household level. It takes two days for a respondent to fill in the questionnaire of a living standards survey, and considerable training, monitoring and checking is required to ensure that the data are of high quality. Not surprisingly, the samples of such surveys tend to be small. The first Vietnam Living Standards Survey (VLSS) for Vietnam, conducted with funding by the United Nations Development Program with technical assistance from the World Bank included 4,800 households, on a nationally representative basis. With 6,000 households covered, the sample of the VLSS 1998 was only slightly larger. A major expansion took place with the 2002 Vietnam Household Living Standards Survey (VHLSS), when 30000 households were surveyed. At 9,189, sample sizes were also quite large for 2004 and 2006 VHLSSs. However, sample sizes remain too small for really accurate provincial estimates (the errors are quite large) and certainly too small to measure poverty at lower levels of administration. Building on data from household surveys and population censuses, statistical methods known as poverty maps allow estimating defensible poverty rates at the district and perhaps the commune level. But in a country characterized by rapid growth and massive rural-urban migration, population censuses become quickly outdated and so do poverty maps. The MOLISA approach, on the other hand, deals with almost all poor and near-poor households in the country, one by one. Every year the list of poor households is updated in each community or thon. This involves some discussion as to which households might have moved out of poverty and which ones might have fallen back in it. A careful assessment of local household classifications was conducted across 41 communities all over Vietnam, relying on wealth-ranking exercises, using the same methodology in all of them. Wealth-ranking is a standard participatory approach derived from applied anthropology. Households in these communities were also subject to more quantitative techniques, in the spirit of those used by GSO to measure household expenditures. In statistical jargon, these techniques are known as means testing and proxy-means testing, among others. It appeared that the classification by MOLISA officials outperformed all the statistical approaches, both within each community and across communities. However, poverty rates computed based on the MOLISA classification did not perform well at ranking communities. It is clear that aggregating the classifications further up, at district or province level, can only result in unreliable poverty rates. With the GSO metric being reliable at aggregate levels and the MOLISA metric being better at commune levels, the main question is how to reconcile them at the middle ground. One of the most obvious steps is to use the same poverty lines in both cases. The revision of the MOLISA poverty lines in 2005 is an important step in this direction, as the new thresholds are much closer to the 2,100 calories benchmark. However, reconciling bottom- up poverty figures, obtained by aggregation over local MOLISA poverty rates, and top-down figures, from household surveys and poverty maps, will not be an easy task. MOLISA is currently undertaking research into how targeting procedures can make best use of the comparative strengths of the methodologically rigorous GSO estimates and the more disaggregated, local estimates. For some time, however, there still may be conflicting information on poverty rates at the district and commune levels. Source: Based on information from GSO, MOLISA, and Nga Nguyet Nguyen and Martin Rama (2007). 20 respectively in 1995, and a threshold of 15 To determine which households are below the kilos was introduced in 1997 for mountainous poverty line, local officials from MOLISA use a areas. In 2001 rice amounts were converted into combination of survey techniques and village money terms, at 150, 100 and 80 thousand dong discussions. They also rely on their personal per person per month, respectively. A major knowledge of the families in the community update took place in 2006, when the poverty line they oversee. The result is a list of poor was set at 260 thousand dong for urban areas and households, who are then targeted as 200 thousand dong for all rural areas. beneficiaries under government programs 6 A POVERTY UPDATE Table 1.2: Poverty Rate across Regions 1993 1998 2002 2004 2006 Northern Mountains 81.5 64.2 43.9 35.4 30.2 Northeast 38.4 29.4 25.0 Northwest 68.0 58.6 49.0 Red River Delta 62.7 29.3 22.4 12.1 8.8 North Central Coast 74.5 48.1 43.9 31.9 29.1 South Central Coast 47.2 34.5 25.2 19.0 12.6 Central Highlands 70.0 52.4 51.8 33.1 28.6 Southeast 37.0 12.2 10.6 5.4 5.8 Mekong Delta 47.1 36.9 23.4 15.9 10.3 Vietnam 58.1 37.4 28.9 19.5 16.0 Source: Based on preliminary GSO data. Estimates for 2006 are unofficial. designed to reduce poverty. The process through the Red River Delta and the Southeast, on the which the lists are compiled varies across the other hand, are seeing a marked slowdown. 10,000 or so communes in Vietnam, which This slowdown is all the more notable, since it somewhat affects the comparability of the follows a period of strong and sustained poverty estimated poverty rates. Analysis suggests that reduction. most of those included on the lists were indeed The combination of fast declines in poverty poor, but that, at least until the poverty line was rates in the poorer areas of the country and increased in 2006, the lists often missed many slower declines in the richer ones is resulting in other poor households. a narrowing of the gap between regions. In an The co-existence of the GSO and MOLISA encouraging sign of convergence, poverty has measures of poverty has led to confusion in the fallen more rapidly in areas where its incidence past, with some analysts preferring one metric was initially highest (Figure 1.2). Statistically, to the other. But both approaches have strengths between 1993 and 2002 a region whose poverty rate was 1 percent higher than the average and weaknesses, making the two metrics experienced an annual decline that was 0.037 complementary to some extent (Box 1.1). percentage points faster than the average. Between 2002 and 2004, the decline was 0.082 Poverty and geography percentage points faster. And the fit of the statistical relationship also became stronger Regional differences remain wide, with the over time. mountainous areas being much poorer than the lowlands, and especially than the Southeast The picture is similar when provincial poverty (Table 1.2). But there are also important rates, instead of regional poverty rates, are differences in the speed at which poverty is considered. There are still dramatic gaps falling. The poorest region of the country, the between the poorest and the richest provinces in Northwest, has reduced poverty by 19 the country (Figure 1.3). But even in the poorest percentage points over the last four years and highland areas some provinces are doing the Central Highlands by a staggering 23 considerably better and starting to resemble their percentage points. The two least poor regions, richer counterparts in the lowlands. 7 SOCIAL PROTECTION Figure 1.2: Accelerated Regional Convergence Source: Based on preliminary GSO data, with one observation per region and per inter-survey period. Estimates for 2004 - 2006 are unofficial. Figure 1.3: Poverty Maps at the Province Level Source: Based on preliminary data from GSO. Estimates for 2006 are unofficial. 8 A POVERTY UPDATE Underlying forces cities of Vietnam have been booming for many years now, and yet urban poverty appears to Rapid improvements in wellbeing should not have stabilized, if not increased. It must be deflect attention from the fact that there may noted that the sample of the household survey is still be around 13.5 million people living in such that less than 100 urban households in it poverty in Vietnam, with between 5 and 6 are poor. Under these circumstances, million of them being food poor. One of the movements of the poverty rate by less than one challenges for the medium term is to locate percentage point are not statistically significant. those places and populations where poverty Yet, any increase is bound to raise questions in persists, and to understand the reasons why they a country where measured poverty has fallen are not benefiting from economic growth as the consistently and rapidly over the past 13 years. majority of the population is, so as to reach Understanding why this is so is especially them with the kind of support that stands the important to adjust social protection policies in highest chances to be effective. Small-scale a period of rapid urbanization. surveys from across the country show the diversity of circumstances under which rural Admittedly, the stabilization of urban poverty poverty persists (Box 1.2). could just be a statistical artifact. GSO has been making consistent efforts to better capture rural- While it is clear that poverty is largely rural in urban migrants in the sample of its household nature, the figures for 2006 suggest that survey. This group is poorer than well- economic growth alone may not be sufficient to established urban dwellers. If their relative deal with poverty in urban areas. The major weight had increased in the 2006 sample, urban Box 1.2: Pockets of Severe Poverty in Rural Areas In Ninh Thuan province, research that is part of an ongoing participatory poverty monitoring exercise engaged with poor communities that have seen little positive change in wellbeing for the poor. Asked about their quality of life over the previous 12 months, a vast majority of poor households reported either no change (51 percent) or deterioration (33 percent). Meanwhile, twice as many non-poor households than poor households reported an improvement in the quality of life. Households blamed dependence on an unpredictable weather for continuing poverty or worsening conditions. Kong Chro is one of the poorer districts of Gia Lai Province. Researchers visiting one of the district's 13 communes were impressed by the outreach of government services. A paved road led to the commune centre, health care cards had been allocated, houses built, land distributed and water supply systems constructed. However, conditions of life remained very difficult for Ede people. One family of eight people visited was subsisting on two hectares of land where they cultivated maize and cassava. Besides raising a pig and few chickens, the family was entirely dependent on income from agriculture. Despite the well-constructed road, this family remained disconnected from the economic growth that is encouraging farmers in other parts of the country to diversify their income sources. Pockets of poverty are not confined to the mountains. The Red River Delta region has seen dramatic declines in poverty over the last ten years. But even there it is possible to remain marginalized from the growth process. A coastal commune in Thai Binh province has a number of households that are prospering through investments in shrimp farms. But other households in that same commune have so little land that they are dependent on harvesting shellfish and tiny shrimp that live in the mud flats exposed at low tide. Within moments of the waters receding, whole families (including school age children) appear to dig the mud for shellfish and shrimp, which they sell in local markets to raise supplementary incomes for very meager returns. Every low tide the families return, regardless of the time of day. Source: ActionAid Vietnam & Institute of Economics (2004) and Xuan Thanh Hoang (2007). 9 SOCIAL PROTECTION poverty could have increased mechanically, looking only at urban households that were without it implying that well-established urban surveyed in both 2004 and 2006. Poverty dwellers have become poorer. among this "panel" component of the survey appears to have increased (the poverty rate goes A similar composition effect may result from from 5.0 to 5.7 percent). Biases from land land conversion. Because of rapid urbanization, reclassification can be corrected by focusing formerly rural wards in the fringes of cities have only on wards that were considered urban been reclassified as urban. Those wards were already in 2004. This adjustment makes little presumably wealthier than most other rural difference to the urban poverty rate in 2006 (it areas, but also poorer than the cities next to goes from 3.9 percent using the classification them. Their reclassification could, other things for 2006 to 4.0 with that of 2004). Finally, prices equal, bring the urban poverty rate down. matter too, as the 2006 price deflators lead to However, there may also be more substantive higher urban poverty rates compared to the 2004 reasons why poverty has apparently stopped deflators (from 3.6 to 3.9 percent). Admittedly, declining in urban areas. Increasingly higher all the changes are minor and unlikely to be prices in the cities stand as a plausible significant. But for the first time in 13 years explanation. The boom in real estate prices they are suggestive of a slowdown in urban increases the cost of retail trade and other poverty reduction, if not worse. services in the cities. Therefore, even simple staples can cost much more than in rural areas, Growing inequality? and the gap is presumably increasing. Higher prices in urban areas may result in lower living Vietnam's strong record in terms of growth and standards for those whose earnings do not grow poverty reduction has been associated with only accordingly. They could push into poverty modest increases in inequality. people whose expenditures are barely above the poverty line. Household expenditure data permits compiling some standard indicators in this respect. For The so-called regional deflators used to instance, it is possible to break the population compute poverty rates can be used to assess into five equally large population groups based whether higher prices may be affecting the on their living standards and to estimate the purchasing power of urban households. share of all expenditures that are made by each Regional deflators play the same role as PPP "quintile" (Table 1.4). These estimates show conversion factors when comparing poverty that the poorest 20 percent of the population rates across countries. They help convert makes around 7.2 percent of all expenditures, expenditures measured in nominal terms into compared to 43.3 percent for the wealthiest "quantities" of goods and services purchased. quintile. This implies that the average person in The regional price deflators for 2006 confirm the richest quintile spends six times as much as that prices in urban areas are rising more rapidly the average person in the poorest one. than in rural areas. The trend would be even slightly faster if the regional price deflators Another standard indicator is the Gini index. were constructed so as to reflect population Measured on a scale from zero to one, this shares in different parts of the county, instead of indicator increases with inequality. In the the consumption shares currently being used. command and control economy preceding Were population weights to be applied, urban economic reforms in Vietnam, the Gini index poverty rates would be marginally higher. was presumably very low, as mostly everybody was poor. Market mechanisms usually result in When all these possible explanations are put a differentiation of incomes encouraging work together, it seems unlikely that urban poverty effort and capital accumulation, and this may be falling (Table 1.3). Biases from a better differentiation can be expected to increase in sampling of migrants can be accounted for by periods of rapid economic growth. 10 A POVERTY UPDATE Table 1.3: What Lies behind Changes in Urban Poverty? 2004 2006 All Panel With 2006 All With 2004 With 2004 Panel house- house- regional households regional ward house- holds holds deflators deflators classification holds Vietnam 19.5 20.2 16.2 16.0 17.0 16.2 16.1 Urban 3.6 5.0 4.0 3.9 3.6 4.0 5.7 Rural 25.0 25.0 20.7 20.4 21.9 20.5 19.6 Source: Based on preliminary data from GSO. Estimates for 2006 are unofficial. Figures in parentheses are standard errors. Because of its economic transition and its abundant anecdotal evidence of rapidly remarkable growth performance, Vietnam could increasing wealth at the top of Vietnam's have seen a substantial increase in inequality. distribution. Sales of cars are growing by nearly Yet the Gini index only trended up slightly 80 percent per year. Literally hundreds of those between 1993 and 2004, and even declined a cars were imported by plane by customers who notch between 2004 and 2006. This stability of did not want to wait for normal delivery, adding the Gini index is driven by the good outrageous transport cost to the already high performance of the three middle quintiles of the import duties and taxes. Vietnamese population, which in turn reflects the emergence newspapers now regularly run stories about the of an increasingly important middle class. The market for luxury phones, cars and, most recent middle three quintiles accounted for 50 percent yachts. The year 2007 saw the purchase of two of total expenditures in 2006, a relatively high Rolls Royce cars, with a price tag exceeding one proportion by developing country standards. million dollars, and one luxury yacht, at a cost But the growth of the middle class is not of more than two million. And there are incompatible with a growing gap between the individual investors owning shares with a value top and the bottom quintiles. in excess of 100 million dollars. Whether standard inequality indicators are able to Standard inequality indicators may not be capture developments of this sort is a matter of telling the full story, however. There is debate (Box 1.3). 11 SOCIAL PROTECTION Table 1.4: Share of Expenditures by Population Quintile 1993 1998 2002 2004 2006 Poorest 8.4 8.2 7.8 7.1 7.2 Near poorest 12.3 11.9 11.2 11.2 11.5 Middle 16.0 15.5 14.6 15.2 15.8 Near richest 21.5 21.2 20.6 21.8 22.3 Richest 41.8 43.3 45.9 44.7 43.3 Total 100.0 100.0 100.0 100.0 100.0 Richest/Poorest 5.0 5.3 5.9 6.3 6.0 Gini index for expenditures 0.34 0.35 0.37 0.37 0.36 Source: Based on preliminary data from GSO. Estimates for 2006 are unofficial. Box 1.3: What is Happening at the Upper End? The primary focus of surveys of living standards in developing countries is often on the measurement of poverty. Therefore, tools such as the VHLSS can provide a sound basis to analyze the welfare of households around the poverty line. They are less good at capturing information from those at the very poorest end of the distribution, since the destitute are often absent from samples. They are also less good at describing the richest end. The very wealthy are unlikely to spend two days answering a questionnaire. And even if they did, they could find it hard to recall how much they have spent over the reference period. The questionnaires themselves tend to be designed based on the expenditure patterns of average citizens, rather than the very wealthy. Questions on conspicuous consumption are often omitted, since they have no relevance for the very large majority of respondents. While the omission of the very rich has no relevance for the measurement of poverty, it could have implications for the measurement of inequality. One study from China found that the exclusion of the top one percent of the sample reduced the Gini index from 0.50 to 0.44. In 2006, the top ten percent of the population accounted for 28 percent of total expenditures. This is less than in several neighboring countries. In China, for example, the top ten percent accounts for 35 percent of total income. In Malaysia it accounts for 38 percent and in both Thailand and the Philippines for 34 percent. Nobody can be sure how much expenditure is missing from the VHLSS. One clue is to look at non-response rates of surveys, but unfortunately this is not available for the VHLSS 2006. Another approach is to compute total consumption based on the VHLSS and compare the resulting figure with private consumption as estimated by national accounts. This exercise shows that the VHLSS 2006 might have underestimated expenditures by around 18.8 percent. This is more than in 2004 (16.2 percent) and 1998 (15.9) percent. Perhaps the biggest uncertainties concern the valuation of housing. The VHLSS asks households to report the approximate market value of the place they live in. The average figures reported for large cities suggest that responses are plausible. However, to measure expenditures it is necessary to convert the value of housing into an imputed rent. This conversion is based on information on the households which do rent the place they live in. But there are not many of those in the sample of the VHLSS, and the resulting ratio of rent to property value may be under-estimated. Many urban households have made sizeable capital gains on the value of their housing. Failure to reflect those gains in their imputed rent may lead to an under-estimation of consumption by the richest population groups, hence to an under-estimation of the extent of inequality. 12 2. ETHNIC MINORITIES Ethnic minority groups account for an plots of land. Ethnic minority people are also increasingly large share of the poor in less geographically mobile, which results in a Vietnam. Poverty rates have been declining different impact of investments in steadily for ethnic minorities living in valleys accessibility. In spite of all this, a faster or in the lowlands and engaging in paddy field reduction in ethnic minority poverty can be cultivation. Other groups, less dominant in expected in the coming years, for the simple number, are seeing little improvement in their mechanical reason that the expenditures per living standards. Overall poverty rates capita of a large number of households are remain much higher among ethnic minorities getting closer to the poverty line and could than among the Kinh and Chinese majority. soon be above it. The severity of poverty is much higher too, and hunger remains an issue for several of the Doing better, but... groups. This is despite considerable progress in making ethnic minority communities more Vietnam's remarkable record in poverty accessible and providing them with basic reduction is qualified by slower progress in social services. In fact, geography alone the case of ethnic minorities. Constituting cannot be retained anymore as the main about 14 percent of the population, ethnic explanation for poverty among ethnic minorities currently account for 44 percent of minorities. Even in remote areas, households the poor and 59 percent of the hungry. Over belonging to the Kinh and Chinese majority the last 13 years, their poverty rate has been are doing reasonably well. Differences in declining at an average of 2.6 percentage endowments do not tell the full story either. points per year, against 3.4 points for the Kinh Ethnic minority people tend to have more and Chinese majority. In 2006, 52 percent of annual cropland than the majority group, ethnic minority people were still living in albeit of poorer quality. Their educational poverty, compared to only 10 percent of Kinh attainment is improving, at roughly the same and Chinese people. Poverty for ethnic speed as that of the majority. And there has minorities was also deeper than for Kinh and also been considerable progress in their access Chinese people. The poverty gap, which to education, health and credit. Ethnic indicates the average distance between the minorities are at a disadvantage in relation to household expenditures of the poor and the annual cropland, use-rights to forestry land poverty line, has also remained consistently and schooling levels. But one of the main higher for ethnic minorities. reasons why they lag behind may be the failure of public policies and programs to take There could be a faster decline of poverty into account their specific needs and among ethnic minorities in the coming years, behavioral differences. For instance, however. The distribution of household agricultural research and extension models expenditures per capita across households in well suited to the lowlands may be of little Vietnam can be seen as the combination of relevance for swidden cultivation on sloping two distinct distributions, one for the Kinh and 13 SOCIAL PROTECTION Chinese group and the other for ethnic the second aggregating all other ethnic minorities (Figure 2.1). The distribution of minorities. However, ethnic minorities are a expenditures among ethnic minority relatively heterogeneous group. households in 2006 is similar to the one Representative data are not available for all observed for the Kinh and Chinese population of them, given their small populations. But in 1993, prior to a period of very rapid poverty distinguishing between sub-groups provides reduction. a "higher resolution" picture, allowing a more nuanced assessment of their differences Growth in Vietnam has, to date, been (Figure 2.2). effective at moving those a little way below the poverty line over the poverty line. It is Ethnic minority groups based in the lowlands possible that the next few years of high (Khmer and Cham) have seen dramatic rates growth will see the movement of ethnic of poverty reduction. Groups who tend to be minorities over the poverty line at a more valley-dwelling rice farmers in the Northern rapid pace than in the past. But such Mountains (Tay, Thai, Muong and Nung) acceleration of poverty reduction, if it were to have also seen a significant reduction in happen, should not be automatically poverty over the past 13 years. But other attributed to successes in policymaking or groups in the Northern Mountains, the government programs. To some extent, it Central Highlands and the South and North would reflect a relatively mechanical process, Central Coasts remain extremely poor. The whereby a slight increase in household lack of progress for these groups tends to be expenditures per capita pushes people over what is ultimately an arbitrary line. masked by the more rapid improvements in the larger ethnic minority groups. For Most previous analyses of ethnic minority instance, the poverty rate of the Central poverty have provided a blunt comparison of Highlands fell by a remarkable 23 percentage performance between two populations, the points between 2002 and 2006. However, a first comprising Kinh and Chinese people, closer examination suggests that there has Figure 2.1: Household Expenditures: A Two-wave Story Source: Bob Baulch, T. Hung Pham and Barry Reilly (2007). 14 ETHNIC MINORITIES Figure 2.2: Diversity among Ethnic Groups Source: Bob Baulch, T. Hung Pham and Barry Reilly (2007). been almost no movement in living standards than four to five years earlier. Villagers for minority groups in that region. agreed that better-off households no longer experienced food shortages and that the Some of these groups are actually poor "hungry period" for poor households had enough to be exposed to hunger. In 2006, a fallen from between eight and ten months per greater proportion of ethnic minorities were year in 2003 to between two and six months below the food poverty line than was the case in 2007. But they also reported that most for Kinh and Chinese people in 1993. Raglai households experienced hunger Hunger appears to be particularly acute in between harvests, especially from April to some regions. For example, more than 40 July. During this period, lack of food and percent of ethnic minorities live in hunger in cash would send people into the forests in the South Central Coast, and more than 52 search of sweet potatoes. At other times of percent in both the Central Highlands and the the year, poorer households might have Northwest. At the same time in all of these adequate quantities of rice or maize to eat, regions less that ten percent of Kinh and but would only be able to afford fish or meat Chinese people were below the higher once or twice a month. general poverty line. Not just remoteness Stories from an ongoing research exercise illustrate the problems ethnic minorities still Regional and ethnic patterns of poverty show face. In four poor villages in Ninh Thuan that prosperity is possible in all regions of the province, not having enough rice or maize for country. Kinh and Chinese people have seen daily consumption was a less severe problem extremely rapid reductions in poverty 15 SOCIAL PROTECTION Figure 2.3: Same Remote Areas, Different Histories Source: Based on GSO data. Data for 2006 are preliminary. regardless of which region they live in. Ethnic instance, in 2006 the average travel time to the minority poverty has remained high in both nearest primary school was 18 minutes (2.7 the Northern Mountains and the Central kilometers) for ethnic minority people, against Highlands, but Kinh and Chinese poverty in 10 minutes (1.6 kilometers) for the majority those regions has declined at 4.6 and 2.8 group. The gap was slightly bigger for the percentage points per year respectively nearest lower secondary school, and (Figure 2.3). Analyses that investigate how the substantially bigger for the nearest upper welfare of different ethnic groups deviates secondary school. But even in this case, the from average welfare confirm that differences average distance for ethnic minority people in living standards are not simply an issue of was reduced from 12.8 to 9.8 kilometers in geography. District-specific effects appear to two years, from 2004 to 2006, whereas it only be modest in explaining differences in fell by 0.3 kilometers in the case of the expenditures per capita. majority group. Access to infrastructure and services is Overall, it is fair to say that the efforts to improving throughout the country, for both the improve infrastructure at local levels are Kinh and Chinese majority and for ethnic paying off. Access has increased as a result minorities. By 2006, 80 percent of the ethnic (Table 2.1). minority population was living within two kilometers of an all-weather road, compared to Land and its use 59 percent in 1998. At 91 and 76 percent respectively, figures were much higher in the Ethnic minority livelihoods are highly case of the majority group. But the change dependent on agriculture. More than 90 percent over time was similar in both cases. Travel of households in this group have annual times have also fallen considerably. For cropland. In regions where ethnic minorities are 16 ETHNIC MINORITIES Table 2.1: Public Programs Are Reaching Out Travel time Within 2 km Access to Access to to lower- Travel time of an all- improved improved secondary to market weather road water sanitation school sources (minutes) (percent of (percent of (minutes) population) (percent of population) population) 2004 Kinh and Chinese 14 9 84 82 36 Ethnic minorities 28 36 66 37 4 Vietnam 17 13 82 76 32 2006 Kinh and Chinese 12 9 91 87 43 Ethnic minorities 25 32 80 55 7 Vietnam 15 13 90 82 38 Source: Based on data from GSO. Data for 2006 are preliminary. Table 2.2: Size of Land Used by Ethnicity Northeast Northwest Central Highlands Kinh and Ethnic Kinh and Ethnic Kinh and Ethnic Chinese minorities Chinese minorities Chinese minorities Annual Have (percent) 88 98 75 99 48 90 cropland Size (m2) 2457 4995 5436 11855 7745 11399 Perennial Have (percent) 33 25 30 20 66 38 cropland Size (m2) 2471 3617 3561 3582 12193 10782 Forest Have (percent) 17 50 - 22 2 4 land Size (m2) 13487 17645 - 22199 - - Source: Based on preliminary data from GSO for 2006. Size is computed over households that have at least one piece of that land; "-" means that the sample is too small to provide a reliable estimate. most dominant the percentage is even higher Perennial cropland land has been important (Table 2.2). Those ethnic minority households in allowing rural households to diversify who have annual cropland tend to have more of income sources. Kinh people have more of it than Kinh and Chinese people, but this should it, particularly in the Central Highlands not necessarily be interpreted as an advantage. where coffee and other cash crops have been Much of the annual cropland farmed by ethnic a source of rural development. Though minorities is on sloping terrain and yields only ethnic minorities often reside close to one crop a year. Kinh people are much more forestry land, whether they actually have likely to have higher-quality irrigated land. access to it varies considerably across the 17 SOCIAL PROTECTION Figure 2.4: Educational Attainment by Ethnicity Source: Based on data from GSO. Data for 2006 are preliminary. Figures are in percent of the adult population. country. Half the ethnic minority Social services households in the Northwest region declare that they "use" forestry land. But only 4 The last 13 years have seen improvements in percent of ethnic minorities in the Central educational attainment across the board. Highlands have these use-rights, despite this Though ethnic minorities have not caught up being the region with largest forest area in with the Kinh and Chinese majority, there is the entire country. no sign that they are falling behind either (Figure 2.4). For instance, the proportion of The way agricultural research and extension adults who have not completed primary are organized may result in ethnic minorities education dropped by 30 percentage points reaping lower returns from their land for the Kinh and Chinese majority, and for allocations. Policies that encouraged the 28 percentage points for ethnic minorities. production of wet rice in valleys may have had Over the same period, the proportion of little impact on ethnic minority poverty. By adults who completed lower secondary contrast, limited attention has been paid to education rose slightly faster for ethnic upland agriculture, which is an essential part minorities than for the Kinh and Chinese of most ethnic minority livelihoods. majority. And while there appears to be a Diminishing fallow times present a challenge slowdown in improvements during the last for upland farming systems, one that has not two years for both groups, a closer analysis been sufficiently addressed by public is needed to determine whether this agricultural research. Similarly, agricultural represents a real trend or is rather a product extension is not supporting farmers enough as of flaws in the data. they switch from subsistence to cash crops on their upland plots. More broadly, agricultural There has also been a considerable success in extension is not oriented towards the technical the distribution of health insurance or challenges of upland agriculture and its staff is healthcare cards to ethnic minority people often unable to communicate in ethnic (Table 2.3). Data for 1998 do not allow minority languages. discerning between health cards and health 18 ETHNIC MINORITIES Table 2.3: Access to Health Care by Ethnicity Ethnic minorities Kinh and Chinese Have health Have health insurance Have health Have health insurance insurance and free healthcare card insurance and free healthcare card (percent) (percent) (percent) (percent) Regions 1998 2004 2006 1998 2004 2006 Red River Delta 8 37 72 33 37 49 Northeast 8 42 73 36 42 61 Northwest 3 70 83 44 55 75 North Central Coast 11 66 86 32 38 55 South Central Coast 3 82 97 36 38 56 Central Highlands 1 70 95 27 36 51 South East 41 41 51 32 36 47 Mekong Delta 3 43 55 21 26 40 Vietnam 8 55 78 30 35 49 Source: Based on data from GSO. Data for 2006 are preliminary. insurance, so the comparison with later years compared to only 5 percent among Kinh and exaggerates the expansion in access to health Chinese households. services. Even with this caveat, it is clear that there have been serious efforts to reach Increased access to formal credit by ethnic vulnerable groups. In 2006, nearly 80 percent minorities might be interpreted as proof of the of ethnic minority households had either free success of VBSP in extending its outreach. health insurance or a healthcare card. Such Indeed, it is part of VBSP's mandate to provide directed lending targeted to poor high proportion reflects a rapid improvement households, particularly ethnic minorities. in the last two years alone, and a dramatic Household survey data suggest that this is transformation over the last eight years. actually happening, as VBSP is reaching more There has also been a slight increase in the ethnic minority households and is becoming fraction of ethnic minority households increasingly important in their overall accessing credit sources. About one quarter of borrowing. But there is also a possibility that Kinh and Chinese households and one third of VBSP is crowding out lending by VBARD. These two banks offer different financial ethnic minority households took loans from products, with the latter characterized by the formal financial sector in 2006 (Table 2.4). larger loan sizes, sometimes requiring The latter is increasingly important for ethnic collateral, which are arguably more useful for minority households, accounting at present for investment purposes. Recent surveys indicate 58 percent of all loans and 70 percent of all dissatisfaction with elements of the VBSP resources borrowed. The Vietnam Bank for offer, particularly the small loan size and the Social Policies (VBSP) is becoming the practice of tying lending to particular dominant source of credit for ethnic activities. Current trends suggest there is a minorities, whereas VBARD, which offers possibility that VBSP may evolve into a bank larger loans, is the primary lender to Kinh and for ethnic minorities, who may then face Chinese households. Nearly one third of the difficulties transitioning to the mainstream resources borrowed by ethnic minority credit market and taking advantage of the households actually came from the VBSP, larger loans offered by VBARD. 19 SOCIAL PROTECTION Table 2.4: Access to Credit by Ethnicity Percent of households Percent of resources borrowing from the borrowed from the different lenders different lenders 2006 2004 2006 Lenders Kinh and Ethnic Kinh and Ethnic Kinh and Ethnic Chinese minorities Chinese minorities Chinese minorities Formal 25 36 61 68 70 69 VBSP 7 18 3 16 5 30 VBARD 17 18 42 51 47 38 Other banks 2 0 17 1 18 0 Informal 21 23 39 31 30 31 Employment support fund 1 1 1 0 1 Credit organization 2 1 4 2 4 6 Mass organization 3 3 1 3 2 4 Money lenders 4 5 6 5 4 5 Friends and relatives 12 13 25 19 19 13 Others 1 2 2 3 1 2 Total 41 54 100 100 100 100 Source: Based on data from GSO. Data for 2006 are preliminary. The total is not sum of formal and informal borrowing because a household can borrow from more than one lender. The cultural dimension quantitative techniques found Kinh people in ethnic minority areas to be much more mobile Thanks to the availability of good household than the minorities. Nearly all Kinh people survey data, it is relatively straightforward to interviewed had visited the commune centre and document the endowments of ethnic minorities, well over 90 percent had visited the provincial as well as their access to social services. Data town, compared to 85 percent and a little more of this sort are at times used to "explain" the gap than one half respectively for ethnic minorities. in household expenditures per capita between Most notable was the difference in the histories the Kinh and Chinese majority and the ethnic of visits outside the province. Two thirds of minority population. The result is a Kinh people had traveled beyond province decomposition of the gap between differences in boundaries, but only 18 percent of ethnic endowments (for instance, in the amount of land minorities had done so. Higher mobility may available) and differences in the returns to those give Kinh people advantages not easily picked endowments (in the productivity of such land). up by household surveys, such as greater access However, these decompositions can be to information, broader social networks, misleading, as they tend to ignore the role of increased contact with markets and the community characteristics which are much opportunity to learn from others. more difficult to capture through survey data. Failing to understand this more cultural For instance, looking at differences in dimension of the difference between ethnic geography and infrastructure is easy, but actual minorities and the Kinh and Chinese majority mobility could be more relevant. One recent may easily result in stigma, lead to research exercise combining qualitative and inappropriate policy responses and frustrate 20 ETHNIC MINORITIES development efforts. Anthropological studies modern medicine is more common when the have explored differences in behaviors and have commune health station (CHS) is close by, the found that some ethnic minority responses to doctor speaks the same language as the patient, policies and programs, though unexpected by and the health insurance card is issued with officials used to managing service delivery in correct information on it. majority areas, are entirely rational given the context. In other cases, those policies and Other interventions may be misguided because programs have been found to be conceptually of behavioral assumptions which are not flawed, being based on ill-informed substantiated by evidence. Over several preconceptions. decades, well-funded programs have aimed to settle groups seen as nomadic because of their Government programs to reduce ethnic minority dependence on swidden agriculture. However, poverty are often built on the assumption that a recent study found that only a tiny fraction of activities which worked well for the Kinh and ethnic minority people were living a nomadic Chinese majority should also work well for lifestyle, with most upland farmers being well ethnic minorities. When they do not, lack of settled. In the same spirit, a review of these understanding can lead to the conclusion that sedentarization programs concluded that they the target beneficiaries are backward, or had had little discernible impact on unmotivated, or lazy. livelihoods. For example, the distribution of health There is also a need to look beyond ethnic insurance cards has, in general, improved access minorities as a whole and realize the challenges to healthcare for the poor. If certain ethnic the variety of cultures and contexts across the 53 minority groups make limited use of their cards, different groups present for policymakers. it could be tempting to conclude that they do not Nationally representative household surveys are value modern medicine as much as the Kinh and unable to capture this diversity well, because the Chinese people do. However, a study of ethnic smaller groups get only a tiny representation in differences in health seeking behavior in Son La their sample, if any at all. Studies that have suggests that economic considerations may investigated the situation of a small number of matter more than perceptions. When taking into groups in certain areas show that differences account the distance to the closest provider between ethnic minority groups (in terms of covered by the health insurance program, the endowments, contexts and, by extension, time away from work, the expense and the responses to government interventions) can be likelihood of decent treatment, the decision not even more marked than differences with the to use the card seems sensible. Reliance on Kinh and Chinese majority. 21 3. SHIFTS AND SHOCKS Rapid economic growth has dramatically cutting across several policy areas. And improved living standards for the bulk of the special policies are needed for people living population. But it has also brought other shifts with disabilities (PLWDs), a group which is and transformations, some underway and probably larger and more vulnerable than some embryonic, some domestic and some previously acknowledged. Last but not least, global in nature, which influence the nature of the same market mechanisms which support risks that people face and the policy options to rapid economic growth and dramatically support the vulnerable. Lower fertility and improved living standards can also be a source improvements in health care are rapidly of vulnerability, with price fluctuations in raising life expectancy. By 2040, more than international markets and anti-dumping 20 percent of the population will be above 60 measures by trade partners resulting in large years of age. By then, social change and rural- shocks to household income. urban migration might have undermined the multi-generation living arrangements which Ageing and migration currently sustain the livelihoods of the elderly. Vietnam is also one of the countries most Vietnam has a young population. At the turn of vulnerable to natural disasters, such as floods the millennium, more than half of it was below and typhoons. Given the massive the age of 25. But fertility rates have been concentration of its population along the declining, from 2.33 in 1998 to 2.11 in 2004, coastline and in the large deltas, disasters take partly as a result of the vigorous family planning a heavy toll in lost lives and damaged campaigns implemented over the last two livelihoods. Climate change will make these decades. Meanwhile, improvements in living shocks worse over time. But providing standards and healthcare have increased life insurance against natural disasters raises expectancy. In 1999, a newborn could hope to important implementation challenges. There live to the average age of 68.3. By 2005, life is also a fundamental transformation in expectancy at birth had risen to 71.3. morbidity and mortality. Traditional infectious diseases have receded, whereas With declining fertility and growing longevity, non-communicable diseases and accidents the Vietnamese population is bound to age (including traffic fatalities) account for a rapidly (Figure 3.1). By 2020, only 40 percent substantial fraction of all deaths. This of the population will be below the age of 25 epidemiological transition requires a and 10 percent will be over the age of 60. fundamental transformation of health care, Further ahead, in 2040 only 32 percent of the from vaccination campaigns and other forms population will be under the age of 25, but the of prevention at local levels towards proportion of the population over 60 years of individual- or household-based insurance age will have doubled. The proportion will be provided regardless of location. Addressing higher for women (22 percent) than for men (19 emerging diseases such as HIV or avian percent), an issue of relevance for the debate on influenza also entails innovative interventions, retirement ages. 22 SHIFTS AND SHOCKS Figure 3.1: A Rapidly Ageing Population This demographic transition has a number of implications for policymaking. There will clearly be a much larger number of elderly. The 20 percent of people who will be over 60 years old in 2040 are currently economically active. Most of them are not poor and many of those who are poor are quite likely to exit poverty in the next five to ten years. An important opportunity exists therefore to put in place mechanisms that allow today's youth to be saving for economic inactivity in older age. This will be particularly important as household structures begin to change, with more traditional, multi-generation households becoming less dominant. This change is already taking place in urban areas, where values are shifting along with increased wealth and intensified exposure to the wider world. But it is also likely to happen in rural areas, as domestic migration becomes a mainstream response to economic opportunities in other parts of the country. Traditional patterns of care within a multi-generation household may be less of a safety net when the youth of today retire than they are now. Domestic migration is the other major demographic transformation ahead. Explosive growth in the non-agricultural sector was sustained by a massive movement of people out of agriculture. Quantifying such movement is difficult, however, and there are good reasons to suspect that current statistics underestimate it by a large margin. Indeed, migrants from rural to urban areas are not systematically included in the surveys most often used to assess poverty and vulnerability (Box 3.1). It is not clear how the absence of migrants affects the poverty estimates in urban areas. But it is clear that migrants may be vulnerable in a number of ways which are hard to analyze with existing data. Access to services depending on registration papers (as for education) or on presence in a specified location (as for healthcare) is a case in point. Source: Based on United Nations (2007). 23 SOCIAL PROTECTION Box 3.1: A Vastly Under-Reported Internal Migration Does the VHLSS underestimate the number of rural-urban migrants in Vietnam? In spite of sustained efforts to improve its sampling, there is a concern that migrants working in industrial zones and living in rented bedrooms may not be appropriately captured. There is also a possibility that the sampling frame may be of better quality for the population having permanent forms of registration (KT1 or KT2 status) than for people holding more temporary cards (KT3 or KT4). One way to answer this question is to focus on HCMC, where a population census was carried out in 2004. The census is supposed to be the most comprehensive data source, as it attempts to capture all households. The VHLSS has a question that enables to estimate the population having differing types of registration. The responses to this question in the case of HCMC can be matched to those provided to a similar question in the Census. If migrants with temporary forms of registration were appropriately captured in the 2004 VHLSS, the proportions of respondents with KT3 and KT4 registration should be similar of that of the 2004 Census. As it happens, the Census estimated that households with KT3 or KT4 registration accounted to nearly 20 percent the population, compared to only 3.8 percent in the VHLSS. Registration 2004 VHLSS in 2004 Census for HCMC Status HCMC All Communes Communes in VHLSS KT1 85.5 71.6 71.8 KT2 9.2 9.7 8.6 KT3 11.5 12.6 3.8 KT4 7.2 7.0 Other 1.5 - - Another way to assess the coverage of rural-urban migrants in the VHLSS is to estimate the population employed in different sectors, or holding different occupations. The VHLSS responses can in this case be matched to employment estimates from the Enterprise Surveys conducted annually by GSO. The results suggest, again, that many migrants are omitted from the VHLSS sample. In 2002, the VHLSS underestimated employment in manufacturing in all of Vietnam by about 2 percent. By 2006 the gap had widened to 22 percent. The largest difference is for the leather goods industry, which includes shoe manufacturers. The VHLSS appears to miss about 41 percent of leather workers (160,000 people in all) in 2002 and around 45 percent (or 230,000 workers) in 2004. It is also apparent that the VHLSS surveys miss a substantial number of construction workers, who are often migrants. The gap in this case amounted to 480,000 workers in 2002 and 420,000 workers in 2004. Source: Based on data from GSO. Weather and climate Most of Vietnam's 2,360 rivers are short and steep, so that heavy rainfall in their basins Vietnam's climate and topography make it one produces intense, short duration floods. of the countries most prone to natural disasters. Sizeable portions of the country, and especially Typhoons and floods are by far the most the Central Highlands and the Central Coast, frequent and severe natural hazard. Typhoons are subject to heavy rainfall. Even more hit the long coastal area most directly. unsettling is the substantial variation in rainfall, Flooding is extensive and prolonged especially over the long coastline and in the throughout the wet season in the large deltas. Northern Mountains (Figure 3.2). 24 SHIFTS AND SHOCKS Figure 3.2: Disaster-Prone Areas Source: Quy-Toan Do and Trung Dang Le (2007). The coefficient of variation is the ratio between the dispersion (as measured by the standard error) and the mean. Data are for 2001-2006. With around 70 percent of the population living such as floodplains, coastal swamps, drainage in lowland areas, in the deltas or along the 3,200 channels or other natural buffers, only adds to km-long coastline, these disasters often take a the vulnerability of the population. heavy toll. There are about 700 deaths per year as a result of typhoons, floods and rainfall- Preliminary results from an ongoing study induced landslides. In addition to the human suggest that households living in disaster prone casualties, there are important economic losses areas have been active in diversifying their and environmental damage. Between 1991 and income sources so as to reduce risk. Matching 2000, natural disasters resulted in 2.3 million daily rainfall data from 172 weather stations for tons of food lost, 6 million houses destroyed and the period 2001-2006 with household survey 9,000 ships sank. In total about 3 billion dollars data from the surrounding areas shows that in assets was lost. farmers in areas facing higher weather volatility diversify their labor inputs more. But The encroachment of economic activity and they do not self-insure by accumulating development into marginally suitable areas, livestock or asset holdings, nor do they 25 SOCIAL PROTECTION participate more in social networks. The study whereas the sea level could rise by 18 to 59 also suggests that farmers in these disaster centimeters. prone areas have fewer diversification options, perhaps because they do not have good access Different scenarios have been developed to to land or credit. model the potential impacts of climate change. While they need to be interpreted with caution, Vulnerability to natural disasters may be they provide important warnings. One study exacerbated over time by climate change. considered 84 coastal countries and found that International consensus now exists that global Vietnam was one of the five most vulnerable to warming is linked with human activity. There is sea level rise in the entire world, and the most also consensus that the poor are vulnerable to its vulnerable in East Asia (Figure 3.3). This is impacts, because they live in more exposed because a large proportion of the population, areas and depend more on natural resources for infrastructure and economic production, their livelihoods. But there is uncertainty over including irrigated agriculture, is located in the scale and nature of the possible impacts. coastal lowlands and deltas. Modeling is more advanced in relation to the impact on sea levels than on weather patterns, Digital surface data has been used to improve and especially on how rainfall patterns will the resolution of these scenarios. It appears that change across the country. Climate simulations a one-meter rise in the sea level would affect 39 under a range of scenarios suggest that mean of the 64 provinces in six of the eight economic annual temperature in Vietnam could increase regions of Vietnam. About 20 percent of by between 1.5 to 2.5 degrees Celsius by 2100, communes could be wholly or partly inundated, Figure 3.3: Going under Water? Source: Dasgupta and others (2007) 26 SHIFTS AND SHOCKS with the Mekong River Delta being the most Providing insurance against natural disaster seriously affected area. By one estimate, a one- raises important implementation challenges. meter rise in sea level would affect Pooling risk is difficult when shocks are bound approximately 5 percent of Vietnam's land area, to affect large population groups, which is the 11 percent of the population, and 7 percent of proper of hazards such as typhoons or floods. agricultural output. However, these figures Measuring the extent of damage in an objective should not be taken literally. In particular, more way might be difficult as well. And the detailed analysis is needed before deriving any availability of insurance could discourage policy implications. For example, scenarios of prevention efforts by households and this sort do not consider the contribution that the communities. As a result, market responses existing system of sea dykes provides for flood may be insufficient, and complementarities prevention. between public policy and private provision will have to be explored. Beyond the potential for arguments over assumptions, it appears that without effective Health and disability adaptation strategies, climate change will increase vulnerability among some of the Vietnam is well along in its epidemiological poorest groups in Vietnam. As is the case for transition, with non-communicable diseases natural disasters today, there will be a need to being by now the main cause of morbidity and develop appropriate policy responses, from mortality, by a large margin. Traditional mitigation to insurance. infectious diseases may still be an issue in Figure 3.4: The Evolving Causes of Mortality Source: Based on data from MOH. 27 SOCIAL PROTECTION Box 3.2: New Diseases: Who Is at Risk? Until the mid-1990s, HIV prevalence was low in Vietnam. Rates were largely below 1 percent, even among typically vulnerable groups such as sex workers, the military, clients of sex workers and tuberculosis patients. Injecting drug users were the exception, with rates of 20 percent reported among selected samples of them as early as 1994. The quality of these data is admittedly of concern. The sampling of injecting drug users and sex workers is largely limited to detained populations, who may have higher prevalence. Nonetheless, injecting drug use is clearly central to Vietnam's HIV epidemic, accounting directly for about 60 percent of reported transmission and contributing indirectly to an even larger percentage, through the links between injecting drug use, sex workers and clients. By 2004, HIV rates were low among pregnant women (0.27 percent) and military recruits (0.39). They were intermediate among clients of sex workers (2.56 percent), tuberculosis patients (3.5) and sex workers themselves (6.5%). They were highest among injecting drug users, reaching a striking 32.2 percent. Rates among sex workers rose in the wake of increased HIV infection among injecting drug users. This suggests that injecting drug use was seeding the epidemics among sex workers, a notion supported by behavioral surveillance. There are too few representative, high-quality data to make confident HIV projections in a country as large and diverse as Vietnam, but it is clear that HIV presents a significant health and development threat. Even if HIV was to simply saturate vulnerable populations and their immediate partners, adult prevalence could eventually exceed 1 percent of the general adult population. That is, over 300,000 people. In addition, by being disproportionately concentrated among marginalized groups, HIV may compromise social inclusion and solidarity. As a long-term illness, HIV also creates particular challenges for health services and is a source of serious costs for households. Avian flu is also a new disease but unlike the slow, chronic HIV, its impact on suffers comes in the form of a short, sharp shock. Avian flu, caused by the infection of highly pathogenic avian influenza (HPAI) virus is a devastating disease in poultry, the disruption of its production and trade. An HPAI epidemic outbreak occurred in Hong Kong in 1997, causing not only poultry death, but also 18 human cases leading to six fatalities. Since then, HPAI outbreaks have been reported in many countries in Asia, Europe and Africa. The virus has caused more than 300 laboratory-confirmed cases, with 185 deaths. The HPAI virus is currently considered a potential source for future human influenza pandemics and raises concerns not just for agriculture but also to public health authorities and scientific communities. Since the first human case of HPAI was identified in Vietnam in 2003, the international community has witnessed large-scale mobilization in order to contain the spread of the virus among the animal population, and to prepare governments, agencies and the populations to a possible worldwide pandemic. Despite the amount of international attention, little to date is known about the virus. Its possible routes of transmission are not well identified. Contact with poultry is recognized as a risk factor, however, making the breeders of duck, goose and chicken more vulnerable. poorer, upland parts of the country, but in most Accidents, injury and poisoning now account for areas their incidence fell dramatically over the a sizeable proportion of deaths. Traffic accidents last thirty years (Figure 3.4). Meanwhile, the are responsible for almost half all accidental incidence of cancer, cardio-vascular accidents, deaths, and resulted in 14,800 fatalities in 2006 diabetes and mental illness is rising and poses alone. This amounts to 19 dead per 100,000 the greatest threat to health and length of life. people, a rate which is higher than the East Asia Non-communicable diseases already account average, worse than in any European transition for 62 percent of all deaths, and the proportion country and four to five times worse than in the is likely to increase in the medium term, along best performing countries. The likelihood of with growing per capita incomes and changes in injury or death in the event of a road accident is lifestyle. estimated at 20 to 25 percent. 28 SHIFTS AND SHOCKS Table 3.1: The Economic Consequences of Health Shocks Urban Rural Change expressed in percent, Earned Medical Food Earned Medical Food for the household as a whole income spending intake income spending intake Death (working age) -30.6 n.s. -18.5 -12.5 n.s. -8.3 Hospitalization (working) n.s. 339.3 n.s. n.s. 292.0 n.s. Hospitalization (non-working) n.s. 256.8 n.s. n.s. 322.1 n.s. Serious illness (household head) n.s. n.s. n.s. n.s. n.s. -10.8 Source: Based on Adam Wagstaff (2007a). Figures were estimated using data from the 1993 and 1998 VLSS. Death is computed only if it took place in the two years preceding the 1998 survey; hospitalization, if it spanned at least seven days over the previous year; serious illness if body mass fell by one standard deviation or more between the surveys. Estimates are for changes at the household level, everything else being equal; n.s. stands for not statistically significant. While traditional infectious diseases are 1993, depending on whether they had receding, emerging communicable diseases experienced a major health shock, highlights pose new threats to public health. SARS, how large the effects can be (Table 3.1). The HIV/AIDS and Avian Influenza are the most findings suggest that the impact is stronger in visible examples, especially because of the risk urban households than in rural households. The of rapid scale up (Box 3.1). But the re- study also demonstrates the impact of emergence of diseases which were previously hospitalization on household medical under control, such as Japanese encephalitis, expenditures. In rural areas, serious illness can dengue fever and tuberculosis, also represents a be destabilizing enough to cause a 10 percent challenge for the authorities. drop in household food intake. People living with HIV continue to face Impacts of this magnitude call for policy considerable stigma. By the time their interventions to mitigate health shocks and, condition becomes known, many of them lose when these shocks happen, to reduce their their jobs, often under the excuse that their impact on household living standards. health does not fit their work requirements. However, the epidemiological transition of This critically reduces the income of their Vietnam has implications on the nature of the families, especially when they are the required policies. When morbidity and breadwinners. Women living with HIV tend to mortality were related to infectious diseases, be more highly stigmatized than men due to the vaccination and prevention campaigns with a commonly held assumption that HIV is acquired geographical focus were the most effective through immoral means, and the social interventions. With non-communicable expectation that women should uphold the diseases becoming a more important cause of moral integrity of family and society, while men morbidity and mortality, curative care gains can be more self-indulgent. Stigma is also prominence, and the focus needs to shift to common in other contexts, from the denial of providing insurance to individuals and service in restaurants or shops to the separation households, regardless of where they live. of children at school. Emerging communicable diseases also call for Whether due to communicable disease, non- different policy interventions, involving a closer communicable disease or accident, the impact coordination between health services and other of a health shock on household living standards areas of government, ranging from extension can be substantial. A study comparing the services (in the case of Avian Influenza) to situation of the same households in 1998 and social policies (in the case of HIV) to 29 SOCIAL PROTECTION infrastructure and education (in the case of living standards in Vietnam, but it also has traffic accidents). Community-based responses brought in a number of dramatic changes. The are likely to be important to reach out to groups conversion of land from agricultural to non- that are poorly positioned to know or demand agricultural purposes, the shift from farming to the services they need. wage employment and the movement of people from rural to urban areas are all signs of a PLWDs are another group in need of special vibrant economy that offers opportunity. But policies. The VHLSS 2006 contained a module these changes are also potential sources of which allows estimating how large this group is vulnerability, from landlessness to in Vietnam. The questionnaire probes unemployment to the loss of social networks. respondents on seeing, hearing, walking, remembering, caring for themselves and Some groups are particularly vulnerable communicating. One simple way to present the because they are already poor, or are barely findings is to distinguish between "any level of above the poverty line, or lack sufficient assets disability" and "severe disability". About 15.3 to cope with shocks (Table 3.2). Among them percent of the population falls in the former are farmers, those working for other households group and 3.7 percent in the latter. These figures or for collectives and those who are reveal a much higher prevalence of disabilities economically inactive. Those working for the than suggested by previous, official estimates. government or in a foreign invested company appear to be least vulnerable by these criteria. A closer inspection shows a strong link between disability and age, with the increase in disability Recent changes in land use influence prevalence as age rises being more marked than vulnerability as well. For example, increased in other countries. There are higher reported reliance on aquaculture in coastal zones has levels of disability for women than for men, generated high returns for those with sufficient possibly reflecting their older average age. investment capital and connections with local There are no discernible patterns linking the officials. But this trend may increase the broader category of disability to income group, vulnerability of groups without access to such and no regional patterns either. But a strong assets. This is because aquaculture ponds have association exists between severe disability and replaced or degraded public goods assets (such poverty. as coastal mangroves and wetlands for capture fisheries) that previously helped diversify A small survey of PLWDs in Hung Yen, Quang livelihoods, especially in times of stress. Ngai and Ho Chi Minh City (HCMC) is quite telling about the difficulties they face. The Global integration has opened new outlets for educational attainment of this group is much domestic production, allowing economies of lower than that of the general population, with scale to materialize and supporting massive more than one third of the sample being productivity gains. But exposure to illiterate. This low level of education may result international markets can also result in large from disability itself and its impact on learning price fluctuations, as experienced by coffee capacity. But it also reflects the difficulties faced farmers at the turn of the millennium. As the by this group in accessing education. By the economy diversifies, and more sophisticated time of the survey, only half of the sample was mechanisms are developed to hedge against working, mainly on precarious jobs including risks, some of these vulnerabilities will start petty trade and simple services such as food and receding. However, in the medium term even a stall cooking and motorbike washing. step as positive as WTO accession can be a Market fluctuations double-edged sword. For instance, the removal of subsidies for fertilizers could affect farmers. The adoption of market mechanisms has In relation to global integration, one important resulted in an extraordinary improvement of source of vulnerability is the NME status given 30 SHIFTS AND SHOCKS Table 3.2: Who Is more Vulnerable? Percent of households who Are less than Have less 10 percent than 15 Meet any of above the million dong these three Occupation of household head Are poor poverty line in assets criteria Inactive 12.5 3.8 2.6 14.2 Employed by Government 1.5 1.6 0.7 6.1 State-owned enterprises (SOEs) 4.7 1.1 1.3 3.4 Foreign-invested company 2.5 0.0 0.0 0.4 Private sector 4.2 2.9 2.9 2.4 Households and collective enterprises 20.4 6.8 7.2 13.3 Non-farm self-employment 6.8 2.2 1.7 22.6 Farmers 25.1 5.8 4.8 37.5 Total 16.0 4.4 3.7 100 Source: Based on preliminary GSO data for 2006. to Vietnam at the time of accession. This mentioned as sources of vulnerability. Among status makes it much easier for its trading farmers, crop failure was identified as the partners to impose anti-dumping measures, biggest risk, followed by the lack of which in turn can have a serious impact on the employment and the ups and downs of the living standards of specific population groups market prices for key agricultural products. (Box 3.3). The most common measure taken by a vast The importance of market fluctuations as a majority of households to address the effects of source of vulnerability is highlighted by a recent these shocks was to reduce expenditures. Two survey conducted among poor households in 24 thirds of them also borrowed, mainly from rural communes across eight provinces. Asked private lenders and relatives, friends or about the economic risks they faced, 87 percent neighbors. One third relied on relief programs of respondents declared to have experienced at offered by the government. About 12 percent least one major shock over the previous three of the respondents declared having taken years. On average, each household had suffered children out of school, and almost ten percent two shocks. Floods, droughts and had a family member migrating to urban areas environmental degradation were often for work. 31 SOCIAL PROTECTION Box 3.3: Anti-Dumping and Catfish Farmers in the Mekong Delta In June 2002, US catfish farmers filed a lawsuit against Vietnam, arguing that its exports of frozen catfish (that is, tra and basa) fillets were conducted at prices below the cost of production. A few months later, in January 2003, the US Department of Commerce ruled in favor of the plaintiffs. Because Vietnam is not recognized as a market economy, it is claimed that its domestic prices are distorted and therefore its domestic costs cannot be used to assess whether production or exports are subsidized. Instead, the production costs of a "surrogate", supposedly comparable country are used. In the case of Vietnamese catfish, the US Department of Commerce chose Bangladesh as the surrogate country. The comparison between Vietnamese prices and Bangladeshi costs resulted in tariffs ranging between 37 and 64 percent on imports of frozen catfish from Vietnam. In July 2003, the US International Trade Commission ratified the ruling. Shortly after, Vietnamese exports of catfish to the US plummeted. The impact of this anti-dumping measure on the living standards of catfish farmers can be assessed using household data from the 2002 and 2004 VHLSS, which span the period right before and after the tariffs on catfish were imposed. A key feature of the VHLSS is that thousands of households surveyed in 2002 were interviewed again in 2004. Many among them were aquaculture farmers in the Mekong River Delta, the main area where catfish is produced in Vietnam. Comparing the change in the living standards of these farmers to that of aquaculture farmers in other regions, where catfish is not produced, gives an estimate of the consequences of the anti-dumping measure. Results indicate that the shock has an impact on income (via lower prices and quantities), per capita expenditures, and current use of inputs, especially in fishing (but also in agriculture more broadly). The average income of a typical catfish farmer fell by 3 to 10 percent, whereas per capita expenditure dropped by 11 to 17 percent. Labor supply increased, partly to compensate the drop in fish-related income. However, long-run responses, like expenditure in education and health, and fundamentally school attendance, were not affected. It would then seem that catfish farmers perceived the shock as being temporary. Source: Guido Porto (2007). 32 PART II: POLICIES AND PROGRAMS 4. LABOR MARKETS Abundance of opportunities for gainful number of wildcat strikes over time suggests employment is one of the most important forms that the current mechanisms for dispute of social protection for households in general, resolution are increasingly ill-adapted to and for the poor in particular. Moving out of address labor discontent. Facilitating collective farming jobs into self-employment, and bargaining at the enterprise level may also hold especially into salaried jobs, has been one of the key for Vietnam to exit the NME status the main avenues for poverty reduction in early on, hence to be able to protect its jobs Vietnam. This transition has often taken place from unjustified anti-dumping measures. by moving out of rural areas. Low unemployment rates, high occupational Activity and employment mobility and chances for migrants to do better are thus essential for households to escape Vietnam is a nation at work. At 71 percent, the poverty and cope with adverse shocks. So far, fraction of the working-age population that is labor markets in Vietnam have performed economically active is high compared to other relatively well. But global integration brings in countries and among the highest in the South new challenges and labor market regulations East Asia region. Unlike other countries, the have to be adapted to a more complex and fraction is very similar for men and for women. diversified economy. In this context, the Meanwhile, the fraction of the working-age introduction of an unemployment insurance population in the total population is bound to scheme may not be a sufficient policy response, increase for some time. This is because of assuming that it is at all warranted. The most Vietnam's rapid demographic transition. The urgent priority is to spur the growth of salaried population growth rate declined from 1.9 employment and to support its formalization. percent per year in the early 1990s to 1.3 This entails a delicate balance between percent at the beginning of this century. As a preserving competitiveness and protecting result, the fraction of children will decline, workers from abuse. The revision of minimum while the numerous children of today will be wage policies and the mechanisms for entering the labor market. collective bargaining is crucial in this respect. The commitments made by Vietnam in the With roughly 1.4 million people joining the context of WTO accession imply that the labor force every year, both economic minimum wages paid by domestic and foreign prosperity and social stability hinge on very companies will need to be unified in a few rapid job creation. So far, the performance of years. To avoid penalizing job creation in less the labor market has been quite remarkable in developed parts of the country, a differentiation this respect. Over the decade and a half for of minimum wages by region may be which representative household surveys are considered. Also, to avoid an explosion in available, the unemployment rate has remained government spending, minimum wages should stable and relatively low (Table 4.1). The be dissociated from the setting of salaries for precision of the estimate may be questioned as civil servants and pensions for retirees. As for the available surveys are not well designed to collective bargaining, the increase in the monitor the labor market. But it is clear that 35 SOCIAL PROTECTION Table 4.1: The Structure of the Labor Force 1993 1998 2002 2004 2006 Inactive 19.4 15.3 16.7 17.2 19.5 Active 80.6 84.7 83.3 82.8 79.7 Employed Government 3.1 3.6 4.4 5.3 5.5 SOEs 2.5 2.6 3.3 3.1 3.3 Foreign-invested companies 0.1 1.1 0.8 1.3 1.6 Private enterprises 10.8 10.1 15.7 17.0 17.3 Non-farm self-employment 14.7 16.5 19.1 16.5 19.5 Farmers 49.5 50.2 38.2 38.8 32.6 Unemployed 0.6 1.8 0.8 0.8 Urban unemployment rate (MOLISA) 6.9 6.0 5.6 4.8 Source: Based on data from GSO and MOLISA. GSO data for 2006 are preliminary. All figures are in percent of the population aged 15 to 64, except for unemployment rates, which are in percent of the active population. Employment figures are based on main occupation. Household businesses are treated as private enterprises for their hiring of wage workers, and as a source of non-farm self-employment for the jobs they provide to household members on an unpaid basis. unlike other transition economies, and despite for by domestic private enterprises and foreign- the remarkable transformation of its economic invested companies has increased remarkably structure, Vietnam has not faced a mass over this period. The dynamism of these unemployment problem. companies accounts for the bulk of job creation during this period, and helps explain why To some extent, this relatively good Vietnam has not faced mass unemployment. performance simply reflects the fact that a majority of the active population still works in Not everybody working for a wage or a salary is agriculture. Underemployment, more than in the formal sector of the economy, however. outright unemployment, is the main mechanism In the context of the labor market, formality can for labor adjustment in rural areas. But the be defined in relation to the Labor Code, fraction of the working age population for introduced in 1994 and amended in 2002. In which farming is the main occupation has been theory, the Labor Code applies to urban and declining rapidly, from about half in 1993 to a rural laborers who have signed a labor contract third in 2006. Meanwhile, the fraction working or have a verbal agreement with their for a wage or a salary has increased steadily, employers. The 2002 amendment defined from 16.5 to 27.7 percent. different kinds of contracts such as indefinite term, definite term and seasonal work. Laborers In the early 1990s, civil servants and public working on farms are also theoretically sector workers accounted for the bulk of wage protected by the Labor Code. The reality is that employment. Not anymore. Their share of the few workers in rural areas benefit from such labor force has slightly increased, partly as a protection outside state-owned farms and result of the expansion in the coverage of social plantations. Even in urban areas, non- services and partly because of the recruitment of compliance with labor regulations is village chiefs into government administration. widespread. For instance, a study on migrants But the share of wage employment accounted revealed that only 36 percent of those in paid 36 LABOR MARKETS employment had a labor contract, with the the economy diversifies and further integrates proportion being lower among women. with the outside world will require a high- quality labor force survey conducted on a Assessing trends in formalization, earnings and continuous basis (Box 4.1). working conditions is difficult, however, due to the absence of adequate survey instruments. Labor earnings Household surveys have proved very useful in Vietnam to assess poverty and measure progress in human development. But these surveys are Low labor costs remain one of Vietnam's main too demanding to implement, so that they are sources of competitiveness, especially when conducted only every two years. And their taking into account that Vietnamese workers labor modules tend to be succinct. A reliable learn fast and have a high work ethic. But labor monitoring of labor market developments, as earnings have been increasing rapidly. Box 4.1: Towards a Better Monitoring of the Labor Market How economic transition, rapid growth and global integration are affecting job quality and working conditions in Vietnam remains an open question. One important problem is the lack of high-quality tools to monitor labor market developments. Even assessing the evolution of simple indicators such as the share of wage workers in total employment is difficult. For instance, both the Labor Force and Employment Survey (LFES) implemented by MOLISA and the household expenditure surveys conducted by GSO put that share at around 17 to 19 percent in the years around 1996 to 1998. And both report a substantial growth in the share in subsequent years. But around 2004 the LFES estimates the share to be 26 percent, whereas the VHLSS puts it at 32 percent. Other problems relate to the ambiguity of some of the concepts and definitions used. For instance, some wage earners may work in the informal sector, and some self-employed in the formal sector. A diagnosis of the existing surveys in Vietnam shows that none of them provides reliable estimates of informal employment. The LFES concentrates on main occupation, and does not collect information on labor contract, benefits, or the characteristics of the employer. The VHLSS is probably the best data source at present. Nevertheless, being mainly designed to obtain income and expenditures estimates, it is weak on labor market indicators and the accounting of informal sector production. As for the Annual Household Business Survey (AHBS), its main shortcoming is to exclude non-located enterprises, an important component of the informal sector. This has resulted in unsolved debates as to how many non-farm household businesses there are in Vietnam. In 2007, taking advantage of the transfer of the LFES from MOLISA to GSO, the questionnaire design was significantly improved. The number of questions increased from around 25 to 50, allowing an enlarged coverage of labor markets indicators. The August 2007 survey will for the first time capture informal employment in Vietnam, including both primary and secondary jobs, and both jobs in the informal sector and unprotected jobs in the formal sector. The new questionnaire also ensures a better compatibility with international labor market concepts and definitions (for instance, on unemployment and underemployment). Meanwhile, a household business and informal sector survey has been launched. Following the principle of mixed surveys (covering both households and enterprises), it focuses on unincorporated enterprises identified as employers or own-account workers in the LFES sample. Two representative pilot surveys have been conducted (in Hanoi and HCMC) and scaling up to the national level is scheduled for 2008. In the medium run, the objective should be to consolidate the LFES and bring it up to the best international standards. To attain this goal, it should become a continuous survey, able to provide quarterly estimates of the main labor markets indicators. It should also include a panel component, whereby the same respondents are interviewed more than once over time, so as capture employment and earnings transitions. And it should allow for flexible modules to address specific labor market issues, such as gender, training or job quality. Source: Based on Mireille Razafindrakoto and Francois Roubaud (2007). 37 SOCIAL PROTECTION The exact speed depends on the indicator reported that they were higher or much higher. considered. The average earnings of a full-time worker in domestic private enterprises provide a A comprehensive analysis of job transitions, useful perspective. In 1998, those earnings using a nationally representative sample of stood at 554 thousand VND per month; by households, reveals a more complex picture. A 2006, they had escalated to 1.420 thousand. The comparison of labor earnings across all figures include bonuses and overtime pay, as occupations between 2002 and 2004 shows that well as labor earnings from secondary considerable gains were also made by moving occupations. They exclude the value of out of wage employment into self-employment. payments in kind and benefits such as health This may reflect the creation of own businesses, insurance or access to old-age pension. Based after a phase accumulating savings and on this indicator, nominal labor earnings knowledge in a paid job. The analysis of increased by a cumulative 12.5 percent per year household expenditures, instead of individual during this period. Given that consumer prices labor earnings, yields a similar result, with gains increased by 4.2 percent during the same period, being particularly important for households who the annual growth rate of labor earnings, in real were not engaged in farming before moving into terms, was 8.3 percent. opening their own businesses. At the other end, those who had to complement their farming Part of this growth in labor earnings was activity with some wage employment, and those associated with workers getting better paid in who remained engaged in farming exclusively, their jobs; part resulted from workers moving fared much worse. from lower to higher paying jobs. Occupational and geographical mobility have actually been a Another way to understand the mechanisms fundamental part of this moving-up process. which have allowed so many Vietnamese workers to secure higher earnings is to look at Needless to say, labor migrants to urban areas their individual characteristics. An analysis of often face considerable hardship, and their pay the determinants of labor earnings highlights the and working conditions are far from ideal. A importance of education (Figure 4.1). In the study of labor migrants to cities conducted in early stages of economic transition, there was 2004-2005 reported earnings substantially relatively little differentiation in labor earnings. below those of city residents. Most of the Pay compression was deliberate in the public migrants made less than 1 million VND a sector. Out of it, the redistribution of month; two out of five less than 600 thousand a agricultural land to rural households had month. Women were much more likely to be in generated relatively similar opportunities for the lowest paying group. More than half of the mostly everybody. However, with the migrants interviewed also reported receiving development of the market economy, those with unstable income. However, when asked if they higher skills have faced increasingly better knew about the expected difficulties before they earnings opportunities. Beyond fluctuations migrated, four out of five said they did. And an that may be attributed to quality data, the astounding 95 percent of them declared that in increase in the returns to education is one the spite of those difficulties migrating had been a salient trends of this period. In 1993, an sound decision. additional year of schooling resulted in labor earnings being only 1 percent higher on These findings indicate that labor migration in average. By 2002, the corresponding gain was Vietnam takes place with high risks and costs, 6 percent and since then it roughly stabilized. but on average the payoffs still outweigh the This widening gap stresses the importance of hardship, especially taking into account the equality of opportunity in education. limited opportunities for gainful employment in rural areas. When asked to compare their The development of a market economy might earnings in the cities with those they could make have been more of an equalizer in other had they stayed in their place, again 95 percent respects. In particular, the gender gap in 38 LABOR MARKETS Figure 4.1: Determinants of Labor Earnings Source: Based on data from GSO. The figures are based on the estimated coefficients of a Mincerian equation explaining hourly earnings in the main occupation of the respondent. earnings appears to have declined during the substantial economic restructuring. Recent 1990s. Other things equal, a woman at work experience also shows that important job losses used to earn 24 percent less than a man in 1993. could result from anti-dumping measures by By 1998, the gap had narrowed to about 15 Vietnam's trade partners. It is therefore percent and it remained at roughly that level important to understand who the unemployed since then. are, and what measures could be taken to facilitate their transition to work. Other salient trends include the catch up of earnings in government, with civil servants So far, the bulk of the unemployed are young, being underpaid in the 1990s and experiencing a urban people with secondary or even tertiary recovery subsequently, especially around 2004 education. Very few of them are unemployed where minimum wages were raised due to job loss or dismissal (Figure 4.2). substantially. Workers in smaller cities also appear to have experienced important gains. Data from the 2003 National Survey Those in larger cities clearly earn much more, Assessment of Vietnamese Youth provide but the fluctuations of the estimate (especially in valuable insights into the employment 2002) cast doubts on its precision. opportunities and paths of this population group. Overall, 57 percent of working youth are Unemployment self-employed, with two thirds of them doing unskilled agricultural jobs. Even after two While unemployment has not been a major issue decades of economic reforms, only a modest 6 in Vietnam so far, rapid global integration is percent of the employed youth reported working expected to lead to increased competition across in domestic private enterprises and 5 percent in all sectors, and this in turn could result in foreign-owned companies or joint ventures. 39 SOCIAL PROTECTION Figure 4.2: Who Are the Unemployed? Source: Based on 2005 data from MOLISA. A statistical analysis of the probability of being as well as engineers. According to the labor at work shows that it is higher for males than for force survey conducted by MOLISA, domestic females, and it also increases significantly with companies face similar challenges, especially age. Although marital status is not strongly for craftsmen. associated with work, younger single youth are more likely to be looking for jobs than older High job turnover is a key factor explaining the married youth. Education also exerts a strong perceived skill bottlenecks. In the survey of influence on the probability of working. More Japanese manufacturing firms operating in educated youth are both less likely to be Vietnam, low worker retention was mentioned working and more likely to be looking for as an issue by about 30 percent of respondents. suitable employment. This is especially so in In the same vein, a study on foreign-invested the case of university graduates. companies showed that over the period from 2001 to 2003, the labor turnover rate was more Protracted job search among educated youth than 43 percent. Turnover was highest in may not necessarily reflect a shortage of textiles, garments and footwear enterprises. employment opportunities, as there is at the According to information provided by the same time a dramatic shortage of skilled affected companies, 32 percent of the departing personnel in Vietnam. A series of surveys workers had moved to other foreign companies, suggest that unemployment coexists with large 23 percent had gone on to establish their own numbers of vacancies. Lack of skilled labor business and 18 percent had taken jobs in was mentioned as a severe bottleneck by one domestic enterprises. fifth of respondents to the 2005 Investment Climate Survey, and as a moderate bottleneck Which policies should be put in place to assist by an additional fifth. Skill shortages were a the unemployed while they search for a job is serious obstacle in electronics, but they also not so clear. Vietnam has an affective safety net appeared to be significant in chemicals, for redundant workers from SOEs. This safety machinery, furniture and even textile. A 2006 net was set up in 2001 to overcome resistance survey of Japanese manufacturing companies to the equitization process and provide fair operating in Vietnam revealed difficulties in compensation to those who stood to lose recruiting staff at the middle-management level, because of it. Since then, it has supported more 40 LABOR MARKETS than 4 thousand enterprises and paid relatively would cover workers in the formal sector and be generous severance to close to a quarter of a funded through higher social security million redundant SOE workers (Table 4.2). contributions. The readiness to implement a Two tracer studies conducted over this period program to support the unemployed is have revealed relatively high levels of commendable at a time when global integration satisfaction among the assisted workers, and may lead to unexpectedly large shocks, maybe especially among women and the old, who resulting in substantial job losses. If that were were potentially the most vulnerable to to occur, having a mechanism in place to restructuring. provide assistance to the unemployed without delays would be of great help. However, The activity of this safety net slowed down whether such mechanism should be activated recently because the time frame for its operation for now is questionable. was based on the optimistic assumption that equitization would be completed by the end of The main risk in introducing an unemployment 2005. With some of the biggest SOEs still insurance scheme is to raise the cost of labor, needing to go through this process, there was a hence to slow down the formalization of need to reactivate the fund and provide it with employment, without really helping those who an appropriate budget allocation, which only lose their jobs. happened in early 2007. The decision was also Unemployment insurance is clearly ill-adapted made to extend its scope to redundant workers to assist young, educated job seekers. Most of from state-owned forests and farms, and to them have a limited work experience, or no update the design of the severance pay package. work experience at all, so they would not have Implementing these changes has also taken contributed to the scheme long enough to longer than expected. But the political will to qualify for any benefits. Moreover, if keep this safety net fund in operation is clear. unemployment among this group reflects a protracted job search for opportunities that are There is also a decision, stated in the Social commensurate with their qualifications and Insurance Law, to introduce an unemployment expectations, financial support could simply insurance scheme, in the spirit of those that can result in extending their unemployment spell, by be found in industrial countries. Such scheme making job seeking more affordable. Table 4.2: The Safety Net for Redundant SOE Workers Spending Year Workers assisted SOEs Average package assisted (VND million) (VND million) 2002 1,147 34 29,262 25.5 2003 18,445 453 534,973 29.0 2004 43,659 873 1,298,738 29.7 2005 86,483 1,445 2,812,933 32.5 2006 49,683 953 1,748,329 35.2 2007 19,401 367 948,616 48.9 Total 218,818 4,125 7,372,851 33.7 Source: Based on data from Corporate Finance Department, Ministry of Finance (MOF). 41 SOCIAL PROTECTION Unemployment insurance is also ill-suited to from abuse and keeping formal enterprises help other potentially vulnerable groups of competitive. Policies influencing wage setting workers. The entry of large supermarket are especially important in this respect. chains in Vietnam, with their more efficient supply chains and management techniques, Vietnam introduced a minimum wage for could affect the earnings of household foreign-invested companies in 1990 and one for businesses engaged in retail trade. But the domestic enterprises in 1993. The latter is the members of these households would not be same for the entire country. Since 1992, the eligible to claim unemployment benefits former varies by region, depending on the because they would not have contributed. As province and on whether the companies are for workers in the formal sector, under the located in urban or rural districts. Minimum Labor Code they are entitled to a reasonable wages were adjusted in several opportunities amount of severance pay. Admittedly, that over the years, with the minimum wage for entails an additional cost for the restructuring domestic enterprises growing faster (Table 4.3). enterprises. But the additional contribution The ratio between minimum wages and average towards an unemployment insurance scheme wages in domestic enterprises has increased would represent an additional cost for all moderately, from 25 percent in 1998 to 32 enterprises, not just those undergoing percent in 2006. restructuring, at a time when minimum wage In spite of these adjustments, the minimum hikes and reforms to social insurance programs wages applying to foreign companies remain already risk increasing labor costs substantially above the domestic minimum substantially. wage. The margin varies from 53 to 93 percent An unemployment insurance scheme would not depending on the region. This margin is be a useful substitute for the safety net for incompatible with the national treatment redundant SOE workers either. No standard principle of the WTO. Vietnam is thus scheme would be able to match the generosity confronted with the need to unify the minimum of the current safety net. Without such wage applying to domestic and foreign generosity, the equitization of large SOEs could companies. And because the latter cannot be be stalled. The efficiency gains that can be reduced, unification will require that the expected from bringing business-oriented minimum wage applying to domestic outsiders into the management of large SOEs enterprises be increased rapidly. The prospect are bound to be sizeable. In fact, completing the of affecting employment creation and restructuring of the state sector is one the last formalization is thus real, and requires the major hurdles for Vietnam to fully become a adoption of a sound policy strategy. market economy. The safety net for redundant A first step is to identify who the minimum SOE workers can be seen as a compensating wage earners are. The characteristics of full- mechanism, to avoid making losers along the time workers whose earnings fall within a way, rather than as a pure social protection narrow range of the applicable minimum wage mechanism. From that perspective, replacing it provide a good indication in this respect. Using by unemployment insurance benefits would not a 20 percent band (from 10 percent below to 10 be advisable. percent above) and data from 2004, it appears that some 2 to 2.7 percent of all wage earners Wage setting are paid the minimum wage. The range in the estimates is due to the various defensible ways Supporting rapid employment generation, to compute labor earnings. If part-time and especially by the formal sector of the economy, casual workers are considered, and the is the most effective way to address the risk of minimum wage is computed on an hourly basis, unemployment. To achieve this goal a good some 3.1 to 6.1 of wage earners appear to be at balance is required between protecting workers the minimum. 42 LABOR MARKETS Table 4.3: Minimum Wages and Average Wages Domestic enterprises Foreign-invested companies Average wage Minimum wage Minimu Average m wage Household Private Region Region Region wage Year enterprises enterprises SOEs 1 2 3 1998 144 552 554 572 464 530 597 680 2002 210 606 771 1002 487 556 626 1037 2004 290 649 852 1077 487 556 626 1044 2006 450 914 1420 1680 810 790 710 1774 Source: Based on data from MOLISA for minimum wages and from GSO for average wages. Figures are in thousand VND per month. Region 1 includes Hanoi and HCMC. Region 2 covers their suburbs, several other large cities and some districts with industrial zones. All the rest of the country is under Region 3. About half of these minimum wage earners are down the formalization of the economy. between 15 and 25 years of age, and more than 62 percent are women. An impressive 85 The 2002 amendment of the Labor Code and percent have completed lower secondary subsequent regulations stipulated that the education, but less than a fifth has any technical minimum wage had to be adjusted over time in qualification. Not surprisingly, almost half do accordance with consumer prices, economic unskilled jobs. Minimum wage earners are growth, and the balance between labor supply heavily concentrated in manufacturing and demand. industries (nearly 42 percent) and in foreign- While this is an eminently sensible approach, invested companies (13 percent). In absolute implementing it in practice is not numbers, however, the largest share (more than straightforward. The study by ILSSA tried to 55 percent) corresponds to household translate the criteria introduced by the 2002 businesses and small enterprises. About half of amendment of the Labor Code into specific minimum wage earners belong to poor or near- indicators. For consumer prices, the study poor households. But the fraction of household considered the cost of a consumption bundle heads and spouses is declining rapidly. securing living standards close to the poverty A second, more difficult step is to assess the line for the worker and one dependent child. In effects of minimum wage increases on labor relation to economic growth, the research market outcomes. Preliminary findings of an estimated the relationship between the analytical effort led by ILSSA suggest that the minimum wage and Gross Domestic Product (GDP) per capita. And for labor market effect on average wages is substantive. But conditions, it estimated trends in the average there is no noticeable effect on poverty rates labor earnings of workers with the same among minimum wage earners. This may be characteristics as minimum wage earners. due to a negative impact on formal sector employment. This impact appears to be minor, For 2010, these different methodologies yield and disappears when the overall probability of levels of the minimum wage ranging from 790 being employed is considered, instead of the to 930 thousand VND per month. But this could probability of working in the formal sector. be an underestimate, for two reasons. First, the These preliminary findings suggest that assumptions used by ILSSA in its simulations minimum wage hikes may not hinder are too optimistic in relation to the inflation rate. employment creation much, but they could slow Second, the forecasts for 2010 are close to the 43 SOCIAL PROTECTION current minimum wage levels in foreign- percent of SOEs, one third of private invested companies. It is unlikely that workers enterprises, and half of foreign-invested in these companies would be satisfied if there companies had a trade union by 2005. And even was no increase of the applicable minimum this could be an overestimate. wage between now and then. Vietnam is thus confronted with the need to double, if not to Most importantly, there are serious deficiencies triple, the domestic minimum wage over just a in the representational funcion of existing trade few years. unions at the enterprise level. The majority are led by the management staff of the enterprise. Absorbing such an increase without losing in Trade unions at enterprise level also have competitiveness could be a serious challenge for problems of capacity in terms of representation many domestic enterprises. The government and bargaining skills. VGCL estimated that 96 has sensibly chosen to introduce a regional percent of SOEs had collective agreements in distinction for all minimum wages, in the spirit force in 2006. But the reported fraction dropped of the one in force nowadays for foreign- to 40 percent in foreign-invested companies, invested companies. In this case, the increase in and further to 25 percent in domestic private minimum wages would be fastest in Hanoi and enterprises. HCMC, and slowest in the least developed provinces. A study of industrial relations in selected provinces reached similar conclusions. Only The doubling or even trebling of minimum 10.4 percent of enterprises in Ha Tay had wage levels over a short period of time could collective bargaining agreements in 2005. The also have an adverse impact on public finance. share was as low as 3.7 percent in Khanh Hoa. The salaries of civil servants are determined Not only there is a limited number of enterprises through formulas involving the domestic with collective bargaining agreements, but the minimum wage and coefficients related to coverage of existing collective agreements is occupation and seniority, plus a series of also quite restrictive. Many of them are simple bonuses and additional payments. Because of replications of the Labor Code rather than the the latter, a minimum wage hike would result in outcome of a real negotiation between workers a smaller increase of the wage bill. But the and employers. Wage bargaining is thus the increase could still put public finance under exception rather than the norm. strain. The domestic minimum wage also enters in the benefit formulas determining the pensions Meanwhile, labor conflicts have been on the of retired civil servants. Therefore, government rise, especially among foreign-invested liabilities would also increase substantially in companies (Figure 4.3). The above-mentioned the event of a minimum wage hike. This study of industrial relations found that 90 stresses the importance of separating minimum percent of strikes had happened due to violation wage policies from civil service pay and of labor regulations by employers, and 80 pension formulas. percent was related to wage issues. However, in the absence of a well-established negotiation Industrial relations mechanism at the enterprise level, most of these conflicts had been processed outside the legal According to Vietnam's labor regulations, any framework and without trade union enterprise employing five workers or more must involvement. The increase in the number of have a trade union or a temporary trade union labor conflicts has been related to issues such as executive established within its first six months salaries, working time, meals and working of operation. Yet, compliance is very limited. conditions, and they do not necessarily entail a The majority of enterprises in Vietnam do not violation of legal rights by employers. Labor have a representative with the right to bargain conflicts also result from the lack of adjustment on behalf of workers. According to the Vietnam of the minimum wage for foreign companies. General Confederation of Labor (VGCL), 90 Many workers have been paid the same wage 44 LABOR MARKETS Figure 4.3: Number of Strikes by Enterprise Ownership Source: Based on data from MOLISA. level for many years, while the cost of living has of Commerce and Industry (VCCI) and the considerably increased. Vietnam Cooperative Association are appointed to be the national representatives of employers, The lack of workers' representation at the much the same as VGCL represents workers. enterprise level and the prevalence of wildcat But no provisions exist for negotiation at strikes suggest that the legal framework for sectoral or regional levels. For instance, there is collective bargaining of Vietnam is not well no legal representative actor of employers in adapted to the maturity of its economy. Current sectors such as garments, textiles or cement. regulations are too stiff in only allowing a trade union executive committee or a temporary trade Other shortcomings of the legal framework union to raise proposals for negotiation and refer to the settlement of disputes. For engage in labor collective agreements. They do instance, there is no time limit for negotiation. not allow for more than one collective In the event of disagreements, the law requires bargaining agreement at the enterprise level. the Labor Conciliation Council to assist in They do not permit those working on piece finding a solution. But the Council is simply production to be associated with wage made of the representatives of the two parties in negotiations, or to engage in industrial action. the negotiation. In the absence of an And they do not protect workers serving as trade agreement, the involvement of an independent union representatives from harassment by third party would be required. Labor abusive employers. administration agencies could play this role. However, they are not allowed to intervene Current labor market regulations also ignore unless requested to do so by the two parties. the possibility of collective bargaining taking And even then, their responsibility is described place at higher levels. The Vietnam Chamber in vague terms. 45 SOCIAL PROTECTION Box 4.2: Trade Unions in Vietnam and in China Trade unions in China and Vietnam continue to encompass all types of 'workers', including top managers. This originated from the nature of socialist employment relations where it was perceived that no conflicts of interests existed between management and workers. This perception has easily become a convenient tool for management to dominate the workplace and control the unions. In both countries, trade union leadership at the enterprise level is disproportionately dominated by staff members representing the management. However, there are important differences as well. Rank-and-file workers in Vietnam have displayed a great degree of spontaneous solidarity, outside the official unions and the official legal framework, to defend and advance their rights and interests. Field research indicates that most wildcat strikes were not only well planned and coordinated, but were also participated in by a majority of the workers within the enterprises concerned. The so-called contagion effect of wildcat strikes illustrates the Vietnamese workers' ability to take well coordinated collective actions. For example, on December 28, 2005, 18,000 workers of a Taiwanese owned company in Linh Trung Processing Zone in HCMC went on a wildcat strike demanding a wage increase, this move was immediately followed by strikes in two other factories in the same Zone on the same issue. What is also noticeably different from China is the response of the public authority to wildcat strikes. In Vietnam the local labor administration will usually send its officials to investigate the situation and to resolve the strike situation by offering a compromise. This intervention tends to legitimatize workers' actions. For instance, managers are quietly advised by local government officials not to take any retaliatory action such as dismissal against labor leaders, as it would only provoke another round of strikes. Of 50 strikes for which detailed information is available, workers' key demands were met in 48 cases. Favorable coverage of the collective actions by local media also generates pressures on the local government to intervene. Trade unions at higher organizational level in Vietnam tend to display stronger support for workers on strikes than their counterparts in China. Local trade union cadres usually accompany the government conciliators, putting pressure on the management side to accept reasonable demands while also persuading workers to return to work. In response to the strikes in Linh Trung Processing Zone in December 2005, VGCL actually criticized the government for its failure to raise the minimum wage for foreign-invested companies for a long period of time. The strong solidarity of workers, the tolerant attitude of the public authorities, the supportive response of the general public and the relatively autonomous behavior of higher level trade union organizations all differentiate Vietnam from China. Source: Chang-Hee Lee (2006b). In late 2006, a number of articles of the the new authority responsible for settling Labor Code regarding the resolution of disputes in the second stage of the resolution collective labor disputes were amended, with process. several implementing regulations issued during 2007. The changes separate resolution Improving the legal framework for collective mechanisms for rights-based and interest- bargaining would not only provide a better based disputes. Importantly, they allow channel for the resolution of industrial conciliation by the district labor officer, the disputes: it could also help Vietnam counter first step of the dispute resolution process. anti-dumping measures by trade partners. The Both employer and employees need to agree absence of satisfactory wage negotiation in written form on the selection of the mechanisms at the enterprise level is invoked district-level labor conciliator. The chairman as evidence that Vietnam is still an NME. of the district People's Committee becomes Under WTO agreements, an NME is more 46 LABOR MARKETS vulnerable to anti-dumping measures, because make this move (Box 4.3). A more adequate the production costs of other (so-called framework could include more flexible surrogate) countries can be used to measure mechanisms for employees and employers to the extent of subsidization. appoint their representatives at various levels. It could also allow piece workers to be part of Vietnam has accessed the WTO with an NME collective agreements, and provide protection to status which will expire automatically in 2018. employees' representatives. Introducing time Introducing a more adequate industrial relations limits for negotiation and replacing the Labor framework could hold the key to advancing this Conciliation Council by conciliators who are date. And there are valid reasons to believe that truly independent from both parties would be Vietnam is in a relatively strong position to another change worth considering. 47 5. TAXES AND FEES Before any resources can be spent by the should also allow a better recovery of the costs government on social protection policies and of infrastructure development, targeting its most programs, they have to be raised through direct beneficiaries. But property tax requires taxation. Securing stable government revenue good registration and reliable valuation of land is one of the main priorities of the current tax titles. More generally, a proper operation of the reform agenda. But the tax base is evolving tax system has strong administrative rapidly, requiring an adjustment of both tax requirements. Vietnam has adopted an instruments and tax administration. SOEs ambitious agenda to modernize tax continue to account for the bulk of revenue administration, but implementation will be collection, but the share of foreign companies challenging. Changing business processes from and private domestic firms is growing rapidly. systematic inspection towards self-declaration The structure of revenue by tax instrument is will be necessary to expand the number of changing as well, with tariffs and other trade- taxpayers, reduce the cost of compliance and related taxes losing share as a result of global mitigate the risk of corruption. This change is integration. The way those tax instruments is now being rolled out, with encouraging results. designed has an impact on both equity and efficiency. Taxes and fees introduce a wedge An evolving tax base between gross and net income and thus redistribute resources across households. In As part of the transition to a market economy, Vietnam, they are progressive overall, with the two major tax policy reforms were undertaken tax burden increasing with household in Vietnam in 1990 and 1998. A expenditures. But there is a wide variation comprehensive legal framework was across instruments. Value Added Tax (VAT) is developed as a result. Taxes imposed on state, progressive, some local fees are regressive, and non-state and agricultural sectors were Personal Income Tax (PIT) affects only a tiny integrated into major instruments, notably fraction of households. The burden on turnover and profit taxes. Subsequently, these enterprises varies too. Disturbingly, it appears were transformed into VAT and Corporate to be higher for domestic private enterprises, for Income Tax (CIT) respectively. By now, the small firms and for those located in rural areas. system in place also includes natural resource The cost for enterprises to comply with tax taxes, PIT, excises, customs duties, and a regulations is high as well. But estimates vary number of minor taxes, fees and charges. widely, suggesting the need for reliable Considerable efforts have also gone into indicators to monitor the efficiency and integrity reforming the legal framework for tax of tax administration. In a country where administration. wealth is very much associated with the ownership of valuable urban land, property Despite the reforms already undertaken, many taxes have the potential to redistribute resources difficulties remain. The number of potential across households without introducing the taxpayers is growing faster than current inefficiencies associated with taxes on arrangements can handle. The large set of household or enterprise income. Property taxes minor taxes, charges and fees results in 48 TAXES AND FEES unnecessary complexity. Efforts have been enterprises and by 4.3 in the case of foreign- made to shift the responsibility for collection owned companies (Figure 5.1). Yet, in spite of of these taxes to local governments, but the buoyancy of the mainly small-scale decentralization creates additional accounting domestic private sector, the 673 largest and supervision problems. Taxpayers taxpayers still contribute more than 70 percent complain about ambiguities and contradictory of total domestic tax revenues. provisions in the tax system, which increase the compliance burden and the discretionary There are also ongoing changes in the power of tax inspectors. To address these structure of revenue by tax instrument. CIT concerns, a third wave of reforms to tax policy and VAT remain the workhorses of the system and tax administration is currently underway. (Figure 5.2). In particular, VAT is a highly "productive" tax in Vietnam, in the sense that Amidst these transformations, the structure of the revenue (measured in percentage points of government revenue is changing as well. GDP) per point of the basic rate (in percent of Overall revenue has increased slowly but value added) is among the highest in the steadily in recent years, from 20.5 percent of region. However, the planned revision of the GDP in 2000 to 23.4 percent in 2007. This VAT, which includes the establishment of an increase is mainly due to the growth in oil exempt threshold and an enhancement of the revenues and in non-tax revenue. Tax revenue refund mechanism, may result in lower proper actually experienced a minor decrease revenue. during the same period. SOEs remain the most important taxpayers, accounting for On the other hand, greater global integration some 54 percent of CIT and 42 percent of VAT has led to a diminished importance of taxes revenue. Revenue from SOEs doubled related to international trade. The share of between 2001 and 2006, whereas it multiplied import and export duties in total tax revenues by 3.3 in the case of domestic private has decreased from 28.9 percent in 1996 to Figure 5.1: Who Pays Taxes in Vietnam? Source: Based on data from MOF. 49 SOCIAL PROTECTION Figure 5.2: Tax Revenue by Source Source: Based on data from MOF. 17.8 percent in 2003 and further to 13.1 several taxes and fees. As consumers, they percent in 2007. The implementation of WTO pay VAT and excise taxes. As residents, they commitments will reinforce this trend, face a range of local fees such as agricultural although not necessarily in the short term. taxes, education fees and taxes on household The commitments made by Vietnam do not businesses, among others. Overall, these entail dramatic reductions in tariffs. As the taxes are progressive (Table 5.1). In 2006, a reductions are mainly at the high end, import household in the richest quintile of the volumes (the base on which taxes are population devoted 14.8 percent of its computed) are bound to growth so that spending to paying taxes and fees, compared revenue could actually increase, at least for to 8.7 percent for a household in the poorest some time. Moreover, WTO commitments quintile. But not all taxes and fees are equally allow raising the tariff lines for a range progressive. VAT, excise taxes and taxes on number of items from zero at present to a household businesses are, while agricultural maximum of 5 percent. This should also fees are not. And not all households pay all result in higher revenue. taxes and fees. For instance, mostly everybody is confronted with VAT whereas a From gross to net income meager 0.1 percent of households pay PIT. Taxes and fees not only raise revenue for the In industrial countries, PIT is one of the main government: they also introduce a wedge tax instruments for redistribution. Its role is between gross and net income, and thus typically more modest in developing redistribute resources across households. countries, given that many households are Some times, such redistribution is a deliberate engaged in farming or work in the informal choice, accomplished through a variety of sector, where incomes are difficult to exemptions, thresholds and differential tax document. The potential for evasion is high at rates. Quite often, however, the complexity of the top of the distribution as well, as profits the tax system leads to unplanned (and even and capital gains are relatively easy to hide unknown) redistribution. from tax agents, some times with their complicity. That leaves civil servants and In Vietnam, households are directly subject to workers in the formal private sector as the 50 TAXES AND FEES Table 5.1: How Progressive Are Taxes and Fees? Taxes as percent of household expenditure per capita House- All holds Poorest Near Middle Near Richest house- paying quintile poorest quintile richest quintile holds taxes (percent) Value-added tax 4.7 6.0 6.7 7.4 7.6 7.0 100.0 Excise taxes 1.2 1.4 1.6 1.7 2.2 1.8 97.3 Taxes on household enterprises 0.1 0.3 0.6 1.8 2.4 1.6 15.4 Educational fees 1.2 1.6 1.8 1.9 1.3 1.5 62.1 Agricultural fees 0.7 0.9 0.6 0.3 0.1 0.4 40.2 Personal income tax - - - - 0.1 0.1 0.2 Other fees 0.8 0.8 0.8 0.9 1.1 1.0 99.0 Total tax paid 8.7 11.0 12.1 14.0 14.8 13.4 Expenditure per capita (thousand dong per year) 2,469 3,982 5,529 7,980 17,106 7,411 Source: Jonathan Haughton and Phuong Viet Ngo (2007). main taxpayers for PIT. Because it falls income, with the proportion being higher mainly on formal wage and salary earnings, among richer population groups. However, PIT can be seen as disincentive to not all wage income comes from the formal formalization. sector, and only a fraction of it (presumably among the upper quintiles of the population) is In spite of these well-known problems, the likely to be subject to PIT. government of Vietnam is trying to scale up its PIT. A new law was promulgated in late A more comprehensive assessment of tax 2007, and from 2009 onwards all individuals incidence should take into account the indirect making more than 4 million dong per month impact of taxes on consumer prices, through will be subject to PIT. Implementation is production costs. For instance, taxes on fuel expected to increase the number of taxpayers affect households who do not pay them to about 2.3 million, and result in 13 trillion directly, because they raise the cost of dong in additional tax revenue. This transportation. To assess the impact of compares with about 300 thousand high- upstream taxes on inputs it is necessary to use income taxpayers and six trillion dong in information on the cost structure of goods and revenue at present. The main source of services. household income targeted by PIT is wage earnings. But the PIT law also applies to The only study of this sort available for income related to equity shares, giving Vietnam was conducted using household individual investors two options: a 20 percent survey data from 1998 and the input-output tax on the profits they make or a 0.1 percent table for 1997. This study was able to trace tax on the sale of stocks. the incidence of approximately half of all tax revenue. Its findings suggest that taxation In Vietnam, wage employment is the source of remains progressive even when considering roughly one quarter to one third of household indirect effects. But the gap between the 51 SOCIAL PROTECTION burden of the richest and the poorest quintile payments have moved closer to the benchmark is lower than when direct taxes alone are over time, which is reassuring (Figure 5.3). But considered. they also suggest that the gap with the benchmark is wider in the case of SOEs, large Taxing capital firms and enterprises located in urban areas. Whether this pattern is due to deliberate Taxes can also affect the incentives faced by incentives, to weak tax administration or to firms. In the case of Vietnam, the tax burden evasion is unclear. But whatever the reasons, should be the same across all enterprises, and imposing a higher (actual) tax burden on small amount to 28 percent of corporate profits. In firms, private businesses or enterprises located practice, however, the tax burden may differ in rural areas does not seem to be in the best from that benchmark. Policies to encourage interest of Vietnam. investment in laggard regions, or dynamic However, there are other, more indirect ways in sectors, or labor-intensive technologies, usually which taxation can become a burden on rely on tax incentives, which in turn should enterprises. The actual cost of complying with result in a lower burden for the encouraged taxes, in terms of hours of skilled personnel enterprises. But poor enforcement, arrears and devoted to filling forms and negotiating with tax even corruption may lead to substantial officials, has been a hotly debated topic in disparities in the actual tax burden across Vietnam in recent years (Box 5.1). enterprises of different sorts. The burden from corruption is also difficult to Individual records from enterprise surveys assess, but not less real. A diagnostic study allow comparing actual tax payments to the prepared in 2005 by the Central Committee for hypothetical benchmark. The latter can be Internal Affairs of the Party identified tax computed as 28 percent of the declared pre-tax administration as one of the government profits. The comparison shows that actual tax agencies most vulnerable to corruption. This Figure 5.3: The Tax Burden on Enterprises Source: Based on data from the GSO enterprise survey. Figures are for 2005 unless otherwise indicated. 52 TAXES AND FEES Box 5.1: How Costly Is It to Comply with Tax Regulations? Assessments of the cost of complying with taxes can be sensitive to the methodology used. Three recent attempts to measure such cost in Vietnam have yielded substantially different results. Investment Climate Assessment (ICA). Conducted by the World Bank in 2005, it surveyed a representative sample of firms in manufacturing across 25 provinces in Vietnam. The focus of the survey was not to estimate the tax burden, but rather to assess the broader investment climate However, one of the questions referred to the time spent dealing with tax inspections, including the preparation of relevant documentation. According to the results, firms were subject to an average of 2.5 inspections per year, lasting 9.5 hours on average. From these figures, it could be inferred that firms need to spend about 24 hours of work to comply with taxes. But tax inspections, no matter how burdensome, are only part of the overall compliance cost. Therefore, the ICA produces an under- estimate of tax compliance costs. Doing Business. Also conducted by the World Bank, this is an attempt to quantify various dimensions of the overall business environment across countries. One of those dimensions refers to the costs of complying with taxes. In this respect, the Doing Business methodology relies on expert assessments for a hypothetical enterprise with 60 employees and a turnover equivalent to 0.7 million dollars per year (in Vietnam's case). In all the assessments conducted from 2005 to 2007, the time needed for this enterprise to be current in its obligations was 1,055 hours per year. This estimate refers to CIT, VAT, PIT and labor taxes. It includes the time to prepare tax returns and to keep books. However, the latter may be incurred anyway, regardless of tax payments. Also, few enterprises pay PIT in Vietnam, and the estimate assumes that electronic filing is not used. Therefore, the Doing Business methodology over-estimates compliance costs. Central Institute for Economic Management (CIEM). This government-led evaluation was conducted in 2007 using a "standard cost model" developed by the Dutch government. The model relies on 360 interviews with small and medium enterprises. In the case of Vietnam, the average time to comply with taxes was estimated at 1,959 hours. Most of this cost (1,733 hours) was associated with VAT, which respondents seem to consider particularly annoying. However, this evaluation mixes the one-time cost of obtaining a taxpayer number with the recurrent cost of paying taxes. Also, the methodology makes it difficult to distinguish between activities which are part of doing business, activities specifically related to the payment of taxes, and more subjective "annoyance costs". A closer analysis shows that the estimate for the time spent on inspections is four times higher than in the ICA. The cost of dealing with most taxes is twice as high as in the Doing Business report, and the cost of dealing with VAT about 30 times higher. Therefore, the CIEM study is likely to have vastly over-estimated tax compliance costs. In order to avoid similar discrepancies in the future, a consistent methodology to monitor tax compliance cost and service delivery should be developed. Such methodology should be anchored on representative samples and well- tested questionnaires. Estimates should be produced on a regular basis, focusing on the costs (time and money) faced by different groups of taxpayers. Similar surveys of tax officers would allow for a more complete picture. Key metrics should include not only the cost of complying with taxes, but also indicators related to the efficiency of administration, the integrity of its officials and the satisfaction of taxpayers. conclusion was corroborated by an investment climate survey suggests that the size of the climate survey conducted among a bribes was not particularly large, but the representative sample of firms in 25 provinces, practice was prevalent also in 2005. Corruption was facilitated by the business processes of the tax administration Taxing property agency at the time, which involved inspections by tax agents to each individual taxpayer firm. A potentially more effective way to redistribute Those inspections were prone to negotiation on income across households while avoiding the profits to be declared. The investment distorting incentives is to tax property, and 53 SOCIAL PROTECTION especially valuable land. In Vietnam all land housing, to land rent and to the transfer of formally belongs to the state but land-use right LUCs. In addition, there are registration fees certificates (LUCs) can be transferred, rented, and cadastral fees. donated or used as collateral or as capital. As stipulated in the Land Law of 2003, local Across many countries, taxes on land and governments are in charge of approving and property are a reliable source of revenue for supervising land use. They can also apply taxes local governments, financing local expenses on and fees on land, and retain the revenue (Table key public services such as infrastructure, 5.2). The most common taxes apply to the education and health care. This is also the case value of agricultural land, urban land or in East Asia. For instance, land tax represents Table 5.2: Revenue Assignment by Level of Government Central Budget Local Budget Non-shared 1. Export/Import duty 1. Housing and land tax sources of 2. Value added tax on imported goods 2. Natural resources tax (except that on revenue 3. Excise tax on imported goods crude oil) 4. CIT 3. Licenses 5. Tax and other revenue from crude oil 4. Land transfer tax 6. Proceeds collected from government 5. Agricultural land use tax lending and capital contributions 6. Land-use fees 7. Non-refundable aid from central 7. Rental of land government 8. Proceeds from sale and lease of state- 8. Fees and charges owned houses and buildings 9. Unused revenue of previous fiscal 9. Registration fees years 10. Proceeds from lottery 10. Others 11. Proceeds collected from government lending and capital contributions 12. Grants for local government 13. Fees and charges 14. Proceeds from utilization of public land 15. Proceeds mobilized from business and individuals in accordance with regulations 16. Unused revenue of previous fiscal years 17. Supplement from central budget 18. Contributions from local people for construction of infrastructure in accordance with regulations 19. Voluntary contributions Shared 1. Value added tax sources of 2. CIT (except that of nationwide business activity) revenue 3. Tax on high income earners 4. Profit remittance tax 5. Excise tax on domestic goods and services 6. Petroleum fees Source: MOF. CIT from nationwide business activities, in which profits can not be assigned to a single province such as a post and telecommunication, electricity generation and distribution enterprises are assigned exclusively to the central budget. 54 TAXES AND FEES a quarter of local revenue in Japan. Even an encouragement to informality. Taxing among developing countries in the region, 13 capital returns, if pushed too far, can be percent of the revenue of local governments is detrimental to competitiveness. Both workers accounted for by land and building tax in and employers can move elsewhere to avoid Indonesia, and by real property tax in the paying taxes. But land cannot be moved. Philippines. Admittedly, property tax creates competition among localities, encouraging households and Property tax has several appealing features. It enterprises to settle in places where can be strongly progressive, as land and infrastructure and amenities are good while housing are among the most valuable assets a property taxes are not too high. But given the household can have. In Vietnam, fortunes enormous agglomeration effects taking place have been made at extraordinary speed thanks in the process of development, some places to capital gains on property. Such gains are are simply unique. There are not two Hanoi, or difficult to tax through PIT. At the same time, two HCMC. At most, there is a case to tax it is straightforward to exempt small plots of land more heavily than the buildings on it, so agricultural land from taxation, so as to avoid as to avoid discouraging residential and penalizing poor farmers. One legitimate commercial investment. concern refers to households with limited income who happen to live on plots which The main problem with property tax is that it appreciated considerably, simply because of requires good registration systems and reliable urban development. Paying tax on land could pricing mechanisms. Good registration be beyond their financial capacity and force systems avoid ambiguities on ownership and them to sell their property, thus destabilizing physical delimitation. Reliable pricing their livelihoods. However, this concern can mechanisms help set tax payments at a level be addressed by delaying the payment of the that is commensurate with the situation of the tax (plus interest) until the property is real estate market. To improve registration, it transferred, through sale or inheritance. is necessary to complete the issuance of LUCs on urban property, to update cadastral maps Property tax also allows recovering and to modernize the information systems of infrastructure development costs, at least land administration offices. Setting up reliable partially. The market value of land is very mechanisms for pricing is more complex. closely related to the density of the area and Administrative procedures may prove the quality of the surrounding infrastructure. insufficient given the vast number of plots to Every time a road is build, those living nearby oversee. As in other aspects of tax make a capital gain, some times substantial. administration reform, self-declaration may be But if the road is financed out of general tax a better alternative. There is obviously a risk revenue, they do not revert much of that gain that unrealistically low prices could be to society. Property tax helps recovering part reported, to avoid paying high taxes. However of the capital gain for government, under the this risk can be addressed by establishing a link form of a stream of revenue computed on the with rights for compensation in the event of new (higher) value of the land. From this land reclamation. If the government can set perspective, it contributes not only to fairness land compensation at the reported price, the but also to the development of local owner faces an incentive to align such price to infrastructure, one of the main priorities of actual market conditions. Vietnam at this stage of development. Tax administration Perhaps one of the biggest advantages of property tax is that it does not distort A comprehensive tax administration reform incentives as much as other taxes do. In a program was adopted by the Politburo in 2004. developing country, taxing wage earnings is Its objectives are to develop a modern, fair and 55 SOCIAL PROTECTION transparent tax system, promoting voluntary increase substantially the number of taxpayers compliance, improving taxpayer service and it can handle. It should also reduce the scope enforcement, and enhancing revenue for negotiation between enterprises and tax collection. These objectives seem largely in agents, hence for corruption. line with the values and expectations of the Vietnamese population (Box 5.2). There are some indications that this approach Implementing such ambitious reform program is paying off already. For several years, a will be challenging, however. mail-in survey of enterprises has been conducted to compare the quality of Revamping business processes so as to change governance at local levels, under the form of the relationship between tax agents and a so-called Provincial Competitiveness Index taxpayers will be critical to the success of the (PCI). The survey contains the following reform. A pilot project for self-assessment of statement: "Negotiations with tax authorities VAT in selected localities is now being rolled are an essential part of doing business" and out at the national level. It leads to the gradual asks respondents whether they agree. The replacement of tax inspections by the fraction that agrees declined from 75.2 declaration of tax liabilities by taxpayers percent in 2005 to 61.1 percent in 2006 to themselves. Tax filings are then subject to 44.7 percent in 2007. The last two surveys analysis, with inspections taking place only included all provinces in Vietnam. Their when objective indicators suggest an improper comparison is telling about the progress assessment or an intention to fraud. This made through the introduction of self- change in business processes should make it assessment (Figure 5.4). But it also suggests possible for the tax administration agency to that more progress can be made. Box 5.2: Attitudes towards Tax Evasion Opinion surveys can be used to gauge attitudes towards taxation and to compare them across countries. The Human Beliefs and Values Survey is a potentially valuable tool in this respect, as it covers 200 thousand respondents in 81 countries accounting for 85 percent of the world's population. Of those respondents, 989 are from Vietnam. The survey questionnaire includes a range of statements and asks the respondents to comment whether they agree with them. One of such statements refers to "cheating on taxes if you have a chance". Answers can be in a scale of one to ten, where one means that it is "never justifiable" and ten "always justifiable." A vast majority of Vietnamese participants declared that tax evasion was unethical, with 87.8 percent considering that it was never justifiable. The mean score was 1.32. Admittedly, self-reporting high ethical standards is not surprising and may not be particularly informative. But it is interesting to note that the share considering evasion to be never justifiable was substantially higher in Vietnam than in other countries and territories in the region, including China, Hong Kong (China) and Thailand. Variation across population groups in Vietnam is interesting as well. The belief that tax evasion is never justifiable appears to be slightly stronger among young adults and the elderly, whereas there is more willingness to compromise among respondents aged between 30 and 49 years. Also, the category most receptive to tax evasion is the most highly educated group. Another suggestive breakdown is by religion. Ancestral worshippers are most likely to believe that tax evasion is never justifiable. Buddhists are somewhat more flexible on the ethics of tax evasion, whereas Christians are the most accommodating. Source: Robert McGee (2006). 56 TAXES AND FEES Figure 5.4: Negotiating with Tax Officials Source: VNCI and VCCI (2007). Figures indicate the percent of survey respondents who agree with the statement: "Negotiations with tax officials are an essential part of doing business". 57 6. BUDGET ALLOCATIONS Opportunities for people to make a living, to the equivalent of 50 percent or more of their have access to social services and to get support GDP in equalization grants. Such availability of in the event of bad shocks are better where there resources bears the potential to make a is an effective government. It takes more than difference, helping to lift people out of poverty just budget resources to deliver on and to reduce their vulnerability. The main infrastructure, health care or targeted programs. challenge is to use them effectively. The small But the availability of such resources can make size of the typical Vietnamese province raises a difference, especially in areas which are too important coordination problems in relation to poor to raise them on their own. Public public investments. And the pro-poor nature of expenditures in Vietnam are not particularly budget allocations from the center to the large in relation to GDP, but they have been on provinces is only now starting to be replicated a slightly upward trend, with spending for transfers from provinces to districts and increasing on investment and education, and to below. At the commune level, progress is still a lesser extent on health. The level of public needed to improve grassroots participation in expenditures also appears to be sustainable over the budget process. time, as the ratio of public debt to GDP is expected to peak barely above 51 percent Overall spending around 2016 and to start declining afterwards. While neither the functional composition nor Total government expenditures are not the relative level of public expenditures are particularly large in Vietnam. By international changing dramatically in Vietnam, the standards, where they accounted for 29.8 mechanisms used to allocate budget resources percent of GDP in 2006. This fraction had been to lower levels of government went through a gradually increasing since 1990, but has profound transformation since 2004, as part of a remained relatively stable in the last four years. move towards increased decentralization. By The government's declared intention is to have now, almost half of public expenditures are expenditures grow annually by 2 to 3 percent decided at provincial levels and below. less than GDP. However, it is unclear that this Importantly, transparent formulas have been will happen in practice, given the willingness to introduced to allocate budget resources to the upgrade the skills base of the country, the provinces. Initially they covered recurrent escalating costs of health care and the pressing expenditures, but they were recently extended to needs in terms of infrastructure development. capital expenditures as well. These formulas rely on indicators such as provincial population Some of these upwards trends have already been and the level of development; they also take into clear for quite some time in the budget of the account geographic conditions and the central government (Figure 6.1). Between 2000 organization of government administration. In and 2007, spending on infrastructure by the practice, budget allocation norms result in a central government has increased by about 2 massive transfer of resources from richer to percentage points of GDP, spending on poorer provinces, some of which may receive education by about 1 percent, and spending on 58 BUDGET ALLOCATIONS health by half a percentage point. If spending increase in the coming years, prudent on pensions and social transfers appears to be macroeconomic policies would ensure that it stable, or even declining, it is mainly because remains manageable. the social insurance agency takes care of a portion of it, with the funding coming from Under plausible assumptions, Vietnam should social security contributions and insurance remain at low risk of debt distress. The rate of premiums. Only targeted programs and the total public debt to GDP is projected to reach a pension entitlements accumulated by civil maximum of about 51 percent in 2016, but servants before 1995 are paid for by the budget. should decline subsequently provided that Moreover, pensions were not readjusted in borrowing continues to be cautious and 2007, a year of rapid growth and accelerating contingent liabilities are kept under control inflation. (Figure 6.2). Success crucially depends on completing the reform of the state sector. While government expenditures are not External debt is projected to decline from an particularly large, there have been concerns already manageable 31 percent of GDP in 2007 about their implications for the sustainability of to about 27 percent in 2012. These figures refer public debt, especially in light of Vietnam's to the nominal value of public debt, but most of ambitious growth and investment targets. Vietnam's borrowing is in concessional terms. Massive infrastructure development, larger If net present values are considered, instead of spending in the social sectors, and possible nominal values, total public debt could peak at support to large economic groups, could result 47.4 percent of GDP in 2016 and external debt in faster economic growth in the medium term, at 25 percent in 2007. Given the openness of but at the expense of larger government the Vietnamese economy, the service of liabilities in the longer term. While this risk external debt may not exceed 4 percent of cannot be ruled out, and public debt is bound to exports at any point in the coming years. On Figure 6.1: A Breakdown of Budget Expenditures Source: Based on MOF data. Figures for 2007 are forecasts. 59 SOCIAL PROTECTION Figure 6.2: Not Overburdening Future Generations Source: Debt sustainability assessment by IMF and the World Bank. the other hand, the fraction of government coherence between capital and recurrent revenue needed to service public debt could expenditures. To foster transparency, transfers increase from 10 percent at present to 15 are based on numerical formulas, involving a percent by 2017, to start gradually declining series of measurable indicators at the province afterwards. level. Initially, allocation norms were applied to recurrent expenditures only, based on 11 Spatial allocation formulas by policy area. Since 2007 the same principle has been adopted for capital The allocation of public expenditures across expenditures. levels of government changed dramatically The formulas use provincial population as one with the Budget Law of 2002, which came into of the main determinants of budget allocations, effect in 2004. Vietnam became a much more but they modulate them based on the level of decentralized economy as a result. By now, 45 development of the province, the difficulty of its percent of public spending decisions are made terrain, its administrative organization or the at the provincial level and below. The National presence of a regional development pole, among Assembly and People's Councils are in charge other criteria (Table 6.1). of budget appropriations. This has also entailed an important transformation in the role of line Other redistribution mechanisms for ministries, from doing to steering. For government expenditures are also at play, in instance, provincial governments now decide addition to these formulas. Among them are a on their investment plans using their available series of targeted programs such as the budgetary resources, instead of submitting exemption of education fees for poor "wish lists" of projects to MPI for screening households, the distribution of health and approval. insurance cards for disadvantage people or the support to communes facing extreme The move towards decentralization has been difficulties. Each of these programs has its accompanied by the implementation of an own formula mapping resources to locations. equalization mechanism, aimed at transferring In most cases there is a noticeable trend budgetary resources to provinces so as to meet towards greater simplicity and transparency. the government's development objectives. This For instance, in the case of health insurance mechanism also seeks to achieve greater cards (under Program 139), provinces receive 60 BUDGET ALLOCATIONS Table 6.1: Budget Allocation Norms for Provinces Capital expenditures Category Criterion Recurrent expenditures Equalization Targeted transfers transfers 1) Total + + + 2) School-age children + Population 3) Ethnic minorities + + 4) School-age children + in P135 communes 5) Poverty rate + + + 6) Local revenue + Development 7) Revenue transferred to state budget + + 8) Industrial output + + 9) Total surface + + Geography 10) Disadvantaged location + + + 11) Growth pole + + 12) Administrative units at district level + + Administration 13) Districts in disadvantaged + locations 14) Civil servants + Source: Based on Decision 151/2006/QD-TTg for the allocation of recurrent expenditures in 2007 and Decision 210/2006/QD/TTg for the allocation of capital expenditures in 2007-2010. The former does not include NTPs, Health Care Funds for the Poor (HCFPs), and Program 135. 70 thousand dong per beneficiary per year. similar to the sum of resources set aside by all And in the case of support to disadvantaged the communes in potentially affected areas. In localities (under Program 135) each targeted the event of a natural disaster, affected commune receives 800 million dong for communes can tap on that common fund. infrastructure and 200 million for production development. Pro-poor transfers Support to communes affected by natural Budget allocation norms, combined with the disasters has a design more akin to insurance. All communes located in areas considered authority for People's Councils to decide how to prone to typhoons, floods and other calamities appropriate the resources allocated to them, could have to set aside either 3 or 4 percent of their be a powerful tool for social inclusion. Unlike budget as a solidarity contribution. The targeted programs, these norms transfer to local applicable percentage is set based on weather levels of government not only the funding but statistics from the period 2001-2005. The also the responsibility for its use. Provided, this central budget, in turn, includes a relief fund is, that the norms actually redistribute resources for natural disasters whose size is roughly from richer to poorer provinces. 61 SOCIAL PROTECTION Figure 6.3: How Large are Equalization Grants? Source: Based on data from MOF and GSO. By referring to the provincial poverty rate, or to empirical question. But once the numbers have the number of districts in disadvantaged been worked out, the answer is clearly locations, the formulas underlying these norms affirmative (Figure 6.3). ought to do so. For instance, transfers for investment purposes increase by 50 thousand The size of the transfer is actually remarkable, dong per year for each ethnic minority person both in absolute and in relative terms. The living in the province. Similarly, the allocation poorest provinces in the Northwest or in the of recurrent health expenditures per capita is 72 Central Highlands receive equalization grants in percent higher in the uplands and in low areas excess of 2 million dong per person per year. populated by ethnic minorities than in urban Everything else being equal this amounts to 25 areas. But not all the criteria considered in to 30 billion dong more per year than for a rich budget allocation formulas are progressive. For province in the deltas. example, more resources are transferred to provinces which have greater capacity for While figures on sub-national accounts are not revenue collection or harbor regional totally reliable in Vietnam, equalization grants development poles. could exceed 50 percent of provincial GDP in some cases. Beyond those extreme cases, it is In the end, whether the mechanism set up by the clear that poorer provinces in the Northern government of Vietnam actually transfer Mountains, the Central Highlands and the resources from richer to poorer provinces is an Central Coast receive much larger equalization 62 BUDGET ALLOCATIONS grants than provinces in the richer Red River their transfers. According to the new investment Delta, Southeast and Mekong Delta. plan of the Government, in 2008 the net transfer to the Northern Mountains could increase by Not only do budget allocation norms favor only 0.8 percent, compared to 6.4 percent for poorer provinces: they have also made the the Central Coast, 11.1 percent for South East, allocation of government expenditures more 11.0 percent for the Central Highlands and 17.2 progressive over time. This is reflected in the percent for the Red River Delta. correlation between net transfers and poverty rates by province. Provincial poverty rates can Local decisions be measured using the household expenditure surveys produced by GSO. Net transfers are The decentralization of authority on budget defined as the difference between the resources appropriations to provincial governments channeled by the central government to the provides an opportunity to better serve local provinces and the tax revenue channeled by populations. But it also raises new challenges. provinces to the central government. Net One of them is the possible lack of coordination transfers are highest for the poorest provinces, between provincial decisions. including Lai Chau, Bac Kan and Ha Giang. Richer provinces like Binh Duong and Dong The median population of a Vietnamese Nai, on the other hand, are net contributors to province is less than one million people. This the central budget. may result in over-investment in some cases and under-investment in others. The The correlation between net transfers and willingness of most coastal provinces to have poverty rates has become both stronger and their own deep-sea port, or their own steeper over time. A comparison between 2002 university, could undermine the necessary and 2006 is telling in this respect (Figure 6.4). economies of scale. In other cases, none of the In 2002, resources were channeled to the individual provinces may be large enough or provinces on the basis of allocations by line rich enough to undertake critical projects, such ministries, investment projects approved by as an international airport. Wasting resources MPI and targeted programs. In 2006 they still through duplication, or failing to undertake relied on targeted programs, but were mainly critical investments, could be detrimental to based on quantitative formulas for recurrent and economic growth and job creation, thus capital expenditures. The correlation has making local populations more vulnerable. become stronger in the sense that it explains a Regional planning, bringing coherence to the larger share of the variation in net transfers. It investment decisions of groups of related has become steeper in that an increase in the provinces, is one way to address this poverty rate is associated with a larger net coordination failure (Box 6.1). transfer. In 2002, one extra point of the poverty rate was associated with an additional transfer Insufficient budgetary resources push many of roughly 20 thousand dong per person per poor localities to rely on local fees, charges or year. By 2006, the corresponding net transfer even voluntary contributions to raise revenues. had almost doubled. Even when inflation is According to a recent study, the majority of taken into account, this almost doubling reflects communes receive transfers accounting for 50 a very substantial increase in the real amount of to 70 percent of their budget needs. This study resources transferred. However, this increase listed about 28 different types of fees and may not be sustainable and should rather be contributions the farmers have to pay to local seen as a one-off change. The formulas for the government and cooperatives to make up for the allocation of budget resources are such that shortfall. On average, these fees and provinces with a stronger capacity to raise contributions would cost a poor household revenue will be favored, whereas poorer nearly 10 percent of its total expenditure. In provinces will experience a slower growth in some extreme cases, they represent 18 percent 63 SOCIAL PROTECTION Figure 6.4: Net Transfers and Poverty Rates Source: Based on data from MOF, MPI and GSO. Hanoi, HCMC and Vung Tau are excluded from the figures although not from the correlation analysis. 64 BUDGET ALLOCATIONS Box 6.1: Regional Plans, Infrastructure Master Plans and Public Projects With much greater emphasis being placed on sustainability, regional development planning has also received increased attention in the last few years. The Socio-Economic Development Plan (SEDP) 2006-2010 explicitly sets out the requirements for the reform of regional planning, by instructing government to "pay attention to spatial planning, technical infrastructure... to prevent overlapping efforts and conflicts between sectors and territories". Shortly after the approval of the SEDP, the government has promulgated Decrees 92, 99 and 140. These documents for the first time lay out the regulatory framework for regional planning. According to Decree 92, master plans should have their targets built around three pillars: economic, social and environmental. The decree also requires the evaluation of the strategic environmental impacts of master plans, although it provides no guidance as to how this should be done. Decree 99 specifies the process and the responsibility of relevant authorities in the implementation and monitoring of development plans and master plans. Decree 140 lists out the requirements of environmental protection in the formulation, appraisal and implementation of development strategies, master plans, socio-economic development plans, development programs and projects. The importance attached to improving master plans was highlighted again in the report of the Prime Minister before the National Assembly, in October 2007. Improving the quality of the development of strategies and master plans was included as one of the main tasks for 2008. Through Official Letter 950, issued in July 2007, the Prime Minister instructed MPI to take the lead in the formulation of the Socio-Economic Development Strategy to 2020 and the regional master plans for the period 2010-2020. Decree 92 is now being revised to increase transparency, to remove duplications and to increase consultation along the way. of the income of the "average" household in the sum by district. But some more refined commune. schemes are emerging as well. Son La, for instance, foresees the share of investment Other problems arise at lower levels of resources allocated by districts growing from 15 government. Much progress has been made in percent in 2007 to 30 percent in 2010, with the adopting predictable and transparent formulas share allocated by the provincial government to allocate resources from central to provincial declining accordingly. Guiding principles such governments. But until recently allocations as transparency, equity and compliance with from the province level down remain dominated master plans are common. Priority is given to by the old budget mechanisms. the repayment of debt over the approval of new projects. When district-level indicators are used Only gradually are provincial governments to determine the amount of resources to be starting to emulate the national formulas for their transferred, the criteria are very similar to those budget transfers to districts and communes. As chosen by the central government. They of August 2007, 21 one of them (including involve the population of the district (including Hanoi) had approved budget allocation norms ethnic minorities), its development level, its like those used by the central government. geography and the size of its civil service. Encouragingly, more than 30 others have Economic zones, industrial zones and world submitted proposals in this direction to their heritage areas also get special treatment. respective People's Councils. Further down, the decree on the exercise of The budget allocation norms being developed democracy in communes, issued in 2003, by the provinces are somewhat simpler. enshrined the right of citizens to supervise and Investment expenditures tend to be set as a lump to be informed about "the annual estimates and 65 SOCIAL PROTECTION settlements of the commune budget". The had heard about the commune budget but, decree also recognized the right to be informed disturbingly, all of them reported that they did about "the estimates and settlement of revenues not understand it. Bearing in mind the and expenditures of funds, programs, projects complexity of budget issues, the findings and contributions mobilized from the people for indicate that there is still a long way to go in the construction of infrastructure and public implementing grassroots democracy. welfare projects of communes, villages and hamlets, as well as their implementation results" In April 2007 an ordinance was issued replacing and to supervise "the collection and spending of the decree on the exercise of grassroots various funds and fees according to the State democracy. The Ordinance provides citizens regulations as well as various contributions of with the same rights as the Decree with regard the people". to the transparency and supervision of local budgets, but it contains new elements that could Progress in the implementation of these increase transparency and improve provisions has been limited, however. A study implementation. Unlike the decree, the based on surveys and focus group discussions in ordinance indicates when the information must 320 villages across four provinces in Vietnam be posted and the duration of the posting, it found that two thirds of participants either rationalizes the means through which disagreed or strongly disagreed with the information must be disseminated, and statement that "there is plenty of timely introduces the right to submit complaints and information on the commune budget". Another denunciations in relation to the supervised study found that 40 percent of the interviewees activities. 66 7. TARGETED PROGRAMS Vietnam has a range of targeted programs schooling. The impact is larger in poorer whose goal, directly or indirectly, is to communes and in those located in more alleviate poverty. Some of them provide cash difficult or remote areas. It is smaller in transfers to vulnerable groups such as communes where ethnic minorities represent orphans, the very elderly or war invalids. a larger share of the population. There are Others support poor households through good reasons to believe that the subsidization access to credit or through subsidies to of participating in mainstream programs by participate in mainstream social programs. poor households is having a positive impact Yet others focus on disadvantaged as well. But there is less clarity in relation to communes, financing local infrastructure the subsidization of micro-credit. In some development and initiatives to improve cases, it is simply too early to tell. The livelihoods. And there are also programs targeted program for rural sanitation and the focused on improving access to services and initiatives to increase the landholdings of boosting the assets of ethnic minorities. ethnic minorities fall in this category. These programs differ in several important However, there are signs suggesting that their ways. There is a wide variation in the implementation will be challenging. resources available, both across programs and for the same program across provinces, A range of programs depending on the local funding they can mobilize. The result is a considerable A number of social assistance and anti- dispersion in the support received by various poverty initiatives exist in Vietnam. Some of beneficiaries, from modest in the case of cash them provide cash support to vulnerable transfers to sizeable in the case of local groups, others deliver assistance on a targeted infrastructure development. There is also basis to disadvantaged groups or regions. The considerable diversity in targeting methods. pricing of utilities such as water can also be In some cases, beneficiaries are a small based on the ability to pay of different fraction of the population at large, and population groups. presumably of the intended target group as well. In others, the coverage of the target Cash transfers are mainly handed over by group is quite substantial, and the extent of Social Guarantee Funds, which provide leakage to other groups is not excessively regular relief to defined vulnerable groups large. Assessing the impact of all these and to war veterans and invalids. Subsidies programs on household living standards is are provided through the National Target not an easy task from a methodological Program for Poverty Reduction (NTPPR), perspective. A rigorous evaluation is formerly the Hunger Eradication and Poverty available in the case of local infrastructure Reduction program (HEPR). Development development. The results suggest that this assistance to poor communes is managed intervention has a sizeable impact on local through Program 135, whereas Program 132 access to markets, household expenditures and Program 134 seek to support ethnic and even social indicators such as children's minorities with agricultural and residential 67 SOCIAL PROTECTION Box 7.1: A Variety of Targeting Approaches The NTPPR 2006-2010, much the same as its predecessor the HEPR 2001-2005, targets households who have been listed as poor using procedures set out by MOLISA. Poor households are formally identified on the basis of an official poverty line, defined as a certain level of income per person per month. In practice, whether household income is above or below such threshold is decided through a combination of surveys and community discussions. This procedure is currently under revision, based on the results of comprehensive field work conducted in 2005. The new approach first identifies those households who are "certainly poor" and "certainly non-poor" and applies tests to the remaining households to establish whether or not they are likely to be poor. MOLISA is currently testing the results of these exercises against household expenditure data from the VHLSS. At aggregate levels, the new approach yields poverty rates which are similar to those produced by the GSO using data from the VHLSS. Program 135, which provides support to communes facing hardship in ethnic minority and mountainous areas, adopts a geographic approach to targeting. The first phase of the program allocated 700 million VND per year to selected communes. In general these allocations were made in a pro-poor manner. Poverty maps constructed using small area estimation techniques show that the communes targeted by Program 135 were generally those with the higher poverty rates. Targeting issues arise at lower levels, as resource allocation may favor wealthier villages within a commune and not everyone in those villages is poor anyway. The second phase of program 135 is tackling these issues by requiring the inclusion of both poorer villages within the poorest "zone III" communes and the poorest villages in the less poor "zone II" communes. The new guidelines governing resource allocation specify the need to develop clear, transparent criteria for channeling resources to the poorest villages and communes and allow flexibility in allocating more funds to poorer places. Programs 132 and 134 have targeted vulnerable ethnic minorities using different criteria. Program 132 aims at distributing agricultural and residential land to ethnic minority households in the Central Highlands, so as to bring landholdings up to a specified minimum. IPSARD is currently assessing the impact of this program in two provinces. Program 134 operates on the same logic, but it targets all ethnic minorities (not just those in the Central Highlands) and its specified minimum for landholdings is lower. In addition to providing land, this program also identifies ethnic minorities living in temporary housing as eligible to receive 5 million dong for housing construction. Ethnic minority households living in high mountains or in areas lacking clean water are also eligible for support to improve their water sources. Villages without clean water supply get a varying level of assistance depending on the proportion of ethnic minorities in the village population. land, housing and water supply. All these Although funding has increased, reaching initiatives differ not only in the nature of the 750 billion dong in 2007, it remains an benefits they provide, but also in the targeting important constraint. The resources techniques they use to identify their supporting these cash transfers come from a beneficiaries (Box 7.1). combination of national and local sources, The overall number of beneficiaries from with their amount varying widely across the targeted programs is not easy to estimate. In country. As a result, some in the target the case of the Social Guarantee Fund, it has groups cannot be covered and, when covered, increased more than three-fold over the first allowances tend to be small on average and to part of the decade (Table 7.1). The number of differ considerably across recipients. beneficiaries has risen in all categories. A recent review of the implementation of the Yet, regular relief in the form of monthly Ordinance on Disabled Persons interviewed cash transfers to vulnerable groups still intended beneficiaries in two provinces. In covers only a small fraction (about 0.5 Tuyen Quang, about 40 percent of the percent) of the Vietnamese population. disabled people interviewed were receiving 68 TARGETED PROGRAMS Table 7.1: A Growing Number of Beneficiaries 2001 2002 2003 2004 2005 2006 Isolated elderly and poor disabled 68 73 77 103 110 122 Orphaned children 24 39 32 45 47 61 Disabled 90 111 120 155 179 180 Elderly above 85 26 70 88 People with HIV/AIDS 10 10 Other 9 Total number of beneficiaries 182 223 229 329 416 470 Source: Based on data from MOLISA. Figures are in thousands. cash allowances of an average amount of 79 of officials working on poverty alleviation thousand dong per month. In Nghe An, on the programs, and on monitoring. other hand, only 17 percent of the interviewed were receiving allowances. But at 221 A total of 43.5 trillion dong has been allocated thousand dong per month the average cash to this program over a five-year period, with 60 support was much higher. percent of the resources going to the provision of preferential credit. For the remainder, the One relevant question, then, is whether cash central budget covers 29 percent of the transfers should be scaled up under the form envelope, with local governments and local of a more coherent program, both in terms of communities expected to contribute a further 5 beneficiaries and in their allocation of and 6 percent respectively. resources to local levels. Most of the resources flow through existing Poor households policies managed by a number of line ministries. For example, MOH is responsible for providing free health insurance cards to poor Assistance to poor households under the households and honoring the obligations these NTPPR and its predecessor HEPR has been cards bring. Likewise, the Ministry of running since 1998. The change in the name of Education and Training (MOET) is responsible the program reflects the conclusion by for ensuring that exemptions and reductions of MOLISA that hunger has been already education fees are implemented according to eradicated in Vietnam. In its new phase, the policy. program seeks to halve poverty between 2005 and 2010. Other programs provide a subsidy allowing the poor to participate in the mainstream services Activities under the NTPPR fall into three areas. accessed by the non-poor. In a context where Two policies and four projects aim to create poverty continues to decline, the assumption is conditions for the poor to develop their that subsidies can be reduced over time and productive activities. This area includes the those who move out of poverty can fund their largest component of the NTPPR, the provision own participation in education or health of preferential credit through VBSP. The insurance. However, ensuring the coherence second area comprises four policies covering between mainstream social services and access to basic services, including healthcare, targeted subsidies can be challenging (Box 7.2) education, housing, water supply and legal aid. The third area focuses on building the capacity Household expenditure surveys can be used to 69 SOCIAL PROTECTION Box 7.2: Linking Targeting Programs and Mainstream Social Insurance Decision 139 (passed in 2002) led to the establishment of HCFPs, with 75 percent of the funding to be provided by the central government and the rest by local and international sources. Initially, resources could be used to either purchase health insurance cards for beneficiaries or to directly reimburse providers. From 2006 onward, HCFPs can only be used to purchase health insurance cards. The government is also to increase the resources per intended beneficiary and to introduce a partial subsidy for the heath insurance premium of the near-poor. Although the implementation of HCFPs was slow to take off in many provinces, the number of beneficiaries has increased substantially in the last few years. Several problems remain, however. Funding is one of them. The government support for the HCFPs was increased from 50 thousand dong initially to 70 thousand in 2006. But this is still too low compared to the cost of the benefit package for the poor. And only a few provinces have mobilized significant levels of funding for their HCFPs. Another problem is the limited extent of risk pooling. All the beneficiaries from this program fall under the same fund in VSS, regardless of their province. But such fund is still managed separately, offering no scope to diversify risk with other population groups with different patterns of morbidity. Pooling would make it possible to increase the average healthcare spending on the poor which is extremely low compared to that of other participants in the health insurance program. Concerns have also been expressed about equity, due to the limited capacity to identify the beneficiaries at local levels and to use this information in budget planning. Experience with a health support project in the Mekong River Delta shows that better-off provinces obtain more resources for their HCFPs by applying looser targeting, hence reporting a higher number of beneficiaries. assess the coverage of some of the NTPPR Poor communes programs and projects. Data from the 2006 VHLSS show that that 13.7 percent of all Some targeted government expenditures do not households had been classified as poor by have poor or vulnerable households as their local authorities (Table 7.2). Though this was intended beneficiaries, but rather poor not much lower than the estimated number of communes. While the emphasis of public poor households using consumption-based investment programs during the 1990s was on approaches to measurement, there were the countries' growth poles, in recent years important shortfalls in some places. For considerable resources have been spent in example, according to the 2006 VHLSS, 49 linking poor regions to nationwide transport and percent of the population of the Northwest electricity networks. Accurate estimates of region is below the poverty line. However, expenditure are complicated by the blend of only 22 percent of people had been officially central and local funds. But over the past decade classified as poor. On a more positive note, the government has probably spent about 1.2 the fraction of people officially classified as billion dollars to connect 6.3 million households poor was higher among those with to the electricity grid. A further 2.6 billion was expenditures below the poverty line (46 channeled towards the construction of roads in percent), and considerably lower among those rural areas over a five-year period. By now, less above (7 percent). In the parlance of than 300 communes are without road access. targeting, this reflects a reasonably good There is evidence that these investments can "coverage" of the program, and a modest have a significant impact on the welfare of the amount of "leaking". By way of comparison rural population (Box 7.3). in 2002 the corresponding fractions were 28 and 5 percent. In addition to directing investment resources 70 TARGETED PROGRAMS Table 7.2: Coverage and Leakage of Targeted Programs Household type Percent who: Total Food poor Poor Non poor Are officially classified as poor 13.7 53.6 46.6 7.5 Hold a healthcare card 10.5 41.6 35.1 6.2 Covered by health insurance 39.9 31.4 31.7 41.3 Have access to preferential credit 5.8 21.4 18.7 3.4 Are exempt from education costs 8.3 38.8 31.6 3.9 Source: Based on preliminary 2006 data from GSO. A household is classified as poor if its total expenditures are below the cost of a food and non-food basket securing 2,100 calories per person per day. It is classified as food poor if its total expenditures are below the cost of the food component of such basket. towards the provision of mainstream owners by now. This is almost twice the level infrastructure to rural areas, the government has achieved under the first phase. Tuyen Quang also developed mechanisms to channel province has already attained the goal of significant funds to some of the country's assigning all beneficiary communes to be poorest localities. In its first phase, Program investment owners, demonstrating to other 135 allocated around 430 million dollars to provinces that this is feasible. 2,362 disadvantaged communes over a five-year period for local infrastructure investment and Among other achievements to date, a number of livelihood support. provinces have developed needs-based criteria for resource allocation for 2008, despite the The program is now in a second phase, with absence to date of specific central-level about one billion dollars scheduled for guidance on this issue. Financial reports are disbursement before 2010. While it remains being prepared that track resources down to the heavily focused on the provision of basic level of spending agencies. The individual infrastructure, it now foresees significantly accounts of these agencies are being reconciled increased financing for capacity building. This with local treasury accounts in many provinces includes support for agricultural production and implementation audits have been conducted aimed at farmer groups, support for in eight of them. On other hand, further progress participatory planning and investment is needed on issues such as public disclosure of management at the commune level, and financial and audit reports, guidance on communications outreach to promote public activities to support production for markets and access to information about the program. The appropriate procurement methods for small- program has also been emulated by a similar scale infrastructure. initiative targeted to poor coastal communes. Water and land The second phase of Program 135 is still in its early stages, so that relatively few hard data are Other government interventions aim at available on actual implementation. However, a providing access to services to the most range of good practices have been noted in disadvantaged populations, or at boosting their particular provinces. For example, on the issue assets. In terms of basic services, Vietnam is at of decentralization of investment ownership to risk of not attaining the Millennium local levels, more than 28 percent of all Development Goal (MDG) on water and communes under the program are investment sanitation. As of 2004, only 48 percent of rural 71 SOCIAL PROTECTION Box 7.3: The Impact of Rural Roads Assessing the impact of local infrastructure investments on the livelihoods of local populations is not easy from the methodological point of view. Infrastructure development often coincides with rapid economic growth, which in turn increases the availability of goods and improves access to markets. How much of the change observed in living standards at local levels can be attributed to infrastructure development alone becomes unclear. Also, the same intervention could have different impacts on different places, depending on their characteristics. For example, communes with better educated households could derive larger impacts from improved access to markets. The extent of local corruption matters as well, as it may affect the way funds are used. Another way in which the workings of government can dull impacts is fungibility. Aid or central government spending for road projects may substitute for local government spending intended for that purpose. A study of a road project implemented in Vietnam between 1997 and 2001 addressed these methodological problems in a convincing way. The project aimed at rehabilitating rural roads which were in very bad condition, some with missing bridges and impassable sections year round. The study constructed a baseline before the project started, in 1997. It conducted subsequent surveys in 1999, 2001 and 2003, tracing the implementation process and schedules of prior rounds. This approach provides sufficient controls for comparison and follows the beneficiary communes for long enough to capture full impacts. The findings reveal significant impacts of rural roads on the development of markets and commercialization. Some outcomes, such as food goods availability, responded rapidly to new and improved roads. Others, such as the presence and frequency of markets and the availability of non-food goods and services, took two years more on average to emerge. The project resulted in households switching from agriculture to non-agricultural activities, mostly service-based. Perhaps most notable, the project had significant and sustained impacts on primary school completion rates. On the other hand, there are many indicators of local area development which cannot be easily attributed to better roads. The study also found enormous impact heterogeneity, depending on local circumstances. Other things equal, poorer communes tend to benefit more. But poorer communes are generally not equal, and some of their characteristics tend to lessen impacts. For example, location in mountainous areas, holding other commune characteristics constant, boosts road impacts. But many mountainous communes in Vietnam have a high share of ethnic minorities and high illiteracy rates which are both negative in their effects. The role of the initial level of local market development appears to be crucial. The complexity of the way in which initial conditions interact with road improvements to affect outcomes suggests that simple formulas to allocate resources for local investment may be elusive. Source: Ren Mu and Dominique van de Walle (2007). households had access to clean water sources on Rural Water Supply and Sanitation. But in and a mere 16 percent had hygienic sanitation. spite of rapid progress, meeting such ambitious It is clearly the ambition of the government to goals will be challenging. redress this situation. The National Rural Clean Water Supply and Sanitation Strategy sets the Alongside the revised Program 135, the target for access to clean water (based on government has also embarked on an attempt to national standards) at 85 percent by 2010 and strengthen the economic base of ethnic minority 100 percent in 2020. The strategy also households. The focus is on productive and anticipates that 70 percent of the rural residential land. Introduced in 2002, the population will be using hygienic latrines by original program (known as Decision 132) 2010, with full coverage achieved by 2020. The aimed to address the pressing land problems of main government instrument to implement this ethnic minorities in the Central Highlands. The strategy is the National Target Program (NTP) more recent Decision 134 extends land 72 TARGETED PROGRAMS allocation activities to ethnic minorities across resources have been inadequate to obtain the the country and expands the scope of support to necessary amount of land from other include the provision of clean water and households. housing. Local authorities are required to identify households that have landholdings Even if land can be allocated satisfactorily, below a certain specified minimum, which research suggests that there is no easy, varies for different qualities of land. They are automatic response in terms of increasing welfare. One study follows the process of re- then expected to make up the identified shortfall allocating forestry land away from SFEs to by allocating land from state-owned farms, by households and user groups in two villages in reclaiming previously unused land, or by the Central Highlands. The research describes redistributing land from households with bigger land allocation taking place in a context of endowments. disputes over rights to use forest land, with prior On first analysis, these actions seem well or customary claims sometimes over-riding the conceived. Ethnic minorities often lack the newer, formal claims. Though the handover of productive land assets that have supported the land to local people brought some welfare gains, majority group to move out of poverty. From they were highly variable across households. this perspective, the allocation of additional The degree to which households were able to land would provide an obvious solution. convert their new rights into higher incomes depended on wealth, the size of upland fields A survey-based review of the implementation of outside the allocated forest, labour capacity and these decisions is currently being undertaken. the location of the village. Though field work is not yet complete, findings so far suggest that the impact on ethnic minority Different issues may influence the livelihoods may be less advantageous than implementation of the housing component of originally foreseen. Decision 134. The case of a commune in Gia Lai province is telling in this respect. The One obvious difficulty lies in the fact the most commune had been allocated funds to build productive and most easily cultivable land is houses for 121 households with low-quality already allocated and farmed. Though it may be dwellings. Each house cost six million dong possible for authorities to find extra land, in and was built by a contractor hired by the practice it may be distant or poor quality or commune. The houses were built of brick, in both. Further complications arise because the the Kinh style, next to the traditional ethnic process of reclaiming land from state forestry minority houses. But they were not considered enterprises (SFEs) has been more cumbersome to be appropriate substitutes by the intended than expected. Provinces also argue that beneficiaries, who chose not to move. 73 8. HEALTH CARE If health care systems are to be judged by their government health units through their eligibility outcomes, Vietnam deserves high marks. Key for reimbursement by the social insurance indicators, especially in relation to child system, and fostering the use of generic drugs mortality, morbidity and nutrition, are indeed and the development of pharmaceutical comparable to those of countries at a much research, are steps worth considering. higher development level. And progress on these indicators has been sustained over time. Good but unequal However, on a closer examination the picture is more complex. Average indicators have Health systems can be judged by their ability to improved, but the gap between rich and poor improve the health of the population. They may has widened. And the grassroots health care not the only thing that matters for health network that served Vietnam so well in difficult outcomes, and it is difficult to quantify the times is increasingly under strain, unable to individual contribution they make. But it is cope with the growing importance of non- unlikely that a country could exhibit good health communicable diseases and the demand for outcomes if its health system was in disarray. more advanced services by an increasingly wealthy population. Meanwhile, the scale and From this perspective, Vietnam is doing well; heterogeneity of the hospital network, combined as well as countries at a considerably higher with the emergence of large numbers of private development level. One key indicator for health units, makes the regulation of the health which internationally comparable data are system increasingly challenging. The available is child mortality. China, Indonesia, government's overall strategy combines the the Philippines and Thailand did better than reliance on market mechanisms (the so-called Vietnam during the 1980s. However, socialization policy) with the subsidization of Vietnam's performance, in absolute terms and the participation of the poor in mainstream relative to these other countries, picked up in programs (the socialist orientation). Increased the early 1990s, and has improved further since reliance on market mechanisms has aligned the then. Between 2000 and 2005, the infant incentives of service providers with the mortality rate declined from 36.7 per thousand demands of those who can afford user fees, but to 17.8, and the mortality rate for children it has led to a relative neglect of social mandates under 5 years from 42.0 to 27.5 (Figure 8.1). and a ballooning of health costs. The As a result, life expectancy at birth climbed subsidization of the poor may not have resulted from 67.8 years to 71.3. in lower out-of-pocket expenditures. For the government strategy to be successful, it needs to Other indicators show Vietnam under an pay attention to market imperfections requiring equally positive light. Maternal mortality rate, specific interventions, and not just increased which stood at 85 per 100,000 live births in competition. Monitoring the delivery of 2002, had declined to 80 by 2005. The fraction services by autonomous units (especially of of children with low weight at birth fell from social mandates), certifying private and non- 7.3 percent in 2000 to 5.1 percent in 2005. And 74 HEALTH CARE Figure 8.1: Child Mortality in International Perspective Source: Based on data from World Bank (2007c). malnutrition among children under five years Most worrying of all is the growing gap in child of age dropped from 33.8 percent in 2001 to survival prospects between the poor and the 25.2 percent in 2005. better-off. Between 1997 and 2002, the largest reductions in both infant and under-five Admittedly, this was a period when household mortality rates occurred among the middle and income increased substantially. But even after near-poor quintile of the population, with the taking into account its strong economic growth, poorest quintile seeing the smallest gains. Vietnam still remains a good performer in terms Inequalities in child vaccination grew over this of infant and under-five mortality. During this period too, with the poorest fifth of the period there was also an increase in the population barely seeing any improvement, proportion of people with access to safe water. whereas the richest quintile experienced a 55 However, this may not be the full explanation percent increase of its immunization rate. for the progress in health outcomes either, as the fraction of households with hygienic latrines A sector in transition and access to safe water in rural, mountainous, remote and isolated areas is still low. And the Historically, grassroots care was the backbone quality of water remains an issue elsewhere. of Vietnam's health system. During the war, Performance is less remarkable when Commune Health Stations (CHSs) and Village considering the distribution of gains in health Health Workers (VHWs) played a key role in outcomes across regions, income groups or reducing communicable diseases and bringing medical conditions. Infant mortality has fallen under control dangerous ones such as small faster in richer southern regions and the Red pox, polio and leprosy. Setting up and River Delta. It remains stubbornly high in the maintaining such a network of local level Northern Mountains and the Central Highlands personnel during difficult times was a (Figure 8.2). There are also concerns in relation remarkable accomplishment. In remote to specific diseases. In the area of pneumonia localities, the assistance of the military health treatment the quality of care appears to have sector was crucial. deteriorated during the 1990s, and there is a re- emergence of tuberculosis. The quality of care provided at the grassroots 75 SOCIAL PROTECTION Figure 8.2: Infant Mortality across Regions in Vietnam Source: Based on data from MOH. level is high by international standards. A study network of health care providers, ranging from by the World Health Organization (WHO) large public hospitals to private practitioners. found that a high fraction of children with The scale of the system, the heterogeneity of diarrhea were correctly assessed, and a high health care units and the overall decentralization fraction of these were correctly rehydtrated of government decisions to local levels add to (Figure 8.3). A large share of caregivers with the complexity of the situation faced by health such children was also adequately advised and authorities. the children were properly managed in a high proportion of cases. A full 100 percent of Vietnam's hospitals are numerous and diverse. children with dysentery were treated correctly. In 2007, there were 1,030 of them run by The picture was less impressive in relation to different levels of government, and an children presenting with pneumonia. additional 49 owned and operated by the private sector. That makes for a total of 16.3 beds for However, grassroots care is under strain at every 10,000 persons. General and specialized present (Box 8.1). And it is not well placed to hospitals under the direct management of MOH, cope with the challenges posed by the growing all located in Vietnam's main cities, are the importance of non-communicable diseases. The largest and most technical up-to-date. At the increase in household income resulting from province level, 324 hospitals handle medical rapid economic growth also results in a demand conditions requiring special treatment. Ranging for more advanced services than CHSs can from 300 to 500 beds, they are managed by deliver. provincial health departments. The remaining 600 hospitals, at the district level, have an Increasingly, the focus of health policy has average of 80 beds. These hospitals remain shifted from running an effective health system highly dependent on the government budget for at the grassroots level to managing a complex their operation. 76 HEALTH CARE Figure 8.3: Quality of Care to Children with Diarrhea Source: WHO (1998). Meanwhile, there has been an explosion in research institutions and professional medical private health care. In 2004, there were 65,000 associations. Their activities range from the private health units licensed throughout the implementation of preventive and curative care country, including 30,000 private health to health promotion and advocacy. Some have clinics, 23,000 pharmaceutical units and focused on specific health issues such as 12,000 traditional health units. These figures HIV/AIDS, reproductive health, sanitation and under-estimate the importance of private the quality of care. providers, as there are large number of unlicensed facilities. Private health units are Overall, there are around 200,000 health staff of mainly established in big cities, and especially all categories in Vietnam, including medical doctors, pharmacists, nurses, nurse aids, in Hanoi and HCMC. But home visits by midwives and technicians. The number of private providers without a clinic or by highly qualified personnel has increased over traditional healers are quite common, time, and includes by now 982 staff with especially in rural areas. masters degrees, 337 with PhDs and 188 The private sector plays an important role in the professors and associated professors. About 70 delivery of out-patient services, where it percent of the doctors working in the private handles 60 percent of visits, focusing on sector are also employed in state-owned interventions allowing easy cost recovery. On facilities. the other hand, it only accounts for 4 percent of in-patient services and 11 percent of preventive The reform strategy care. The average Vietnamese person has 1.8 contacts with private health units per year, 1.2 The Doi Moi process entailed a profound contacts with CHSs, and 0.8 with public transformation of the health care system of hospitals. Vietnam. The last decade and a half have been characterized by continuous experimentation in Non-government actors are gradually emerging. search of new approaches. But beyond the They include international and local NGOs, specifics these attempts have been guided by 77 SOCIAL PROTECTION Box 8.1: Health Care at the Grassroots Level CHSs are the basic technical unit in the public health system. Their main task is to provide primary health care to local populations, including preventive care, family planning, normal delivery, the supply of basic drugs, and overall health promotion. In principle, the CHS is the primary health facility for most of the poor benefiting from health insurance, before they can be referred to higher levels. CHSs also manage the operation of the local network of VHWs. The average CHS employs 3 to 5 health workers, including a medical doctor in roughly 70 percent of cases, an assistant doctor, a midwife and a nurse. Nationwide, around 50,000 staff are employed at the CHS level. They receive a salary and get social insurance coverage. Around 80 percent of villages in Vietnam have a VHW. But not all of them are qualified, especially in remote, mountainous and border areas. More than a quarter of VHWs did not receive the required, three-month training. And most of them perform other jobs assigned by the village or commune, so that they cannot devote much time to health care activities. VHWs receive a small allowance of roughly 35,000 VND per month. The grassroots level of the health care system is fundamental for the implementation of National Target Health Programs (NTHP). There are 10 such programs at present, dealing with malaria, goiter, tuberculosis, leprosy, dengue fever, immunization, child malnutrition, mental health, food safety and HIV/AIDS. Several NTHPs have gained positive results, especially in relation to extended immunization, malaria and leprosy. These programs have also helped to reduce the morbidity and mortality of some dangerous diseases and epidemics, and contributed to reductions in malnutrition. About three quarters of the operating budget of CHSs is paid for by the government. The norm for recurrent costs is set at 10 million dong per CHS per year. Funding is provided by the central government in the case of disadvantaged communes and by local governments elsewhere. However, more than half of the country's CHSs have expenditures below the regulated norm, and about 72 percent of them have to collect service fees from the population. On average, revenue from fees amounts to 3.7 million dong per CHS per year. Spending on drugs and medical supplies barely reaches 10,000 dong per person treated per year. Given the under-funding of CHSs, spending by the health insurance program on the poor remains very modest. Since 2006, CHSs no longer report to district hospitals, which only kept responsibility for medical supervision. Now, preventive health bureaus are in charge of the implementation of NTPHs, whereas the health office of the District's People's Committee handles organizational matters. This overlap of responsibilities has substantially weakened the coordination of core CHS activities. Besides, CHSs are not recognized as institutional entities. Contracts with them can only be made through their supervising institutions and proper accounting is not performed. two main principles. First is to rely on user fees during the 1990s. The additional income as a mechanism to reduce the dependence of derived from these fees was especially health care units on budget transfers and to important to raise the earnings of health increase efficiency in the allocation of personnel and to finance investments in health resources. And second is to help the poor cope facilities. The scope for health care providers to with out-of-pocket health expenditures. The manage themselves was considerably enhanced first of these principles is often referred to as the through Decrees 10 (in 2002) and 43 (in 2006). "socialization policy". The second one reflects Hospitals could then be operated like SOEs, the socialist orientation of the government of with their own bank accounts and tax Vietnam. exemptions, with responsibility over personnel decisions, including hiring and firing, and with User fees were introduced in 1989 and became the authority to mobilize or borrow resources. a crucial source of income for the health sector Further deepening the move in this direction, 78 HEALTH CARE HCMC is now preparing the equitization of its "People's Health Protection, Care and hospitals. Other provinces can be expected to Promotion in a New Context". Citing concerns follow over time. about developments in the sector, and attributing the problems to a disconnect As user fees were rapidly raising out-of-pocket between demand and supply, this document health expenditures, province-level pilots tested proposed to rapidly increase the share of fee waivers and cards identifying patients eligible government resources devoted to the health for free or discounted services. Some of those sector, and to channel a larger fraction of those mechanisms rode on social health insurance, a resources through health insurance. It even set program introduced in 1992 that did not have universal health insurance coverage as the target many takers and was still stagnant one decade for 2010 (the reference to a specific date was later. Fee waivers did not work well either, as subsequently dropped). It also advanced the hospitals were often unwilling to accept them; need to upgrade health care facilities, giving delays, lack of funds, and information gaps priority to the lower tiers of the system, spelled trouble for approaches relying on the particularly in the Central Highlands, the issuance of cards. Drawing lessons from these Northern Mountains, the Central Region and the failures, in 2003 the government passed Decision Mekong River Delta. 139, which created province-level HCFPs. More recently, it was decided that HCFP would directly Who pays what? pay the health insurance premium of the poor and targeted groups, whereas a partial subsidy would be offered to the near poor. The emergence of a market economy, the epidemiological transition, the increase in In 2005, the politburo of the Party took the household income and the overall direction of unusual step of consolidating the overall reforms in the sector have resulted in a approach to health sector reform into a single fundamental transformation of health finance in strategic document, the Resolution 46 on Vietnam. Figure 8.4: Health Expenditures in International Perspective Source: Based on data from World Bank (2007). 79 SOCIAL PROTECTION Total health spending is in line with what could Hospitals in Vietnam receive their income from be expected, given the country's development three main sources: the budget, fee income level (Figure 8.4). Total health spending is also received from patients and the health insurance growing rapidly. In 2003, it was estimated at agency. The share of the latter has increased in 25.4 dollars per person per year. By 2006, it recent years. As for the budget, it remains the may not be too far off the 45 dollars benchmark most important income source for hospitals, but proposed by the WHO for poor countries. its contribution to the health system as a whole has been on a declining trend (Figure 8.5). A rapid increase in health spending is not Within government, there has been a transfer of surprising, as this is a common pattern as responsibilities to the provinces. By 2000, countries grow richer. Across countries, a one- around 60 percent of public spending on health percent increase in GDP per capita is associated was undertaken by provincial governments, on average with a 0.71 percent increase in the with better-off areas raising an even greater share of health spending in GDP. However, in share of revenues from their own resources. the case of Vietnam spending may be growing even faster. Out-of-pocket expenditures remain the main source of income for the health care system in From an input perspective, it is largely wage Vietnam. Of total household spending on costs that have been driving hospital health, 30 percent is used to pay user fees (both expenditures up. Despite wages making up for official and unofficial) in health facilities. The only 40 percent of total hospital costs, they remaining 70 percent is spent on medicines and account for half of the growth in health medical consumables in retailers. An analysis spending. Another 30 percent is accounted for of spending associated with inpatient stays by rising prices for drugs. From an output shows that only 48 percent of it corresponds to perspective, the number of both inpatients and user fees. Indirect expenses on medicines and outpatients is growing steadily, perhaps at extra services such as food, lodging, travel and around 6 to 7 percent per year. But hospitals are gifts for health workers account for the rest. also spending more per patient, especially on diagnostic tests such as X-rays and ultrasounds, Out-of-pocket expenditures are large enough to and on surgery. potentially result in higher poverty. Of course, Figure 8.5: Mostly from Out-of-Pocket Payments Source: MOH (2007a). 80 HEALTH CARE it is always difficult to assess what would have expenditures suggest that important, unresolved happened in their absence, as households would problems remain in the organization of the have probably behaved differently. For health system. instance, if they did not need to save so much for health care perhaps they could have chosen The reform strategy adopted by the government to work less hard. But a crude estimate of the is clear but its success will hinge on the poverty impact of out-of-pocket expenditures organization of health insurance, the main can be obtained by assuming that the money mechanism chosen to extend access to health could have been used to support consumption. care and finance the delivery of health services In that case the poverty rate of Vietnam would over time. Given its importance, this is an issue have been 1.1 percentage points lower. This that deserves to be treated separately. But estimated impact is bigger than in other success will also depend on a good countries in the region, except for China where understanding of the peculiar nature of markets it is more than twice as large. Also, according in the health sector, where information to one estimate Vietnam has the highest asymmetries and monopoly power are common, incidence of catastrophic health care spending and may result in inefficient outcomes. In this in the world, together with Brazil. context, just relying on market incentives and financing participation in mainstream programs User fees may not be the main factor behind by the poor (the two main principles of Doi Moi large out-of-pocket expenditures in Vietnam, as reforms) may prove insufficient. shown by the importance of spending on drugs and other medical consumables. But in One area in need of attention is the delivery of combination with the increased autonomy for health services to those who are exempted from service delivery units, user fees might have the payment of user fees, or simply cannot contributed to this outcome. afford them. Decree 10 aimed to enlarge the operational and financial autonomy of hospitals, User fees are a mixture of charges per service providing their managers with wide latitude in (for instance, in the case of diagnostic tests) and all areas, except in setting fees. Hospitals were charges per day (for inpatient stays). There are allowed to earn net income, distribute it among authorized ranges for each type of fee and also their staff and invest it. For instance, they variations according to the type of hospital. The establish wards for fee-paying patients, introduction of user fees did not result in higher providing better-quality care than in regular charges by district-level hospitals, which were hospital areas. not initially subject to Decree 10. But in provincial hospitals it led to a 47 percent This latitude in the use of revenue from fees increase in out-of-pocket income by the third aligns the incentives of health staff with the year of implementation. Whether such an interests of fee-paying patients. But in the increase was justified in medical terms is absence of tight monitoring, it can be expected difficult to tell. But there is a risk that to result in a relative neglect of those who autonomous hospitals simply induced demand cannot afford the fees, or are officially exempt for their services. from paying them. To the extent that hospitals offer health care interventions not available Market imperfections elsewhere in their vicinity, there is also an incentive to set fees at an excessively high level. And with patients depending on health Solid health outcomes on average, good quality personnel to tell them what they need, the of care at the grassroots level and the rapid temptation to prescribe too many drugs and expansion in the availability and quality of diagnostic tests could be difficult to resist. health services are all accomplishments that Vietnam can be proud of. However, inequalities Decree 43 tried to realign incentives towards in health outcomes, the ballooning of health compliance with social mandates. Under this care costs and the importance of out-of-pocket decree, hospital staff only get a share of the 81 SOCIAL PROTECTION revenue from fees after recurrent costs are percent of outpatient visits. But patients are recovered (partly or fully) and after the first 25 poor judges of the quality of care they receive. percent of the revenue is set aside for the In this context, the certification of providers upgrading of facilities. A cap was introduced on becomes crucial. Certification will also be the wage bill of hospital personnel, which required if health insurance is to become the key reduces their incentive to charge excessively mechanism to pay for health services, as high fees. Decree 43 also refers to the need to foreseen by Resolution 46. Making a private monitor the quality of services provided. physician eligible for the reimbursement of However, Vietnam lacks credible quality control expenses by VSS is a straightforward way to mechanisms in health care, and no reliable signal his or her medical expertise. But with at monitoring system exists. Setting up such least 65,000 private health units in Vietnam, system would be a major task. Consumer going through this certification process in any associations should be involved in it. substantive way would be a vast logistic challenge. A second, related problem with the current direction of reforms concerns the certification The certification of providers should also of providers. By now, a large fraction of out-of- extend to non-government actors. Their right to pocket expenditures is on services delivered by provide social services was officially private health units, which account for 60 recognized in 2004. However, their role does Box 8.2: Drug Production, Commercialization and Pricing The patent system, developed under the WTO Trade Related Aspects of Intellectual Property Rights (TRIPs) agreement of 1995, established a global standard of 20 years of patent protection for all forms of technology, including medicines. Under the patent system, pharmaceutical companies have not always offered medicines at affordable prices in developing countries, thus placing financial strains on national health programs and on people who must pay for medicines out of their pocket. Under the original TRIPs Agreement, developing countries were therefore provided with safeguards and flexibilities to promote and protect public health. The right of developing countries to protect public health was reaffirmed under the Doha Declaration, which was agreed upon by all WTO Member States in November 2001. Generic competition is an effective approach to reducing medicine prices. TRIPs safeguards introduce generic competition during the patent term and ensure rapid introduction of generic medicines upon patent expiration. Thailand, for example, exercised a specific safeguard, known as compulsory licensing, for three medicines to treat HIV/AIDS and cardiovascular disease. Before it did so, patent holders were unwilling to reduce the price they charged for these medicines. But in response to the use of this safeguard, the Ministry of Public Health received significant discounts. This confirms the importance of compulsory licensing as a tool to prevent the abuse of monopoly power by pharmaceutical companies . Developing countries are under pressure to introduce stricter levels of intellectual property rights protection, especially in the context of bilateral or regional free trade agreements. For instance, since 2002 all free-trade agreements with the US include "TRIPs-Plus" articles. However, there are concerns that the implementation of TRIPs-Plus could lead to increased expenditure on medicines. And it is not clear that higher protection of intellectual property rights will encourage pharmaceutical research that is relevant for the needs of developing countries. Indeed, less than 2 percent of the new medicines launched since 1999 address developing country diseases. As public health demands continue to place a strain on public health systems in developing countries, it will be increasingly important for intellectual property right rules to closely adhere to the letter and sprit of the Doha Declaration on TRIPs and Public Health. Source: Oxfam International (2006). 82 HEALTH CARE not seem to have received the importance it However, there is also scope for a more active deserves, and there is a perception that they stance in relation to cheaper, generic drugs. have been treated marginally in designing Monopoly power by laboratories is intrinsic to the policies for the health sector. system through which pharmaceutical research is encouraged. Drug discovery and development is A third area which needs attention relates to a complex, lengthy and costly process. drug policies. A large fraction of out-of-pocket Intellectual property rights are used to restrict the expenditures in Vietnam goes into buying imitation of discoveries for some time, thus medicines. This is partly due to limited allowing laboratories to recover their research and competition in the domestic market for drugs. development costs through high prices. But from Single distributors often control large segments a humanitarian perspective this is harder to justify. of it, which results in high markups over This is why developing countries can regulate imported prices. From this perspective, it is intellectual property rights in ways that favor the reassuring to see the emphasis of the development of cheaper, generic products, while Pharmaceutical Law of 2005 on fostering encouraging domestic pharmaceutical research competition. The Law approaches market (Box 8.2). The government of Vietnam should regulation on the basis that manufacturers and explore such ways as it strives to implement its traders are entitled to compete in terms of price. commitments towards the WTO. 83 9. HEALTH INSURANCE After a decade of sluggish growth, participation more efficient, and establishing dedicated funds in Vietnam's health insurance program is now to cover the cost of chronic illnesses and life- expanding rapidly and could soon reach half of threatening diseases for the population at large. the population. The program is actually made of several schemes. The original ones covered the One program, several entries formal sector of the economy on a compulsory basis, and the rest on a voluntary basis. The In 1992, four years after the introduction of user main innovation behind the current rapid fees under Doi Moi, the government launched a growth in enrollment was to create a new entry health insurance program. At the time, it modality, whereby the government directly pays comprised two schemes: one compulsory, for the premium on behalf of targeted population workers in the formal sector, the other groups (disadvantaged people, children below voluntary. six years of age...) This approach has been effective at increasing coverage among the poor, The compulsory scheme covers mainly civil to whom it has provided better access to health servants, workers in SOEs and those in services. The same principle will be applied to relatively large private sector firms. Initially, provide partial subsidy to the near-poor so that only enterprises with ten workers or more were they can participate on a voluntary basis. obliged to participate, but this size threshold However, the health insurance program has also was eliminated in 2005. The contribution to the resulted in an escalation of health care costs, to scheme is set at 3 percent of the salary, of which the point where the program ran a deficit for the 2 percentage points are formally paid by the first time in 2007. Two main forces underlie employer. Protection by this scheme does not this trend. First, providers face an incentive to extend to family members. But people of merit deliver both more treatments and more and pensioners are covered at the government's expensive treatment than may be warranted on expense. medical grounds. Second, the expansion of the program suffers from adverse selection, The benefit package is quite generous, as it meaning that heavy users of health care services includes outpatient and inpatient services at all are more likely to enroll. Addressing these two levels of health care. It also includes laboratory problems through a higher premium for the exams, X-rays and other diagnostic imaging voluntary scheme or a higher social security procedures, as well as drugs listed as contribution for the compulsory scheme would reimbursable. The health insurance fund even only offer temporary relief. Such response covers expensive high-tech services such as could undermine the expansion of coverage and Magnetic Resonance Imagery, hemodialysis and the formalization of employment. More open-heart surgery. Insured patients are eligible promising alternatives include strengthening the for health care provided not only at public procedures through which patients are referred health facilities, but also in private units that to more advanced and expensive facilities, have contracts with the health insurance agency. modifying the rules used to reimburse health care providers so as to encourage them to be Despite the low contribution rate and the 84 HEALTH INSURANCE generosity of the package, until not long ago Group enrollment was encouraged, to better compliance with the compulsory scheme was pool risks between healthy participants and only partial. By 2005, mostly all SOEs those who are more likely to become heavy participated in the program, but only a fraction users of health services. Education facilities, for of private enterprises did (Figure 9.1). instance, can enroll their pupils only provided Moreover, assessing compliance on this basis that at least 30 percent of them participate. In is potentially misleading, as even among 2007 the focus shifted to household participating enterprises only a fraction of the membership. Benefits were also revised with workers tends to be declared, and only a the introduction of waiting time for high fraction of their wage is used to calculate the technology treatment and maternity. Given the contribution. If the average contribution rate slow expansion of the voluntary scheme, there paid by the enterprise is considered instead, is a debate as to whether to remove all the shortfall appears to be much larger. conditions on group participation. Including contributions towards old-age Resolution 46 gave a new impetus to the health pensions and other benefits, enterprises should insurance program. It was followed shortly be paying 17 percent of their wage bill to VSS. after by Decree 63, which significantly In 2005, the actual figure is between a third increased the scope and depth of the benefit and a half of that. package for both schemes. Importantly, the The voluntary scheme is aimed at the rest of the government chose to directly finance population, including the families of contributors participation in the program by increasingly to the compulsory scheme. Payments are on a large population groups. In practice, this flat rate. Since 2007 the premium varies by amounted to a new entry modality into the health insurance program region, according to their development level. Co-payments had been eliminated in 2005, but Vietnam was the first country in Asia and the were reintroduced in 2007. second in the world to ratify the UN Convention Figure 9.1: A Varying Degree of Enterprise Compliance Source: Based on data from the 2005 Enterprise Survey by GSO. 85 SOCIAL PROTECTION Figure 9.2: A Rapid Expansion in Coverage Source: Based on data from VSS for 2006. Series on population groups are measured in millions of insured (left axis); the series for the total is expressed in percent of the population (right axis). on the Rights of the Child. In connection with to 3 percent of the minimum wage is being this move, in 2005 it was decided that public considered as well. health facilities would provide health care free of charge to children below six years of age. A partial subsidy would also be extended for the Such facilities were also mandated to hand over near poor to enrol. In late 2007, the decision free health insurance cards to children in need of was made to support half of the health insurance medical attention. The government, in turn, premium for this group. Several provinces had committed to making the necessary transfers to already embraced this approach and were using finance the additional costs. their own resources to top up the HCFPs and support participation in the health insurance A similar approach was adopted in the case of program by the near poor. disadvantaged groups. Provincial HCFPs had been created under Decision 139, in 2002, to Who is covered? finance the health costs incurred by the poor, by ethnic minority groups in six northern and five Growing formalization, increased enrollment highland provinces, and by people living in in the voluntary scheme with government difficult communes covered by Program 135. support, and above all the extension of the In practice, HCFP resources had been used to program to targeted population groups have directly reimburse public health care units for resulted in a rapid expansion of health the services rendered to these groups. But in insurance coverage, from around 25 percent of 2005 provinces were instructed to use HCFP the population in 2004 to more than 40 percent resources to directly purchase health insurance at present (Figure 9.2). cards for everyone in the targeted population. The initial allocation, out of the central budget, It is often argued that health insurance is was 50 thousand VND per year per beneficiary. unlikely to result in substantial coverage. In It will be raised to 130 thousand from 2008 developing countries, the argument goes, the onwards. Increasing the government support for formal sector of the economy is small, so that the contribution by the poor to a level equivalent compulsory schemes can only reach a fraction 86 HEALTH INSURANCE of the population. As for voluntary schemes, quintile of the population, and declines as they suffer from adverse selection, meaning that increasingly better-off groups are considered. only the elderly and those in need of health Admittedly, there is some leakage too, in the treatment have an incentive to enroll, whereas sense that individuals who are in principle not the young and healthy majority of the eligible do get support from HCFPs. But population is unlikely to do so. leakage appears modest overall, and it happens mainly among the poorest population groups While all this is true in Vietnam's case too, anyway. financial support by the government and the active involvement of mass organizations in The conclusion is similar when considering the encouraging participation make the word distribution of HCFP resources across regions, "voluntary" lose some of its original meaning. rather than across population groups. Three From this perspective, discussing the merits and quarters of the funding goes to the Northern demerits of social health insurance in general, or Mountains, the Mekong River Delta, the North its potential to eventually reach universal Central Coast and the Red River Delta, in that coverage, may not be particularly enlightening order. On a per capita basis, however, the South in the case of Vietnam. A more relevant East receives almost four times the allocation of discussion is whether the health insurance the Northern Mountains, and six times the program is reaching those in need of support, allocation of the North Central coast. and whether it is helping them gain access to affordable health care of reasonable quality. The bad news is that the program is not covering all vulnerable groups. Based on the criteria set On the program's coverage, there are both good by Decision 139, as of 2004 roughly 31 percent and bad news. The good news is that resources of the population ought to have been covered by from HCFPs are relatively well targeted towards HCFPs. But by the end of 2006 the real figure the intended beneficiaries (Figure 9.3). was only around 22 percent. The situation Participation is highest among the poorest might have improved since then, both due to a Figure 9.3: Targeting and Leakage in Decision 139 Source: World Bank (2007c). 87 SOCIAL PROTECTION Figure 9.4: Vast Disparities across Regions Source: Based on data from VSS. better operation of HCFPs and to the fall in the health insurance coverage does not affect out- proportion of the population living in poverty. of-pocket expenditures or, if it does so, the However, some vulnerable population groups effect is not large or statistically significant. are unlikely to be covered. Migrants are one of Only one study, on the voluntary scheme in Hai them. According to a recent study, for the small Phong, found a very large impact. A couple of portion of migrants that was sick and treated at other studies estimated out-of-pocket savings in health facilities, 84 percent had to pay for the the range of 20 to 35 percent, whereas more cost of services and medications out of their than a dozen of them reported no effect. Also, own pockets. Only 12 percent had the cost health insurance might have reduced the covered by their family. None of them had incidence of catastrophic health spending, but health insurance. not dramatically. If the estimates are to be interpreted literally, HCFPs result in Access to services catastrophic spending declining by 3 to 5 percentage points. The main goal of health insurance is to reduce the cost of medical treatment, especially in the However, looking only at out-of-pocket event of catastrophic shocks. In an actuarially expenditures is potentially misleading. If health fair program, the total amount insured people insurance coverage makes access to health spend on health, on average, should not be services more affordable, participants can be affected by such shocks. The effect of insurance expected to make a more intensive use of such would be to even out payments for health care services. The combination of lower costs and across the population. Those payments would higher use is certainly associated with increased not vary so much according to whether people wellbeing. But it is not necessarily associated get sick, but rather according to their earnings with lower out-of-pocket expenditures. The real (in the mandatory scheme) or not at all (in the issue, then, is whether participants in the voluntary scheme). program make a more intensive use of health services. The findings of available studies are Most estimates from Vietnam suggest that more consistent in this respect. 88 HEALTH INSURANCE Health insurance coverage increases the health insurance cards. In fact, limited probability of admission to hospital and the understanding extends beyond this group, length of stay in hospital. To a lesser extent, it affecting some times health staff at local levels. increases the use of outpatient care as well. Also, in many opportunities patients who are Studies also find evidence of health insurance covered by health insurance do not use their coverage encouraging the use of public sector cards when accessing services. facilities, instead of private health units and pharmacies. However, the impact varies across Access barriers are less severe in the case of income groups, with the poorest group children below six years of age. Reports by experiencing the smallest increase. provincial health departments claim that 96 percent of those children had received their free Impacts appear to vary across schemes. For healthcare cards. This claim is not reflective of instance, the effect of insurance on the the overall picture. If figures from the probability of admission to hospital is smaller population census are used as a reference, then among participants covered by HCFPs than only two thirds of the eligible children have under the compulsory scheme. And it is been given cards. However, this in itself should smaller still for those in the voluntary scheme. not be a major obstacle. MOH guidelines There is also evidence of impacts varying with stipulate that children under six without health income. Membership of the mandatory scheme care cards can still receive free services at increases the likelihood of hospital admission public facilities as long as they present for all income groups, but the effect is most "equivalent papers", which include birth pronounced for middle income groups. Among certificates and commune birth registration lower income groups, mandatory health documents. insurance has no significant impact on the length of stay. Worryingly, no significant A spending bias impacts on inpatient care are evident among the poorest 10 percent of the population, which Health insurance can also be expected to lead to may result from other barriers faced by this an overall increase in health spending. To some group to access health services, notably extent, this is so by design. Reducing the out- transport and time costs. of-pocket cost of medical attention is bound to lead to an increased use of health services, as There are also some implicit barriers this group has happened in Vietnam. This can be faces, as they have to first seek care at CHSs considered as a positive outcome, especially in whose quality of service is low. Biased the case of disadvantaged population groups treatment and poor attitude of health staff who would have otherwise relied on self- towards patients from this group also present a medication or would not have sought any significant obstacle on their way to seek care. treatment. But there are also less positive Other barriers are related to hidden costs, such reasons why health insurance may lead to an as transport to the newest service provider, loss escalation of health care costs. of earnings or the price it medicines. On the users' side, in the absence of universal Qualitative studies and focus group discussions coverage there is adverse selection, in the sense tend to confirm that health insurance coverage that it is people with poorer health who tend to may not be enough to gain access to services. enroll. Those who are covered by the program Throughout, beneficiaries of Decision 139 state are therefore likely to use medical services more that they receive less attention than patients intensively than those who are not. On the paying on a private basis, particularly from the providers' side, unless the program is very well level of district hospitals upwards. It also designed, there is an incentive to deliver more appears that beneficiaries of Decision 139 are services than is warranted and to choose more not fully aware of the benefits they are entitled expensive services too. This is because of the to, or do not understand the rights associated to practical difficulties in checking all medical 89 SOCIAL PROTECTION interventions whether are justified. In the care, including more tests and more expensive absence of solid, wholesale "purchasing" skills, drugs, to insured patients. the health insurance agency may simply function as a largely passive payer of bills. As for the adverse selection problem, it is apparent in the dispersion in the number of Data from hospital inventories confirm that health care contacts across groups of insured health insurance is associated with an increase (Figure 9.5). This number is modest in the case in the quantity of services provided, relative to of students, who are young and mostly healthy. the number of outpatient consultations and It is modest in the case of the poor as well, inpatient admissions. For instance, the fraction because they cannot afford to stay out of work of births delivered through a C-section is higher for long and also because of the more the larger the fraction of hospital patients who substantial barriers they face in accessing health are insured. And a similar regularity can be services, from knowledge to transport costs. observed in relation to the number of Formal sector workers and merit groups occupy biochemical and microbiological tests an intermediate position. Not surprisingly, the conducted. highest number of contacts is associated with pensioners and with those who enrolled In the case of the poor, household survey data voluntarily in the health insurance program. from 2004 show that health insurance coverage is not associated with lower out-of-pocket Some studies have tried to understand who expenses per outpatient consultation or per enrolls in the voluntary health insurance inpatient admission. This absence of savings scheme. One of such studies focused on could be due to the fact that some expenses, students aged from six to 20 years of age. It such as non-prescription drugs bought from found that the educational attainment of the vendors, are not covered by the program. household head and the level of household However, the high out-of-pocket payments per expenditures had a strong association with contact incurred by the insured likely results enrollment. Other things equal, a female head from providers delivering a different style of of household was much more likely to purchase Figure 9.5: Health Care Contacts by Group of Insured Source: Estimates based on 2006 data from VSS. 90 HEALTH INSURANCE health insurance for her children than a male behalf of 4.5 million public sector employees. head. A short distance to the district hospital In addition, it directly or indirectly covered the and the presence of a public hospital in the premium of another 18 million participants, commune were also positively associated with among them public pensioners, merit groups, enrollment, whereas the availability of private the poor and other disadvantaged people. health units had a negative effect. But overall, the study found little evidence of adverse Securing income from other sources has proved selection among this group, as the height-for- more elusive. For instance, there is a substantial age score was higher among enrolled children. gap between the contributions formal sector firms pay towards social insurance and the The conclusion was different in the case of contributions they report to the tax informal sector workers. A study of enrollment administration office for the purpose of tax in the voluntary scheme among this group found exemptions. Also, not all provinces use the some differences across industries. For instance, resources from the HCFP to purchase health those working on a self-employment basis in the insurance for targeted population groups. Lang health sector were considerably more likely to Son, for instance, provides health insurance have enrolled. There were differences by cards to the poor only when they need medical occupation too, with those who classify attention, keeping its HCFP resources for other themselves as skilled being more likely to enroll uses. Meanwhile, VSS has substantial than the unskilled. Women were more likely to outstanding bills with hospitals. enroll than men, and workers from high-income households were more likely to enroll as well. At a deeper level, however, the problem is the But unlike the case of students, the study found escalation in the costs per enrollee. The some evidence of adverse selection among this expansion in coverage is insufficient by itself to group, as informal sector workers who had been account for the growth in expenditures. Had confined to bed or had taken time off work costs per enrollee stayed at their level of 2003, because of illness were more likely to enroll. expenditures would have gone up by only 30 percent of their actual increase (Table 9.1). Financially viable? Moreover, there is a clear gap between income and expenditure for two groups of insured, At the beginning of 2007 the deficit of the namely pensioners and participants in the voluntary health insurance scheme was growing voluntary scheme other than students (Figure so rapidly that VSS temporarily suspended its 9.6). These worrying developments underscore operation. In barely two years, expenditures had the need to keep treatment costs down, or at doubled and the balance of the health insurance least to keep their rate of growth in line with program as a whole had gone from a surplus of income growth. 400 billion dong in 2004 to a deficit of 1.2 trillion in 2006. The voluntary scheme was the main source of the problem, with the Smarter spending compulsory insurance scheme not being able to compensate for the shortfall anymore. In early Containing treatment costs per insured person 2007, for the first time, VSS was no longer and addressing the adverse selection problem capable to fully reimburse the health hold the key to making the health insurance expenditures of its members. Only after co- program financially viable. Alternatives such payments were reintroduced in the voluntary as raising the premium for the voluntary scheme, and the premium for the voluntary scheme, or the contribution rate for the program was almost doubled, was the issuance compulsory scheme, can only offer temporary of health insurance cards authorized again. relief, because they do not correct the tendency towards ever greater spending. Moreover, The government is by far the biggest source of adjusting on the revenue side would have the VSS funding. In 2006, it paid contributions on unintended consequence of reducing coverage. 91 SOCIAL PROTECTION Table 9.1: Average Spending by Type of Insured Reimbursement per card (thousand dong) Insurance group 2003 2004 2005 2006e I. Compulsory scheme 97 126 180 171 Pensioners 145 325 479 717 Formal sector and merit groups 136 171 238 265 II. Voluntary scheme 16 21 45 151 Students 16 19 26 56 Other voluntary 65 136 701 III. Poor and disadvantaged 34 52 71 56 All 73 91 126 166 Source: Based on data from VSS. Figure 9.6: Costs and Revenues of the Health Insurance Program Source: Based on data from VSS. Higher contribution rates, in particular, would They involve directing patients towards less slowdown the formalization of the economy, expensive health care facilities whenever especially at a time when minimum wages appropriate, removing the incentive for service need to grow rapidly in order to comply with providers to deliver unnecessary diagnostics WTO commitments. and treatment, and moving the funding for chronic illnesses and catastrophic shocks out of In practice, several complementary approaches the program. could be considered to contain treatment costs and address the adverse selection problem. In the parlance of health specialists, these 92 HEALTH INSURANCE approaches concern referral procedures, health care more generally. Its procedures, in provider payments and specialized funds terms of referrals or reimbursements, stand a respectively. At the risk of simplifying, stronger chance to become industry standards. If they referral procedures and better-designed provider were well designed, they could help contain the payments aim at containing treatment costs per broader escalation of health care costs in insured person, whereas specialized funds could Vietnam. be used to mitigate the adverse selection problem. Specialized funds could also The main weakness of Vietnam's referral system contribute to making the system more equitable, at present is its inability to prevent the insured at a time when the majority of the population is from "jumping" levels in the hierarchy of health not yet covered by the health insurance care facilities and choosing unnecessarily program. expensive ones as their first point of contact. In principle, CHSs should transfer patients to Before describing these approaches in more district hospitals, district hospitals should detail, it is important to note that their financial transfer patients to provincial hospitals, and implications could go well beyond the health provincial hospitals should transfer patients to insurance program. VSS is gradually becoming central hospitals (in the South to HCMC and in a major source of income for hospitals, as the North to Hanoi). In serious cases, levels can important by now as user fees (Figure 9.7). be legitimately skipped. But bypassing seems Meanwhile, the share of income directly to be the rule more than the exception. For provided by the budget is gradually declining. instance, a study on paediatric emergency It amounted to 47 percent in 1998 but was down shows that about half of children patients check to 37 percent by 2006. VSS is also increasingly directly into central hospitals, despite many of passing contracts with private health care units. them not being from the city in which the The growing importance of VSS as a wholesale central hospital they have checked into is purchaser of health services makes it a natural located. Such bypassing causes overload in candidate to encourage efficiency gains in expensive central hospitals, distracts the best Figure 9.7: Sources of Hospital Income Source: MOH, Vietnam hospital inventories. 93 SOCIAL PROTECTION specialists towards minor cases, and results in and daily rates for inpatient stay was higher health care costs on average. established. Ceilings apply at low levels in the system, but not higher up. The schedule has One important motivation for patients to skip been revised, with the introduction of new referral levels is to avoid what they perceive as procedures and the pricing of some new drugs. bad quality of service. CHSs, which are But the incentive remains to overspend. officially the first in line, do not have a good Admittedly, the inspection unit of VSS performs reputation among the population at large, with a control of the charges reported, trying to health workers at higher levels or even with the assess whether they are plausible. But in health insurance agency. There is practice, this is a purely administrative check, dissatisfaction even among the poor, despite the resulting in a collective invoice being submitted CHS being their main access point for medical by the provider. The consequence has been an care, above the pharmacy and the VHW. A escalation of refunds at higher levels. For recent survey revealed discontent with the range instance, in 2007 one central hospital in Hanoi of equipment and the quality of examinations at absorbed 5 percent of the total amount CHS level. Perceptions were more positive in reimbursed by VSS nationwide. relation to district hospitals and above, except on the friendliness of health staff which appears It would make sense for VSS to assume the role to worsen. However, there are reasons to be of an informed wholesale purchaser, linking dissatisfied with the quality of care at those refunds to medical conditions more than to levels too. In 2003, only 59 percent of patients services provided and specifying the required referred to provincial hospitals had been quality standards. correctly diagnosed at the district level, and 75 percent of patients referred to central hospitals Experimentation with alternative provider had been correctively diagnosed at the payment mechanisms has been going of for provincial level. some time already. Capitation was introduced in Hai Phong (since 2005) and Hai Duong (since To mitigate the skipping of referral levels, a top 2007). After six months of operation, good priority is to improve the quality of health care compliance with the delivery of the services and provided by CHSs. But it is also important to prescription of generic drugs was observed, strengthen referral procedures, so that whereas the administrative workload decreased prospective patients do not unnecessarily jump significantly. However, there were concerns over lower-level facilities. To achieve this goal, about the difficulty to finance the introduction it could be more efficient to provide financial of high-tech procedures. There were also initial incentives to the facilities than to penalize steps towards establishing a pathology-based prospective patients. The health insurance reimbursement system. The analysis concluded agency could apply a discount on the that in the case of conditions with limited cost reimbursement of services to hospitals if the variation, such as appendicitis and C-section, patient was not appropriately referred by the DRG payments would be workable. However, lower level. This could create some pressure DRGs are difficult to introduce on a large scale from district hospitals to their affiliated CHSs to without appropriate (and highly elaborate) ensure that they comply with referral costing and controlling mechanisms. procedures. As for the CHSs themselves, VHS Finally, some consideration should be given to could pay them a flat fee for consultation the mechanism to fund the treatment of provided. chronic illnesses and life-threatening diseases, Paying health care providers on a fee-for- such as diabetes or cancer. These are the service basis is not necessarily a good idea at medical conditions resulting in catastrophic higher levels in the system (Box 9.1). However, shocks for household. The current coverage this is what VSS does since the mid 1990s, of the health insurance program implies that when a mixture of fees per outpatient service more than half of the population is exposed to 94 HEALTH INSURANCE Box 9.1: Alternative Provider Payment Mechanisms Fee for service. The amount reimbursed to facilities is based on a user fee schedule, with fees set at either the same level charged to uninsured patients or below. One important advantage of this approach is that it allows a rapid cost-recovery for investments in new technologies. There is also an incentive to provide all necessary care regardless of the patient's wealth. A major inconvenient is that it encourages the delivery of more services, and more expensive services, than may be warranted. To counter this tendency, the amount paid per patient or per case tends to be capped. In Vietnam, a list of procedures with prices that are reimbursed by health insurance was developed in 1995 and supplemented with new items and prices in 2006. Private health units signing contracts with VSS are reimbursed at the same rate public facilities. A drug list identifies all the medicines reimbursed by VSS, and also the level of the health care system authorized to prescribe them. Capitation. The amount reimbursed is based on the number of members served by the provider. Because of its simplicity, this approach is well suited to areas with limited administrative capacity. It has been successfully introduced in Thailand and Lao PDR, and Vietnam is considering it for its CHSs and district hospitals. It is also useful for general practitioners, who have the responsibility to steer patients through the health care system. However, when budgets are not sufficient, the capitation approach creates an incentive for providers to withhold services or downgrade their quality. This is the reason why capitation is not often used for hospital reimbursement. An alternative worth considering is morbidity-weighted capitation, whereby reimbursements depend on age, gender and other easily observable characteristics of the patient. Case-base refunds. The amount reimbursed is related to a medical condition, determined mainly on the basis of diagnostic tests. This so-called Diagnostic Related Group (DRG) approach can provide strong incentives for efficiency in the delivery of services. But the challenge is to set reimbursement rates at a level that reflects the costs that would be incurred if appropriate care was delivered rather than current costs. In practice, DRG includes "pathways", specifying the treatment to be used after each key diagnostic. However, building such pathways is difficult without sound and reliable treatment data. In Vietnam, Decree 63 called for the development of alternative payment mechanisms. An informally stated goal is to adopt DRG payments at provincial and central levels for around 700 medical conditions. Implementation could take five to seven years. such shocks. Moreover, providing funding for chronic illnesses and life-threatening these conditions through the health insurance diseases. Resources should be provided program can only encourage those who suffer directly by the budget. But they could be from them, or are more likely to suffer from managed by the health insurance agency, them, to enroll. This, in turn, can only using provider payment mechanisms similar aggravate the adverse selection problem. to those applied in the case of health insurance. This would contribute to shifting While Vietnam continues to gradually extend the funding of health care to the demand side, the coverage of its health insurance program, and would strengthen the role of VSS as an one or several dedicated funds could be informed wholesale purchaser of health care established to cover treatment costs for services. 95 10. OLD-AGE PENSIONS The old-age pension program of Vietnam has sector, the compulsory pension program is recently been revamped with the aim of solid financially. Its annual surplus is roughly making it financially sustainable and 3.7 percent of GDP at present, and it could extending its coverage, on a voluntary basis, accumulate a massive amount of reserves in to the informal sector of the economy. There the coming decades. But how large that are few pensioners for now. Almost all of amount will be will very much depend on how them used to work for government or for pension benefits are indexed to inflation. SOEs. But multi-generation family Meanwhile, the management of reserves could arrangements imply that almost one in ten be substantially improved. At present most households in Vietnam receives pension investments are in low-yield instruments, income. The distribution of such income is which results in a return below inflation. clearly regressive, in the sense of being concentrated among the richer part of the Life in old age population. However, other things equal, the poverty rate would be several points higher in A majority of the elderly are women. The the absence of pension income. More female share of the population rises with age, important than the current distribution of from 58 percent for those over 60 years to 66 beneficiaries is the prospect of extending the percent above 80. A majority of the elderly also coverage of the program. The number of lives in extended families. More than half of participants in the compulsory program has them are part of three-generation households, grown steadily, and represents by now 18 including working age adults and children. But percent of the labor force. Whether the 37 percent of them live in households voluntary program for the informal sector will containing elderly people only. have many takers is not clear. The experience with the largest voluntary pension program in Old-age is a period of fundamental operation in Vietnam, in the province of Nghe transformation in life. The amount of time An, is not particularly encouraging. Many of devoted to work, in the household and out, its participants would not be inclined to join decreases quite steadily (Figure 10.1). the new voluntary pension program because Throughout old age, women work longer hours the minimum contribution is too high and than men, with most of the difference due to because they are unlikely to reach the 20 years their higher involvement in household work. of contribution needed to gain access to a But even by age 70, both men and women are pension, as opposed to a lump sum. The still devoting roughly 20 hours every week to formulas used to compute pension benefits non-household work. This reality is in sharp also result in important deviations from the contrast with the rigid retirement rules of many actuarially fair level. Overall, those working old-age pension programs, which impose that for government get a better deal than those people cease working at a specified age. Old- working for businesses. With growth in age is also a period of more recurrent or enrollment coming mainly from the enterprise chronic sickness. 96 OLD-AGE PENSIONS The elderly live in households whose total Pension programs income is slightly higher than the average for the Vietnamese population (6.4 million dong In the period preceding Doi Moi Vietnam only per year in 2004, compared to 6.1 million for had a non-contributory old-age pension the population at large). Their households have program for civil servants and workers in lower income from trade and business SOEs. In 1993 coverage was extended to activities, but this shortfall is more than domestic private enterprises and foreign compensated by larger remittances and social companies, on a contributory basis. The security transfers. About 90 percent of the entitlements of civil servants and SOE elderly live in households receiving workers were then "grandfathered" into the remittances. The urban elderly are more likely new program, with the budget taking to receive remittances from abroad; those from responsibility for their funding. The program ethnic minorities much less so. However, is since then administered by VSS. In remittances and transfers are only one way to addition, a non-contributory pension scheme provide the elderly with resources. Living in is in place for people aged over 85 who lack extended families and pooling incomes remains resources. However, such non-contributory a much more powerful mechanism for now. scheme is administered separately by In spite of their slightly higher income, MOLISA, as one of its targeted poverty households including elderly people are alleviation programs. slightly more likely to live in poverty. This is The system is now undergoing an because they tend to be larger. Other things unprecedented reform. The first ever Social equal, the probability for a household to live in Insurance Law was passed in 2006, with its poverty is 4 percent higher if it includes various building blocks being implemented someone aged over 60. Households with between 2007 and 2009. The reform aims to elderly that are headed by a woman are gradually expand the coverage of the system significantly more likely to be extremely poor. and to ensure its financial sustainability. The Figure 10.1: Working Less, Getting Sick more Often Source: Based on preliminary data from GSO. 97 SOCIAL PROTECTION Box 10.1: Other Benefits in the Compulsory Program Sickness. This is a cash benefit paid to an employee suffering from an illness that prevents him or her from working. The benefit can be paid during 30 to 60 days, depending on the number of years of contribution. A maximum of 180 days per year, irrespective of the contribution record, is available for diseases requiring long-term treatment. The benefit amounts to 75 percent of the wage earned in the month immediately before ceasing work. Maternity. A cash benefit is paid to a breast-feeding woman for to six months after delivery, depending on her working environment and health condition. The beneficiary is also entitled to up to five days of leave for pre- natal health checks. Occupational disease or injury. A benefit under the form of a lump sum or a monthly payment is provided to compensate for the loss of working capacity resulting from occupational injury or disease. The benefit covers injury incurred during business trips or in the journey to and from work. Both the lump sum payment and the monthly benefit consist of two parts. One is based on the minimum wage, the other on the years of contribution and the average salary. In case of injury, the employer is responsible for medical expenses and salary for the duration of the treatment. VSS pays the benefit from the date of hospital discharge. Disability. Participants who lose at least 31 percent of their working capacity are entitled to a disability pension. The size of the benefit depends on the extent to which working ability is affected. The minimum benefit is equivalent to 30 percent of the minimum wage. It increases by 2 percent of the minimum wage for each additional point of working capacity lost. latter was accomplished through a series of their first 15 years of contribution, but it falls parametric changes in the formulas for pension to 2 percentage points subsequently. The benefits. The former through the establishment minimum pension is set equal to the minimum of a voluntary pension program for the informal wage. The maximum is 75 percent of the sector, designed to allow the portability of average salary. Contributions made past the benefits to the compulsory program. point when this maximum entitlement is reached accrue towards a lump-sum payment. The compulsory pension program applies to all Those with less than 20 years of contribution Vietnamese citizens working in any enterprise also get a lump-sum payment when they reach provided that they have at least a three-month the retirement age. The immediate relatives of contract. It also covers government officials, pensioners who die are entitled to a survivor military personnel and public security benefit, ranging from 40 to 70 percent of the employees. The contribution rate towards old- minimum wage per beneficiary, for up to four age pension and survivor benefits is currently 16 beneficiaries. percent of the salary. Formally, 11 percent is paid for by the employer and 5 percent by the The retirement age is relatively low, at 55 years employee. The Social Insurance Law foresees for women and 60 years for men. Early an increase of this contribution rate to 22 retirement is allowed for those losing 61 percent of the salary by 2014. percent or more of their working capicity. It results in a reduction of the pension benefit by In order to receive a pension, participants 1 percentage point of the average salary for must contribute for at least 20 years. The every year in advance. Retirement ages are 50 pension benefit is computed on the basis of an and 45 years for men and women having "average salary". In the case of women, worked for at least 15 years in hazardous jobs benefits accrue at the rate of 3 percent of the as listed by government. average salary per each year of contribution. The accrual rate is the same for men during In addition to old-age pension, the compulsory 98 OLD-AGE PENSIONS program includes a pension in the event of Limited coverage disability, and a series of short-term benefits related to sickness, maternity and occupational Participation in the compulsory pension disease or injury (Box 10.1). An additional 3 program has grown steadily over the years, percent of the basic salary is contributed albeit from a low base (Figure 10.2). At towards the first two benefits. The contribution present, there are about 8 million active rate towards occupational disease or injury contributors representing roughly 18 percent of benefits is an extra 1 percent. Therefore, in the labor force and 54 percent of employment total, participants in the compulsory social in SOEs, foreign invested companies and insurance program pay 22 percent of their private domestic firms registered under the salary towards old-age and short-term benefits. Enterprise Law. The Social Insurance Law of 2006 introduced a It is too early to tell how many participants the voluntary old-age pension program accessible to voluntary program will attract. The answer will any working-age person who is not currently partly depend on the engagement of mass enrolled in the compulsory program. The organizations to promote enrollment. But the features of the voluntary program are largely experience of the voluntary pension program for inspired from the compulsory system. The farmers set up by Nghe An province in 1998 contribution to make in order to participate has sheds some light on possible trends and to be at least equal to 16 percent of the minimum challenges ahead (Box 10.2). wage; this threshold will be gradually raised to 22 percent. Entitlement to pension benefits There are about 1.9 million pensioners in occurs at age 55 for women and 60 for men, after Vietnam at present. This represents about a a minimum of 20 years of contribution. And quarter of the population aged 60 years and immediate relatives are entitled to a survivor above, but many of those pensioners are pension. The main differences with the actually younger. Because of extended family compulsory program is the absence of a arrangements close to one tenth of households minimum pension. receive transfers from the social insurance Figure 10.2: Coverage of the Compulsory Pension Program Source: Based on data from VSS. 99 SOCIAL PROTECTION Box 10.2: Participation in Voluntary Pension Programs As of March 2007, the voluntary pension program of Nghe An province had about 85 thousand participants, of which one thousand had reached or were soon to reach retirement age. More than 60 percent of the contributors to this program are women. There are several important differences between the Nghe An program and the new, national voluntary pension program. The minimum contribution in the Nghe An program is 10 thousand dong, and the average contribution is 25 thousand. This is far less than 16 percent of the minimum wage required by the national program (the equivalent of 72 thousand dong per month in 2007). Only 5 percent of current contributors would reach the minimum contribution threshold at present. Also, in the Nghe An program almost all participants are entitled to a pension when they reach retirement age. The level of the monthly pension is equal to the sum of contributions, accrued at an interest rate of 0.6 percent per month, minus administrative expenses, divided by 120. Administrative expenses are set at 15 percent of the accrued interest. This formula would be close to actuarial principles in the life expectancy at retirement was about 10 years (or 120 months). But for a mainly female population it is probably much higher. With the establishment of a voluntary pension program on a national basis, the question is whether participants in the Nghe An program would be willing to switch to the national program and, if not, what kind of incentives they would need in order to do so. A field study was conducted to this effect in October 2007. It included a survey of 5 percent of participants in the Nghe An program. Responses revealed a high degree of satisfaction among participants, with 84 percent of them saying that they would want to continue with the program. Interviewees were also offered the option to switch to the national program, "buying" years of contributions with their actual contributions to the Nghe An program. Because actual contributions are low, only 54 percent of Nghe An participants would be able to reach 20 years of contribution at they reach retirement age. The option was then offered to the other 46 percent to "buy" the remaining years of contribution. Among the group who could aspire to a pension, 68 percent would be willing to switch to the national voluntary program. This is the equivalent of 37 percent of all participants in the Nghe An program. Among the rest, 46 percent would accept, but only provided that they can "buy" the additional years of contribution to get a pension. Very few would be interested in contributing towards a lump-sum payment at retirement age. The share would increase by an additional 7 percentage points if the minimum contribution was set at 50 thousand dong per month, instead of 72 thousand. Overall, 57 percent of participants in the Nghe An program would accept to switch in the absence of any subsidy. Disturbingly, those least interested in switching to the national voluntary program are poorer or older than the rest. Source: Paulette Castel (2007a). agency (Figure 10.3). Not surprisingly, the regressive. Almost half of pension spending fraction is much higher among urban, non-poor goes to the two richest quintiles and only 2 households, belonging to the Kinh and Chinese percent to the poorest. From this perspective, majority. This is because most of the pensioners pensions are not a pro-poor program. A of today were government employees and SOE different conceptual exercise is to subtract workers prior to 1995, when the new rules for social security income from household contributory pension were introduced. expenditures. Based on data from the 2004 VHLSS, the poverty rate would have been 4.6 Pension payments are an important source of percentage points higher, with most of the income for the households of the elderly. With impact coming from pensions. From this most pensioners being former civil servants and perspective, then, pensions do contribute to SOE workers, these payments are clearly poverty reduction. 100 OLD-AGE PENSIONS Figure 10.3: Households Receiving Pension Income Source: Based on preliminary data from GSO. Figures are for 2006. However, these analyses do not really tell what making it difficult to derive general patterns. the real poverty impact of the pension program But one practical step that can be undertaken is is. In order to assess such impact a to compare those benefits with a hypothetical counterfactual would be needed in which actuarial benchmark, for workers with different households know that they cannot count on characteristics. pension income and need to save on their own for old age. The impact of the program would be Consider the case of a man whose salary has assessed by comparing the actual poverty rate grown by six percent per year in real terms, who with the poverty rate in the hypothetical retires at age 60, and lives until age 78. Assume counterfactual. Constructing the counterfactual for a moment that this man had not enrolled in would require behavioral assumptions in relation the pension program, but had nevertheless saved to labor force participation, savings or household 16 percent of his salary for a certain number of composition, spanning entire life cycles. Not years. Assume also that the savings accrued at everybody is likely to agree on such assumptions an annual rate of 4 percent, in real terms. In this though. And theoretical uncertainties would be hypothetical benchmark, when reaching age 60 compounded further by data limitations. It is the man withdraws the savings and buys an therefore safer not to pass a judgment as to annuity, to be paid to him for as long as he lives. whether the pension program introduced in Such deal would be actuarially fair, in the sense Vietnam in 1995 is generally progressive or that the present value of this man's savings regressive. On the other hand, it should be would be equal to the present value of his possible to assess whether the formulas of the annuity payments. A similar, benchmark pension program implicitly subsidize or penalize calculation could be done for a woman retiring participants with different characteristics. at age 55 and leaving until age 80. Now consider that the savings go into the Weak incentives compulsory pension program instead. In this case, computing the pension at retirement Whether pension benefits are generous or not involves calculating first the "average salary" to depends on the work history of each individual, which the formula applies. And such "average 101 SOCIAL PROTECTION salary" varies depending on whether the person is to assume that it is invested, yielding 4 works for the government or not. If not, the percent per year in real terms, the same as in the average salary is computed over all the years of actuarial benchmark. contribution. But if the person works for government, the average is computed only over The comparison between the estimated pension the last few years before retirement, from a benefits and the actuarial benchmark shows that minimum of five for those who started before the compulsory program of Vietnam creates 1995, to a maximum of ten for those who joined winners and losers (Figure 10.4). Government after 2005. To make salaries comparable over employees are clear winners, in the sense that time, they need to be adjusted for inflation. But their pension benefits are either comparable to here, again, there is a difference between those the actuarial benchmark or higher. Some times working for government and the rest, with past much higher. The findings also show that salaries adjusted by the minimum wage in the women do better than men. This is because former case and by consumer prices in the latter. their pension benefits accrue more rapidly after For the conceptual exercise, it is assumed that 15 years of contribution, but also because their longer life expectancy results in a lower inflation runs at 5 percent per year and real actuarial benchmark. On the other hand, non- wages grow by 6 percent per year. Beyond the government employees stand to lose except if specific figures, it is clear that the "average they choose to contribute for less than 30 years. salary" of someone working for government is much higher than that of a person with the same Different assumptions on the growth rate of real work history but out of government. Note also wages, or the annual rate of inflation, or the real that the pension program of Vietnam only pays rate of return on savings would yield different a lump sum to participants who retire with less patterns. Regardless, considerable gaps would than 20 years of contribution. To be able to remain between actual pension benefits and the compare its value to the actuarial benchmark, actuarial benchmark. And government this lump sum can be converted into an employees and women would most likely equivalent string of payments. One possibility emerge as net beneficiaries anyway. Figure 10.4: Pension Benefits and the Actuarial Benchmark Source: Based on Paulette Castel and Martin Rama (2005), updated so as to reflect the changes made to the draft Social Insurance Law by the National Assembly. Pension benefits include 3 percent contribution paid by VSS towards health insurance. 102 OLD-AGE PENSIONS The comparison between pension benefits and pensioners are mainly former government the actuarial benchmark is relevant to assess employees and SOE workers whose benefits the prospects of extending the coverage of the have been grandfathered. Resources to pay for program. The expansion cannot come from the their benefits come directly from the budget, so government sector, as it is already fully that the actual expenditures of VSS are modest. covered. As for private sector workers, they will be less likely to enroll if the pension The reforms introduced by the Social program does not compare well with saving on Insurance Law of 2006 also ensure a good their own. Admittedly, the actuarial financial health for VSS in the longer term. benchmark would be difficult for them to The sustainability of the pension program match. The market for annuities is thin in hinges on the way the "average salary" of non- practice, so that self-insuring against the risk government employees is calculated. To allow of living longer than resources would be comparisons over time, past salaries are difficult. But it is still distressing to realize inflated by consumer prices. But in a country that the pension program of Vietnam is with the growth prospects of Vietnam, actual unlikely to be appealing for private sector salaries will almost certainly increase much workers who have less than 20 years left faster than consumer prices. Adjusting past before reaching retirement age. And it is also salaries by inflation thus results in the "average clear that the program discourages salary" falling increasingly behind the salary at participation beyond 30 years of contribution. retirement. This is not the case for government employees, whose "average salary" is Managing resources computed only over the last few years of service, with past figures adjusted by the minimum wage, not by consumer prices. But The social insurance programs of Vietnam most of the growth in coverage of the pension seem to enjoy an enviable financial health so program will come from the non-government far. VSS is responsible for managing the sector. Therefore, it is the low pensions of resources of these programs. In recent years its non-government employees that will ensure revenue has far exceeded its expenditures the financial viability of the pension program. (Table 10.1). In 2006 the surplus was 36 trillion dong, or roughly the equivalent of 3.7 The volume of reserves of the pension program percent of GDP. However, this robust will accumulate in the long term will crucially performance should not come as a surprise. depends on how pensions are indexed to The population of Vietnam is very young, so inflation. The Social Insurance Law states that that there is barely one pensioner for every 15 the adjustment will be based on consumer contributors to the system. Moreover, prices and economic growth. But how exactly Table 10:1: Social Insurance Revenue and Spending Year Long-term benefits Short-term benefits Total Old-age Lump Survivor Sickness Mater- Industrial Total revenue pensions sums benefits invalidity nity injury spending Balance 2002 26.7 7.0 0.4 0.3 0.2 0.3 0.1 8.2 18.5 2003 33.7 10.2 0.4 0.4 0.2 0.5 0.1 11.7 22.0 2004 42.6 11.2 0.5 0.4 0.2 0.6 0.1 13.1 29.5 2005 51.6 15.3 0.9 0.4 0.2 0.7 0.1 17.6 33.9 2006 60.7 21.4 1.2 0.6 0.3 1.1 0.1 24.7 36.0 Source: Based on data from VSS. Figures are in trillion dong. 103 SOCIAL PROTECTION Figure 10.5: The Reserves of the Pension Program Source: Based on Paulette Castel, and Martin Rama (2005), updated so as to reflect the changes made to the draft Social Insurance Law by the National Assembly. The real return on investments is assumed to be 4 percent per year. this rule is interpreted can make a difference. health of the pension program does not seem so It could mean that the reference is the growth enviable. By one estimate, the administrative in consumer prices, provided that economic costs of VSS amount to roughly 4 percent of circumstances make it affordable. Or it could total contributions. And yet, 6 percent is be read as saying that pensions should not lose retained, which raises questions as to how the value compared to the salaries of those at remaining 2 percent is being used. More work. In the first case, pensions would be importantly, the reserves of the pension adjusted by the inflation rate; in the second program are yielding a return below inflation. they could follow the minimum wage (a choice Investments have concentrated on safe sources that MOF seems to favor at present). For a with very low rates of return. These are decade or so, this choice would only affect the predominantly in the public sector. According budget transfer the government needs to make to one study, more than half of the portfolio is to VSS to pay for the pensions of former in bank deposits in state-owned commercial government employees and SOE workers. But banks, a third goes to support development the implications of this choice become projects and the rest goes into government noticeable as non-government workers start to bonds and treasury bills. The figures in this retire, in a decade or so (Figure 10.5). study are most probably outdated by now. But it is unlikely that returns have increased much There are other aspects in which the financial since then. 104 11. INSURANCE SCHEMES Market mechanisms can be very effective at approach. In between, government leadership diversifying risk and cushioning shocks. can be combined with private delivery. Financial products such as life insurance, Agricultural insurance is a case in point. savings products for old age or agricultural Reimbursement obligations on commercial insurance redistribute resources from "lucky" loans by farmers could be linked to an objective individuals and households who are not affected indicator (such as upstream water levels), based by shocks toward the "unlucky" ones who are. on a classification of land plots by risk area. But From this perspective they can be seen as part of such products are unlikely to emerge without an the broader social protection system. Since initial government investment. In extreme most of the poor and vulnerable already actively cases, insurance needs to be provided by the participate in markets, as entrepreneurs, government itself. Natural disasters fall in this employees or consumers, in principle at least category. Exposure to multiple hazards, which is they could also benefit from the protection likely to increase as a result of climate change, provided by such financial products. However, requires stronger coordination arrangements the markets for these products also suffer from between all the ministries and agencies imperfections that make government involved. intervention necessary. Without it, the share of the population covered could be thin and Insurance products participants would be exposed to providers defaulting on their obligations. For markets to Vietnam's insurance system has experienced work for those who need protection the most, healthy growth. The sector's liberalization took they must first be able to work at all. In place in 1993 when entry by multiple practice, the required degree of government stakeholders was first authorized. Life intervention varies considerably across insurance started in 1996 and began to grow segments, from industry regulation to active from 1999 onward, when large foreign insurers monitoring to market organization. At one end, joined the market. Reform accelerated from well-performing insurers can provide an 2003 when Bao Viet, a state-owned company efficiency benchmark against which to assess established in 1965 originally to provide only the management of government-run social import-export cargo and shipping insurance, insurance mechanisms. For instance, returns on was restructured into a financial group with life the assets of private pension funds (net of and non-life insurance services. Bao Viet and commissions and fees) are an obvious other state-owned insurers have since then been comparator for the returns of the assets of public equitized into joint stock companies, whereas pension programs. At the other end, there is a private joint stock insurance companies and need to design incentives so that the poor and brokers have been established. other groups who would tend to be excluded stand a chance to participate in those markets. By the end 2006, there were 37 firms of Access to life and disability insurance through diversified legal form and scale operating in micro-credit schemes is an illustration of this the life and non-life insurance, reinsurance, 105 SOCIAL PROTECTION and insurance brokerage sectors (Figure 11.1). Malaysia and nearly 10 percent in industrial Of these, three were fully state-owned, 16 were countries. In Vietnam today, more than 90 joint stock companies, four were joint ventures percent of the population does not have any and the remaining 15 were fully private. Such formal insurance policy. diversification of ownership forms shows how much the sector has matured in the decade For many years the growth of the sector rested since life insurance started in Vietnam. on life insurance, which was opened for foreign investment earlier on and is by now dominated Premium income for the insurance sector was by foreign firms. The products offered number close to 15 trillion dong in 2006, or about 1.5 around 100. Among them, endowment percent of GDP. This represents a growth of insurance has been the dominant product, with 9.6 percent over the previous year. About 57 an 80 percent share in new business premiums. percent of that income was accounted for by The sector is offering increasing benefits, life insurance. Funds invested by the insurance combining insurance protection with savings industry amounted to almost 31 trillion dong, or and investment elements, thus attracting larger 3 percent of GDP. More than half of the total size contracts on average. Yet, the number of was invested in government bonds, close to a new life policies sold decreased in 2006, and so quarter was in deposits, and roughly a tenth has the number of life insurance agents. each in equity shares and trust investments. Financial performance was strong, thanks to To some extent, the healthy growth observed in enhanced risk assessment and loss prevention life insurance for many years reflected the capacity. Insurers have also performed well in scarcity of long-term saving instruments in terms of gross and net claim ratios. Vietnam. By now, however, life insurers are confronted with heated competition from the However, the overall insurance penetration rate thriving banking sector and the stock market. is lower than elsewhere in the region. By way of comparison, premium income reaches 3.5 The non-life insurance sector, which was until percent of GDP in Thailand, 5.4 percent in recently dominated by domestic firms, started Figure 11.1: A Steady-Growing Sector Source: Vietnam Association of Insurers. 106 INSURANCE SCHEMES growing rapidly since accession to WTO. This However, a few commercially-oriented means that protection of businesses and mechanisms have emerged to provide social households against risks of accidents, business protection to low-income populations at local events and natural catastrophes started levels. Two broad schemes, namely micro- improving gradually. Non-life insurers are insurance groups and credit unions, deserve offering about 700 products. Vietnam's WTO especial attention. commitments include allowing participation by foreign investors in the lucrative statutory Several micro-insurance products are already insurance business from 2008 onwards. Among available in the case of Vietnam. Overall, the the areas to be opened up, motor insurance sector is growing rapidly, although from a low accounts for the largest share, followed by base. Microfinance groups and mass property and casualty insurance. organizations have developed programs such as aid funds, hunger and calamity reserves and Products more closely related to social cash allowances for war invalids and merit protection remain marginal. For instance, there citizens. Other groups have introduced the were only 1.4 thousand policies for annuities. write-off of outstanding debts in the event of Even on life insurance, there are half a million death. Sometimes the household of the insured policies active at present, which is the person also receives a funeral grant. equivalent of roughly one for every 40 households. Commercial health insurance looks Some commercially-oriented companies offer more promising. Local firms such as Bao Viet, micro-insurance products as well. Bao Viet Bao Minh and Pjico are implementing schemes proposes products requiring low premiums; as focusing on school children. However, no study low as 15 thousand dong per person per month has yet assessed the coverage, benefit package for people under 31 years of age. This is for a and amount reimbursed of such schemes. policy with a value of five million dong in the Another area with growth potential is private event of death. Mass organizations (in pensions (Box 11.1). particular the Women's Union) and sometimes microfinance institutions are used as brokers. Insurance for the poor There is less information in relation to health insurance funds. The experience in Hoc Son, Commercially-oriented insurance companies Hanoi, indicates the presence of serious adverse rarely offer products that are attractive to the poor. selection problems. Participants are primarily The typical sources of market failure in insurance pregnant, old, or in bad health. There are also markets, related to asymmetric information some signs of abuse, such as healthy people between policy holders and insurance managers, getting medicines for other, noninsured are exacerbated in their case. High administrative members of their family. costs usually lead to sizeable insurance premiums that poor households cannot afford. Premiums are Information is also limited in relation to anyway bound to be higher than for the emergency loans. A 2004 survey of insurance population as large, as the poor have shorter life and micro-insurance products found that interest-based lenders, pawnshops, and two expectancies, higher disability rates and a higher savings-and-loans funds of the Women's Union incidence of disability. Moreover, the relative (in Nghe An and HCMC) were providing complexity of premiums, benefits and contingent emergency loan services. Besides these cases, a payments is often difficult to grasp for detailed study of households in the Duyen Hai individuals and households with limited financial District in Tra Vinh Province also indicates that knowledge. Thus, the typical insurance products it is common for poor households to borrow in offered by commercially-oriented companies kind. Rice is indeed obtained in small quantities only reach individuals and households with from friends and relatives, but small traders also relatively high incomes. This is also true in the lend rice through the shortage period leading up case of Viet Nam. to the harvest. 107 SOCIAL PROTECTION Box 11.1: Private Pension Programs The main risk with private pension funds is to see them defaulting on their obligations towards the insured. Few companies live for many decades. It might therefore be difficult for someone of working age today to trust one such company to administer his or her savings and faithfully pay a pension until death ensues. Private pension funds with defined contributions manage savings on behalf of policy holders up to the point of their retirement. In doing so, they function like an investment fund. Therefore, similar regulations should apply. In both cases, provisions are needed against excessive borrowing. If by design a defined contribution fund does not have debt, it cannot go bankrupt, except for the case of fraud. As for the risk of fraud, it should be mitigated by appropriate regulation and supervision to ensure sound business conducts and professional ethics. Yet, even with limited borrowing and in the absence of fraud there might be cases of poor performance. For instance due to recessions, or to poor management. An important policy issue is how much protection should be given to the insured against such risks. Private pension funds which provide a minimum guarantee have an element of "defined benefit," or entitlement, which creates liabilities and thus risk of bankruptcy. Prudential regulation is required to discourage excessive risk taking by pension funds providing such a guarantee. Such funds must have a strong risk management capability. Aside from a tight limit on borrowing, regulation often includes specific restrictions as to the type of assets that such funds are allowed to invest in. For instance, illiquid assets and unlisted securities tend to be banned. The behavior of pension fund managers should be closely monitored not only by the supervisor but also by policy holders. If a pension fund is legally organized as a company (or an investment company), policy holders would be the equivalent of shareholders. They should participate in the board of directors, having the power to hire or fire a fund manage and overseeing his or her performance of the pension fund manager. Similar arrangements should be made even when a pension fund takes a contractual form such as a business trust. Assets managed on behalf of policy holders should be kept in a competent and independent custodian. Execution of sale and purchase of assets often requires a broker. Regulation should require that such a broker not be affiliated with the pension fund manager. There should also be arrangements in place to ensure a high degree of transparency of business dealings between the fund manager and the broker. Those arrangements are necessary to avoid the appropriation of assets through business practices such as "churning", "front running" or "self- dealing". The securities market regulator should have adequate competency to police such activities in the market. Regulation should promote competition among pension fund managers. To ensure that, pensioners should be provided with good portability so that they can choose and move to a well performing pension manager. This requires a competent individual account management capability by the pension fund manager and its custodian. The regulation should require that they have appropriate electronic data processing. The performance of pension funds should also be made easily and frequently observable. In the case of investment funds, they are typically required to publicize their net asset value (NAV) daily. A similar requirement should be applied to pension funds. In a similar spirit to the case of open-end investment funds, pensioners should be provided with possibilities to liquidate their contributions without a prohibitive penalty. At retirement, a pensioner should have an option of receiving a lump sum or reinvesting it in an annuity or getting a mixture of both. Pension fund managers could continue to manage the annuity if they have appropriate expertise to do so. In many countries, however, life insurance companies manage annuities because they specialize in actuarial business. One of the government's larger-scale credit expenditures with no collateral required. programs, VBSP, targets its lending towards Borrowers must instead be certified by the low-income groups and includes emergency People's Committee of their commune and be loans among its products. Households can part of a credit and savings group. These two borrow up to 10 million dong to cover living requirements aim at reducing the risk of default, 108 INSURANCE SCHEMES but they also considerably increase the time it selection. The experience of Vietnam has been takes to get a loan. As a result of this delay, similar to that of many other countries in this VBSP processes few emergency loans in respect. When Bao Viet experimented with practice. agricultural insurance in the 1980s and 1990s it soon learned that it was extremely difficult to Besides administrate delays and a sometimes control administrative costs due to the small size limited local outreach, the low provision of of the average farm. Since then, the instrument emergency loans by VBSP can also be linked to has not really picked up. its reliance on commune authorities and local mass organizations to seek out the eligible poor Yet, focus groups discussions conducted in households and organize lending groups. Local Dong Thap province in 2007 suggest that Women's Union representatives have raised flooding remains a major concern. The most concerns about the high costs for them to run severe shocks are linked to early flooding, with these activities properly, and have complained numerous participants referring to a particularly that these costs have not been fully factored in. devastating episode in 2000. Back then, not Local authorities also express reluctance to only did many of them lose their crops and incur endorse emergency loans, which are considered additional debt, but some also ultimately lost a high risk because they are not related to any land rights when they were unable to pay off productive investment. their debts. Some of them have responded to early flooding by advancing the harvest. While Supporting micro-insurance products and this approach reduces costs, it also reduces emergency loans requires specific financial benefits due to the poorer quality of rice when it expertise. In the case of micro-insurance, to is harvested early on. determine the level of premiums it is necessary to estimate the frequency of the risks that are The feedback obtained from these focus groups being insured and the magnitude of the losses suggests that farmers do appreciate the idea of associated with bad shocks. Also, to profitably linking insurance to commercial loans. They invest and manage the resources collected, appear to value the lower transaction costs, and assess and process the claims, and rapidly thus lower administrative costs, that would disburse the benefits, some financial expertise accompany this type of contract. And they and administrative capacity are required. But understand the rationale for using the level of such expertise is seldom available at local water in the river as an objective indicator to levels. This explains why most providers of decide on repayment. Not everybody is micro-insurance have turned to commercial supportive, however. Some of the participants insurance companies for the management of think that recent improvements in infrastructure their insurance schemes. This is the so-called would result in less flooding than 2000, even if partner-agent model where the formal insurance mapping work suggests otherwise. Some company assumes the financial risks and the farmers also believe that the government will microfinance institution serves as a forgive debts, despite recent policy changes "matchmaker" to provide lower-cost links calling for the halt of this practice. between the parties. In light of Vietnam's past experience with Agricultural insurance agricultural insurance and also based on this study of Dong Thap province, solutions have been proposed to mitigate the traditional With a majority of the poor living in rural areas, problems of high administrative cost, adverse and their livelihoods depending so much on selection, and moral hazard. Those solutions crop success, agricultural insurance is a hinge on flood-index insurance offered over the potentially important social protection short period of time (June 20 to July 10) when instrument. However, two common programs early flooding interferes with the rice harvest. plague agricultural insurance programs around the world, namely moral hazard and adverse Three insurance products can be considered. 109 SOCIAL PROTECTION One possibility is to index the level of water the province, district and commune upstream (at the Tan Chau hydro- administration. They also include the National meteorological station) and to provide Committee for Search and Rescue, which indemnity payments directly to the dominant reports to the Prime Minister, and the rural lending agency, the Vietnam Bank for Committee for Distribution of Goods and Agriculture and Rural Development (VBARD). Money, with participation of the Fatherland A second possibility is to provide farm-level Front and MOF. Other institutions are also indemnity. In this case, assessment costs can be associated with the effort to reduce and manage kept under control by classifying farm plots into risks from natural disasters are the National flood risk zones and using remote sensing Hydro-meteorological Center and GSO. satellite technology to measure the water level. A third possibility is to pass on the benefit Public funding is provided under different directly to rice producers based on flood risk appropriations at the level of Central mapping, instead of adjusting the benefit based Government. Investments to prevent and on actual flood levels. mitigate natural hazards include structures such as dykes and resources such as hydro- Natural disasters meteorological forecasting capacity. These investments are financed by the concerned Vietnam has a long history of striving to ministries as part of their regular programs. manage and reduce the damage risk form Disaster response activities and post-disaster natural hazards. This history goes back over relief are funded by MOLISA, which tops up centuries, ever since floods were seen as one of the resources mobilized by the affected the "four biggest dangers to mankind", together provinces themselves. with fire, robbers and invaders. By 1248, the These institutional and funding arrangements Red River dyke system had already been have served Vietnam well in the past and have established. At present, a system of river resulted in considerable achievements. Yet the embankments and sea dykes spans thousands of performance of the current disaster risk kilometers. There is also increasing management systems seems to be falling recognition that climate change can exacerbate behind. natural disasters, leaving large numbers of people in need of emergency support and In the absence of a broader consensus on how to pushing many households back into poverty. jointly address exposure to multiple hazards, individual ministries pursue prevention and The government has succeeded remarkably well mitigation investments in isolation. They often in reducing the direct impacts of extreme emphasize disconnected structural measures, climatic events. To manage the disaster cycle, Vietnam has structures and mechanisms that thus forgoing important synergies. Multi- provide prior mitigation, response during the hazard exposure requires the involvement of disaster, and subsequent recovery and multiple ministries and agencies. While there reconstruction. Much of the effort is undertaken seems to be no single best organizational by the provinces and communes where disasters structure for that purpose, it is widely accepted occur, in the name of the subsidiarity principle. that a strong central agency is key in a disaster But there is also recognition that current management system. In Vietnam, the approaches, largely geared towards handling preparation, approval and implementation of floods and storms, need to be reconsidered. public investments that would make provinces more resilient to natural disasters are At present, key institutions to manage and cumbersome, with little sense of urgency on the reduce natural disasters include the inter- part of concerned authorities. These ministerial Central Committee for Flood and investments are often not integrated into a Storm Control, which is replicated at the level of broader national strategy and plan. 110 INSURANCE SCHEMES In addition to structural investments, for construction codes with the aim of disaster- instance in river embankments, drainage canals proofing public infrastructure and private or safe harbors, much can be done at relatively property would help increase resilience to little cost. Introducing disaster risk disasters. management at the community level, through low-cost "Safer Commune" or "Safer Village" Faster implementation of disaster relief and plans, is a promising approach. Another avenue recovery is often hampered by the absence of is to undertake flood modeling, which allows to credible damage and loss assessments. Better better plan the effect of possible infrastructure assessments would provide for better decision investments on flood and water flows. Land use making in terms of resources needed and planning is also key to minimize economic targeting. Credible assessments are also activities taking place on marginal sites instrumental in mobilizing international funding vulnerable to disasters. Similarly, revisiting in the case of larger damage events. 111 PART III: THE WAY FORWARD 12. AN INTEGRATED APPROACH As poverty continues to fall, new forms of generation household including grand-parents vulnerability emerge. The rapid ageing of the and works in agriculture or does informal population, massive rural-urban migration, the activities, such as being a casual laborer or growing importance of non-communicable helping in a household business. Because of a diseases, fluctuations originating in world rapid demographic transition, in a few decades markets, natural disasters and climate change population will be much older on average. And confront Vietnam with unprecedented because of rapid rural-urban migration, it will challenges. But the response to those have moved too. In a couple of decades this challenges cannot be simply to mimic the social average citizen may have health concerns, will protection models of industrial countries. most probably have migrated to a city, could be Poverty still remains an issue. It remains deep part of a nuclear household including no grand- for many in the ethnic minorities; it may be children, will most probably be part of the changing its face in urban areas. Just dealing formal economy as a wage employee or a with risks will not be enough, as some groups businessperson, and will not be so far from are not doing well even in good times. Instead, retirement. it is better to think of social protection as a combination of policies and programs across How well will the institutional arrangements of three areas: opportunity to thrive, support for Vietnam support this transformation? How can those left behind and insurance to cope with they contribute to making the personal journeys shocks. Much has been accomplished by of millions of citizens become stories of Vietnam in all three. As it joins the ranks of prosperity and accomplishment, not of hardship middle-income countries, entering a period of and uncertainty? Jobs will be needed for people fundamental economic and social to move out of agriculture and informal sector transformation, more will be needed to ensure activities. Local development will be required that growth remains inclusive. The key to to boost the living standards of those who do not success may be in a better integration of policies migrate. Social services should be accessible for those who do, and may need them away and programs across all three areas. While from home. Health care should be affordable specific policy recommendations may be and of good quality. Instruments should be offered in each of them, they can be brought available for people to save towards their old together under the form of four main priorities: age. And protection mechanisms should exist to extending coverage, strengthening incentives, cope with multiple risks along the way, from improving information and managing resources. natural disasters to anti-dumping actions by trade partners to catastrophic health shocks. A reform agenda Targeted programs aimed at the most vulnerable Vietnam is at the dawn of an unprecedented will not be enough. Targeted programs were economic and social transformation. The never a critical component of the government average citizen is a mostly healthy 26 years old strategy anyway. They aimed at helping the who lives in a rural area, belongs to a multi- neediest households cope with the costs of 115 SOCIAL PROTECTION social services and to gain access to credit. booming economic hubs may start to take a They also supported the most disadvantaged toll on those whose means are barely above communes to integrate with the market the poverty line. In rural areas, the plight of economy. But they never represented a major ethnic minorities deserves special attention. budgetary outlay, and only reached a fraction of Many in this population group are bound to the poor. The strategy of the government has "cross" the poverty line in the coming years, relied more on economic growth, the which will result in continued poverty redistribution of land and reasonably good reduction overall. But they will still be barely health and education. And when it comes to above the line for the foreseeable future. And social policies, the stated ambition of the meanwhile, many among the ethnic minorities government is to build universal programs. will still be suffering from deep poverty. A From this perspective, targeted programs can policy agenda focused on managing risks and become important in supporting the shocks only would fail to deliver to these participation of the poor and the vulnerable in groups, as they experience hardship even in mainstream social programs. But mere social the best of times. assistance will not do. A common vision on what needs to be done However, the policy agenda of a country at seems to exist among policy makers, academics Vietnam's development level cannot be to and the population at large. It is a vision of mimic the social protection systems of shared and inclusive growth, in the socialist industrial countries either. Poverty remains a spirit of Vietnam. Based on the success of Doi serious issue. It continues to decline rapidly Moi, it involves relying on market mechanisms in rural areas, but for the first time it seems to to the extent possible, to ensure an efficient have stagnated, perhaps even increased, in allocation of resources in the short term and urban areas. The higher cost of living in sustain innovation and growth in the longer Figure 12.1: The Cost of Subsidizing Participation in the Pension Program Source: Paulette Castel (2007b). 116 AN INTEGRATED APPROACH term. But it also entails mobilizing government health insurance program is to remain resources to finance the participation of the poor financially viable. and other vulnerable groups. And it requires market regulation and policy stewardship, to Something similar could be done in the case of ensure that everybody has access to services and pensions, where a minimum contribution that quality standards are met. towards retirement and some basic benefits could be paid by government to the social But the specifics remain unclear, often resulting insurance agency on behalf of each poor in frustration in the daily implementation of household head. A necessary complement to social policies and programs. One way to this approach is the partial subsidization of articulate the reform agenda is under the form of participation by the near-poor. Without it, the four policy priorities: extending coverage, incentive for households to escape poverty strengthening incentives, improving information could be reduced and enrollment could decline and managing resources. among those who do escape. Extending coverage One important question is whether a strategy involving the subsidization of a large fraction of While the vision underlying the social the population can be afforded. The answer is protection agenda is one of shared growth and probably yes. To begin with, part of the inclusion, Vietnam is quite far from anything subsidization simply entails shifting current resembling universal coverage for its social government funding from providers (say, public programs. In the case of health insurance, the hospitals) to purchasers of services (the social initial target was to get there by 2010. By now security agency, or a dedicated fund for chronic close to half of the population is covered, which illnesses and life-threatening diseases). In the is certainly an accomplishment for a country at case of pensions, the alternative to the subsidy is Viietnam's development level. But subsequent not savings: it is a contingent liability, as gains could prove increasingly difficult. something will need to be done sooner or later Coverage is even more limited in relation to to support old people without income. The truly pensions, raising the prospect of large liabilities additional spending can be contained as long as when the young of today reach retirement age, if policies are in place to support a rapid reduction budgetary resources need to be mobilized to pay in poverty. for non-contributory pensions. Targeted This is show by a recent study which assessed programs do not reach all of their intended the cost of three, potentially complementary beneficiaries either; far from that. avenues to extend coverage in the case of old- Implementing the vision will require bolder age pensions. The first option is to pay in full steps. The full subsidization of participation by the minimum contribution towards the the poor in mainstream programs, already voluntary pension program for one adult per adopted in the case of health insurance, opens a poor household. The second one is to partially promising avenue. But more budgetary subsidize the participation of one adult per near- resources may be needed as the health insurance poor household in the voluntary pension premium increases to match real health care program. Based on a previous study on the costs. The partial subsidization of the near poor, willingness to pay for old-age pensions, the to be undertaken from 2008 onwards, should subsidy was set in this option at 40 percent of also contribute to the expansion of coverage. As the minimum contribution. Finally, the third for those who will not benefit from subsidies, option is to help the same two groups reach a there is a need to contain the adverse selection. minimum pension at retirement if their It could be tempting to expand coverage by contributions are not sufficient. In this third simply lowering requirements to enroll. case, the subsidy is equivalent to the gap However, mechanisms such as group insurance between the pension or lump-sum they would be or household enrollment will be necessary if the entitled to and half of the minimum wage. 117 SOCIAL PROTECTION Taking household sizes into consideration, the consideration the costs of extending the study estimates the cost of financing the coverage of social protection programs. participation of 1.9 million poor and 2 million near poor today, and the partial subsidization of As for the compulsory health insurance and old- pensions to one million poor and near poor as age pension programs, the extension of they reach retirement age. The simulations are coverage hinges on the gradual formalization of based on the assumption that the minimum the economy. In this respect, the main priority contribution and the pension levels grow at the is to avoid penalizing job creation through same rate as income per capita. The results excessively high contribution rates. Labor costs show that these options are affordable (Figure are already bound to increase rapidly, as a result 12.1). The analysis may need further of the rapid minimum age increases necessary to refinement before reaching a firm conclusion. comply with WTO commitments. The Social But these results are plausible, for two reasons. Insurance Law, by foreseeing an increase in The first one is that the voluntary pension contributions towards old-age pension, and program sets the minimum contribution at a possibly towards unemployment insurance, may relatively low level, which also results in add to this burden. In the same spirit, balancing relatively low pensions. The second is the the books of the health insurance program expected decline in poverty rates, which should through a higher contribution rate would not be reduce the need for subsidization over time. an advisable move. For the subsidization approach to work in Strengthening incentives practice, two complementary policies need to be considered. One of them is to use targeted Increased reliance on market mechanisms, one programs more systematically as a vehicle for of the foundations of Doi Moi, can certainly the poor and the near poor to participate in lead to large efficiency gains. The improvement mainstream programs. This is the spirit of in the quality of care provided by hospitals Decision 139, but there are a range of other operating in an autonomous manner is a case in targeted programs whose effectiveness is less point. However, in sectors which are precisely clear. A consolidation of current program under characterized by market imperfections, the form of a narrower and better defined set of undesirable outcomes cannot be rated out. benefits and transfers, including health Hospitals may become good at catering to those insurance premiums, social security who can pay, but they can also neglect the contributions and possibly cash transfers could mandatory provision of free services to the poor. help in this respect. Improving the monitoring of health care providers is thus an important priority. The other complementary policy to consider refers to the allocation of resources to sub- The mechanisms used by the government to national levels. In a country as decentralized as regulate these sectors can be a source of Vietnam, the implementation of national distortions as well. Health insurance has the policies varies considerably depending on the merit of shifting the funding of health care from capacity and the resources of local the supply side (say, hospitals) to the demand governments. From this perspective, it is side (the insured and VSS). In doing so, it has important to ensure that provincial resources are the potential to foster competition among commensurate with provincial needs. In this providers and increase the overall efficiency of respect, developing a property tax providing the health sector. However, just adopting health resources to local governments would help. insurance is not enough to contain the escalation This move would reduce the reliance of local of costs or to ensure a better quality of service authorities on regressive fees. It is also (Box 12.1). The payment schedules used by the important to make sure that the budget social insurance agency to reimburse health care allocation norms used to redistribute resources providers can encourage excessive diagnosis to provincial levels and below take into and treatment, resulting in an unwarranted 118 AN INTEGRATED APPROACH Box 12.1: The Limits of Mandatory Social Health Insurance Like Vietnam, countries in the former Soviet block had tax-financed health systems in place when they began their transition to a market economy. Many experienced substantial reductions in government revenues along the way, and many looked to mandatory social health insurance (SHI) as a way to raise additional revenues for the health sector. SHI was also seen as an opportunity to move away from paying health facilities through budgets, and instead pay them according to the volume of care delivered, sometimes through a competitive contracting process. In the end, roughly half of the countries in the former soviet block adopted SHI. Some opted for a hybrid model, creating a SHI agency and adopting provider payment reforms, but financing the agency through income taxes or general revenues. A recent study finds that the countries that did adopt SHI experienced increases in government health spending, controlling for other contemporaneous factors, and also saw hospital activity increase somewhat. The transitions did not, however, bring a measurable reduction in mortality or disease incidence, despite the higher levels of health spending. Several of those countries are now busy trying to introduce stronger incentives in the system to ensure that their higher spending levels translate into better health outcomes. All countries that adopted SHI have explicitly maintained the commitment to universal coverage, by using tax revenues to finance "contributions" for people outside the formal sector. In practice, this has worked better in some countries than in others. Fiscally decentralized countries like Russia appear to have faced more difficulties. It does not appear that subsidized participants are denied care or receive worse quality care. But there is a shortfall of revenue for the health system as a whole. China's recent experience is also of potential relevance to Vietnam. In 2003, the government began rolling out a voluntary health insurance program in rural areas following the almost total collapse of the old commune-based cooperative medical system. The new scheme is heavily subsidized for everyone, with government picking up about 65 percent of its total cost. In the poorest provinces, contributions are paid entirely by government. In part because of the very generous subsidies, enrollment has been very high. The program appears to have increased utilization, but seems to have had only modest impacts on the risk of large out-of-pocket payments. This partly reflects a more intensive utilization of services, which is a positive outcome. But the impact of insurance on out-of-pocket payments per contact has been minimal. Because the budget for the program is small relative to total health spending in China, coinsurance rates are high, the list of exclusions extensive, deductibles sizeable, and reimbursement ceilings fairly low. All of which explains the high out-of-pocket payments. But even in urban areas, people with insurance face more costly care partly because they end up being treated at higher levels in the system. It is not clear that this extra expense is warranted on medical grounds. There is certainly evidence in China of extensive "supplier-induced demand". It stems in part from the emphasis on fee-for-service and the fact that facilities have almost complete autonomy. China is now trying to deepen coverage. But the additional resources will come from more generous government subsidies, not from higher household contributions. It is also trying to introduce provider payment reforms and other measures that will allow the insurer to put downward pressure on costs and upward pressure on quality. escalation of health care costs. This is certainly women get more, out of the same stream of an area deserving attention by authorities. contributions, than private sector workers and men do. After 20 years of contributions, the The benefit formulas used for the old-age incentive to remain enrolled starts declining pensions program affect the incentives to save and it is in the interest of participants to under- and to participate in the system. Pensions are declare their earnings. Parametric changes quite far from what an actuarially fair system should be made to the benefit formulas to would provide. Government employees and bring them closer to the actuarial benchmark, 119 SOCIAL PROTECTION so that those who contribute more also get information, insurers may simply not be more. In practice this can be accomplished by interested. In other areas, such as agricultural gradually moving towards a system of extension, finding the right incentives may "notional" accounts, which mimic through involve a much deeper understanding of accounting what could be accomplished by cultural differences between ethnic minorities saving at a given interest rate and buying an and the majority group. The current offer of annuity at retirement. services is based on models which have worked Moving closer to the actuarial benchmark well in the lowlands but may be of little would also help introduce flexibility into the relevance for swidden cultivation on sloping retirement age. At present, mandatory ages are plots of land. too far from the work patterns of the elderly in Vietnam. Many of them would prefer to Improving information continue contributing for a longer time and enjoy a higher pension subsequently. Low The vision of an integrated approach to social retirement ages may also dissuade many people protection is at odds with the current of working age from joining the voluntary fragmentation of information systems (Figure pension program. Under present rules, few men 12.2). Targeted programs use the classification aged 40 or more and women aged 35 or more of households by poverty status, conducted by could be interested in participating. Because local officials of MOLISA. Cash transfers for they would know for sure that they can only vulnerable groups, while also managed by aspire to a lump sum payment, not to a pension. MOLISA, are handled separately. Health And it is the assurance of a flow of revenue in insurance and old-pension programs run on old age that people value most. their own information systems, focused on individuals rather than households. All of these The emergence of private pension funds will programs rely on the place where the household require that attention be paid to the portability of or the person lives, or is supposed to live, benefits across programs. As Vietnam grows through a system of residential registration richer, many may want to consider topping up which often restricts access to benefits in other their publicly run pension by saving through a localities. Such system seems ill-adapted to a private pension program. The more country where a large fraction of the population complementary the two systems are, the will be moving from rural to urban areas in the stronger the incentive to combine them. coming years. Eventually, for some segments of the population the voluntary pension program will have to The fragmentation of information systems is compete with private pension providers. This such that the ministries in charge of formulating can be a healthy development, whose potential policies cannot really monitor the behavior of would be amplified if savings could be the populations they are supposed to serve. For transferred across programs. Equally important instance, MOLISA knows little about the typical is to regulate and closely supervise private jobs and earnings transitions of contributors pension programs, to reduce the risk of them towards old-age pensions. And MOH has defaulting on their obligations, through difficulty figuring out how health care services bankruptcy or fraud. are used by different participants in the health insurance program. Some of the most complex incentive issues might actually be faced in rural areas. Detailed information on the use of health care Sustaining the emergence of flood insurance facilities and services is obviously important to and other market-based tools to deal with adequately reimburse providers. At a more natural disasters will not be easy. Without a fundamental level, this information is critical to proper institutional set up, involving cadastral monitor the delivery of services, assess patterns maps, hazard modeling and verifiable of health care demand, monitor costs, and 120 AN INTEGRATED APPROACH Figure 12.1: A Fragmented Approach to Information Management design appropriate policies for the health sector. commercial banks are paying old-age pensions At present, however, this information does not using cards for automatic teller machines. Lang circulate from providers to policy makers or the Son is relying on sophisticated chip cards for social insurance agency. participants in the health insurance program. In itself, each of these initiatives bears much In the case of MOH, a central division on health promise. But the new systems are not information is the focal point for collecting and connected to a general architecture for analyzing health statistics for the sector. At the information management, which limits their provincial level, there are one or two persons in effectiveness. Further down the road, bringing charge of health care data. And at district levels, common standards into these scattered efforts health departments are responsible to process a may add complexity to an already difficult series of forms compiled at commune level. problem. But there are many such forms, and each NTHP has its own information system. The result is a The main recommendation in this area is to compilation of paper documents with little introduce a single social insurance number, to potential to feed into any usable database. be given to every person at birth regardless of coverage. The poverty status of the household VSS also has a health information system, the person belongs to can be updated on an which records inpatient stays and outpatient visits by the insured. Inspection units, based in annual basis. The social insurance number hospitals, are responsible for data entry. can serve as the platform to document Unfortunately, the impressive amount of personal histories in connection to information being compiled by provincial and employment, earnings and health care use, district hospitals does not seem to be processed among others. With adequate information further. management systems, these personal histories can help design better provider payment There are several ongoing attempts to introduce mechanisms in health, and eventually support modern information technology in parts of the the introduction of individual accounts in the social protection architecture. Some case of pensions. 121 SOCIAL PROTECTION Unlike current registration cards, a single Managing resources social insurance number would allow the full portability of benefits throughout the country. Delivering on the social protection agenda also From this perspective, it could make a requires solid financial management. Targeted contribution to labor mobility, helping programs involve transferring resources to migrants access social services regardless of thousands of communities and millions of location. A single social insurance number households. Extending the coverage of social could also limit the denial of service often programs will require some form of observed at present. Eventually, the social subsidization for the poor and perhaps the near insurance number would reduce the usefulness poor, with the budget covering the contributions of the household registration system, an of a significant portion of the population, or administrative approach which served well in supporting provinces to do so. The its recent history but is not well suited for a development of the health insurance program more dynamic and mobile economy. amounts to shifting an increasingly large share Box 12.2: Managing Public Pension Reserves Public pension reserves are managed poorly in many countries. Funds are often forced to invest in government bonds and housing loans at low nominal interest rates. In countries that suffered from high inflation, real investment returns were negative. Even in stable countries, returns on public pension reserves have often been below market. To address these problems, several industrial countries have in recent years revamped the governance structure and investment management of their public pension funds. Canada, Ireland, New Zealand and Norway are among them. Separate state entities with their own boards of directors have been created in three of these countries. The boards of directors are responsible for formulating the investment policies of the funds, setting the strategic asset allocations, and supervising management. They have adopted corporate governance and conflict of interest guidelines and have set up Audit Committees to ensure the effectiveness of internal control systems. They have appointed auditors and global custodians and adopted appropriate asset segregation and valuation rules. They have also made considerable use of external advisers on a wide variety of topics, ranging from advice on asset allocation strategies to the selection of external asset managers and the adoption of information systems. Public accountability of the funds is buttressed by regular independent reviews of their performance as well as special examinations that may be commissioned by the ministries of finance. All the funds are required to submit reports to the government and give evidence to parliamentary committees. All four funds have placed special emphasis on monitoring and controlling both financial (market) and operational risks. They have all separated investment decision making from back-office operations, including record keeping, settlement, and performance measurement. They have developed detailed control procedures. And they have all installed sophisticated systems to measure the performance of external asset managers. In three of the countries, the funds started by emphasizing passive management through external asset managers. However, this concept was relatively quickly expanded to allow active management, use of customized indexing to limit excessive exposure to index-dominating companies and allow investments in smaller companies. Internal management capabilities were developped, first in passive management and progressively also in active management. The approach was later expanded to encompass investments in private equity, real estate and infrastructure projects as well as emerging markets. All four pension funds have achieved positive investment results with excess returns over their respective benchmarks. However, equity returns were adversely affected by stock market bubbles. The best performer was New Zealand, partly due to its avoidance on unsettled capital markets. Source: Dimitri Vittas and others (2007). 122 AN INTEGRATED APPROACH of the funding for health care from public public hospitals to private clinics to NGO providers to wholesale purchasers of services. facilities to certified doctors. Scrutiny will be The pension program, in turn, is bound to required to avoid abuse or favoritism, as in accumulate a massive amount of reserves any other public procurement decision. during several decades. This is because the population is young and the labor market is An area deserving special attention is the formalizing, so that dependency ratios will management of the reserves of the old-age remain low. pension program. Those reserves already exceed the equivalent of three billion dollars. Transparent financial management will be But they are placed in bonds and liquid assets crucially important in all these areas. There yielding low returns. There is a need for will be a need to make sure that targeted strategic decisions as to what the authorized communes do receive the resources they are investments are, and how portfolio entitled to and manage them properly. It will reallocation decisions are going to be made. be necessary to monitor that those who are Transparent mechanisms could also be classified as poor get their social security considered to report to the National Assembly contributions totally or partially paid for out on the financial performance of social of public resources. The purchasing decisions protection programs, including the of the social insurance agency will involve management of reserves, perhaps on an annual hundreds of thousands of providers, from basis (Box 12.2). 123 BIBLIOGRAPHY The word "processed" describes Development Bank, Manila. informally reproduced works that may not be ADB (Asian Development Bank). 2006b. Social commonly available through libraries. Protection Index for Committed Poverty Reduction. ADB, Manila. ActionAid Vietnam and Institute of Economics. ADB (Asian Development Bank). 2006c. 2004. Listening to People Living in Supermarkets and the Poor in Vietnam. Poverty. Agriculture Publishing House, Asian Development Bank. Making Hanoi. Markets Work Better for the Poor, Adams, Susan J. 2005. Vietnam's Health Care Hanoi. System: a Macroeconomic Perspective. ADB (Asian Development Bank). 2007a. M4P IMF, Washington, DC. Week 2006: Proceedings of a series of ADB (Asian Development Bank). 2001. Social review and planning events held by Protection in Asia and the Pacific. ADB, Making Markets Work Better for the Manila. Poor (M4P) during the week 27 November to 1 December 2006. ADB, ADB (Asian Development Bank). 2005a. Labor Hanoi. Markets in Asia: Promoting Full, Productive, and Decent Employment. ADB (Asian Development Bank). 2007b. Rural Key Indicators 2005. ADB, Manila. Labor Markets and Migration. Making Markets Work Better for the Poor ADB (Asian Development Bank). 2005b. M4P Discussion Paper No. 18. ADB, Hanoi. Week 2005: Proceedings of a Series of Review and Planning Events Held By ADB (Asian Development Bank) and WHO Making Markets Work Better for the (World Health Organization). 2002. Poor (M4P) During the Week 31 October Improving Health Status and Reducing to 4 November 2005. ADB, Hanoi. Inequalities. ADB and WHO, Hanoi. ADB (Asian Development Bank). 2005c. Social Adger, W. Neil. 1998. Indicators of Social and Protection Index for Committed Poverty Economic Vulnerability to Climate Reduction: Multi-Country Report. Change in Vietnam. CSERGE Working Prepared By Halcrow China Limited and Paper GEC 98-02. Centre for Social and Institute of Development Studies, Economic Research on the Global Sussex. TA No. 6120 - REG. ADB, Environment. University of East Anglia, Manila. Norwich, UK. ADB (Asian Development Bank). 2006a. Asia Ahmadi, Nader. 2005. Situation Analysis of Water Watch 2015: Are Countries in Asia Institutional and Alternative Care on Track to Meet Target 10 of the Programmes in Viet Nam. MOLISA, Millennium Development Goals. Asian CIDA & UNICEF, Hanoi. 124 BIBLIOGRAPHY Aksoy, Ataman and Aylin Isik-Dikmelik. 2007. Baulch, Bob, Thi Kim Chuyen Truong, The Role of Services in Rural Income: Dominique Haughton and Jonathan the Case of Vietnam. World Bank Policy Haughton. 2002. Ethnic Minority Research Working Paper No. 4180. Development in Vietnam: a World Bank, Washington, DC. Socioeconomic Perspective. World Bank Alarcon-Caracuel, Manuel Ramon. 2006. Policy Research Working Paper no. Report on the Current Situation and the 2836. World Bank, Hanoi. Development and Improvement Bhaumik, Sumon Kumar, Saul Estrin and Klaus Perspectives of an Integrated System of E. Meyer. 2007. Determinants of Social Security in the Socialist Repubic Employment Growth at MNEs: of Vietnam. Spanish Agency for Evidence from Egypt, India, South International Cooperation, Hanoi. Africa and Vietnam. Comparative Processed. Economic Studies, 49(1):61-80. AusAID. 2006. Quang Ngai Natural Disaster Bijlmakers, Leon and others. 2006. Feasibility Mitigation Project: Lessons Learned. of the Programme Approach in the Hanoi. Processed. Health Sector in Vietnam. Hanoi. AusAID and UNDP (United Nations Bird, Richard Miller and N. E. Slack. 2004. Development Programme). 2005. Impact International Handbook of Land and of HIV/AIDS on Household Vulnerability Property Taxation. Edward Elgar, and Poverty in Viet Nam. UNDP, Hanoi. Cheltenham, UK. Axelson, Henrik. 2007. The Evolution of Health Brassard, Caroline. 2004. Wage and Labour Insurance Policy in Vietnam. Asian Regulation in Vietnam within the Development Bank, Hanoi. Processed. Poverty Reduction Agenda. Policy and Axelson, Henrik and others. 2007. Evaluating Society, 23(2):49-77. the Impact of a Targeted Program on Brassard, Caroline and Scott Fritzen. 2007. Healthcare Utilization and Expenditure Multi-Level Assessments for Better in Vietnam Using Quasi-Experimental Targeting of the Poor: a Conceptual Methods. Paper Presented at the 6th Framework. Progress in Development World Congress of IHEA, Copenhagen, Studies, 7(2):99-113. July 2007. Processed. Bui, Minh Dao. 2003. Mot So Van De Giam Bales, Sarah and others. 2007. The Early Impact Ngheo O Cac Dan Toc Thieu So Viet of Decision 139 in Vietnam: An Nam (Poverty Reduction Issues of Ethnic Application of Propensity Score Minorities in Vietnam). Social Science Matching. Report to the Ministry of Publishing House, Hanoi. Health with the support of Vietnam- Bui, Minh Dao and Thi Bich Lan Bui. 2005. Sweden Health Cooperation Programme Thuc Trang Doi Ngheo Va Mot So Giai and the Asian Development Bank. Phap Xoa Doi Giam Ngheo Doi Voi Cac Hanoi. Processed. Dan Toc Thieu So Tai Cho Tay Nguyen Baulch, Bob, Joe Wood and Axel Weber. 2006. [Poverty and Some Solutions for Hunger Developing a Social Protection Index for Eradication and Poverty Reduction for Asia. Development Policy Review, Ethnic Minorities in Tay Nguyen]. Nha 24(1):5-29. xuat ban Khoa hoc Xa hoi, Hanoi. Baulch, Bob, T. Hung Pham and Barry Reilly. Capuno, Joseph. 2005. Health Insurance for the 2007. Ethnicity and Household Welfare Poor: Lessons from the Philippines and in Vietnam: Empirical Evidence from Vietnam: Paper Presented at Forum 9, 1993 to 2004. Processed. Mumbai, India, 12-16 September 2005. 125 SOCIAL PROTECTION Processed. Ci, Chy Lii, Chien Thang Nguyen, Hing Ceng Carew-Reid, Jeremy and Lothar Linde. 2007. Vo and Hinh Thu Phan. Non-Farm Rapid Assessment of the Extent and Employment in Rural Vietnam After Doi Impact of Sea Level Rise in Vietnam. Moi: VASS - Doi Moi Review Project Climate Change Discussion Paper 1. With Support from UNDP and the ICEM - International Centre for Embassy of Sweden Research Paper No. Environmental Management, Brisbane, 5. Hanoi. Processed. Australia. Processed. CIEM (Central Institution for Economic Castel, Paulette. 2005. Viet Nam: Toward Management) and GDC (German Universal Social Protection: Private Development Cooperation). 2004. Mechanisms to Reach the Poor: Policy Decentralization: National Policies and Issues and Research Implications. Field Experiences on Decentralization of Making Markets Work Better for the Planning and Budgeting. CIEM & GDC, Poor Discussion Paper No. 13. ADB, Hanoi. Hanoi. Clarke, Simon. 2006. The Changing Character Castel, Paulette. 2007a. Nghe An Voluntary of Strikes in Vietnam. Post-Communist Pension Fund: Transition to the National Economies, 18(3):345-361. Scheme: Policy Options. Study Funded Committee for Ethnic Minorities and MPI by AECI - Spanish Cooperation in (Ministry of Planning and Investment). Vietnam. Processed. 2006a. Discussion Papers of Ministries: Castel, Paulette. 2007b. Old-age Security Documents to Be Distributed at the Medium and Short-term Policies: International Consultative Forum on Pension Coverage for the Poor Supporting the Socio-Economic Voluntary Pension System Coverage and Development in the Ethnic Minority and Social Assistance Programs. Study Mountainous Areas of Vietnam 2006- Funded by DFID Vietnam. Processed. 2010. Committee for Ethnic Minorities and Ministry of Planning and Castel, Paulette and Martin Rama. 2005. Investment, Hanoi. Comments on the Draft of the New Social Insurance Law. The World Bank Committee for Ethnic Minorities and MPI in Vietnam, Hanoi. Processed. (Ministry of Planning and Investment). 2006b. Discussion Papers of Provinces: CDI (Center for Development and Integration). Documents to Be Distributed at the 2006. Sinh Ke Thay the Voi Cac San International Consultative Forum on Pham Phi Go Cua Nhom Dan Toc Dao Supporting the Socio-Economic Xa Thanh Son, Huyen Son Dong, Tinh Development in the Ethnic Minority and Bac Giang (Replacing Livelihood, With Mountainous Areas of Vietnam 2006- Non-Timber Products, of Dao Minority 2010. Committee for Ethnic Minorities in Thanh Son Commune, Son Dong and Ministry of Planning and District, Bac Giang Province). CDI, Investment, Hanoi. Hanoi. Conway, Tim. 2000. Current Issues in Sector- Centre for Community Health Development. Wide Approaches for Health 2006. Impacts of the Policy for Development: Viet Nam Case Study. Healthcare Support for the Poor through WHO, Geneva. Evaluation of the Implementation of the Decision 139/QD-TTg in the Central Conway, Tim. 2004. Politics and the PRSP Highland. Asian Development Bank, Approach: Vietnam Case Study. ODI Hanoi. Processed. Working Paper 241. Overseas Development Institute (ODI), London. 126 BIBLIOGRAPHY Cuc Thong ke Thanh pho Ho Chi Minh. 2005. Market Economy, Held by the Institute Dieu tra dan so giua ky nam 2004 Thanh of Labor Science and Social Affairs with pho Ho Chi Minh. Cuc Thong ke Thanh the Support of the World Bank, Da Nang, pho Ho Chi Minh. March 9-10, 2007. Processed. Dang, Boi Huong, Sarah Bales and Khanh Dasgupta, Susmita and others. 2007. The Phuong Nguyen. 2006. Ensuring Health Impact of Sea Level Rise on Developing Care for the Rural Poor in Vietnam and Countries : A Comparative Analysis. China: a State or a Market Approach? World Bank Policy Research Working Medical Publishing House, Hanoi. Paper No. 4136. World Bank, Dang, Boi Huong and others. 2007. Rural Washington, DC. Health Care in Vietnam and China: Davidsen, Soren and Carrie Turk. 2005. The Conflict Between Market Reforms and Citizens' Report Card in Vietnam. World Social Need. International Journal of Bank, Hanoi. Health Services, 37(3):555-572. de Brauw, Alan and Tomoko Harigaya. 2007. Dang, Ngoc Quang and others. 2006. Su Tham Seasonal Migration and Improving Gia, Tinh Minh Bach, Tinh Trach Nhiem Living Standards in Vietnam. American Giai Trinh Trong Tien Trinh Ngan Sach Journal of Agricultural Economics, Xa: Truong Hop Tinh Tra Vinh, Viet Nam 89(2):430-447. (Participation, Transparency, de los Angeles-Bantista, Feny. 2004. Early Accountibility in the Commune Childhood Care and Education in South Budgeting Process). Oxfam GB, Oxfam - East Asia: Working for Access, Quality Hong Kong & Institute of Development and Inclusion in Thailand, the Mangement of Asia, Hanoi. Philippines and Viet Nam. UNESCO, Dang, Nguyen Anh. 2005. Viet Nam Internal Bangkok. Migration: Opportunities and Challenges de Silva, M. J., S. R. Huttly, T. Harpham and M. for Development. In DFID (Department G. Kenward. 2007. Social Capital and for International Development) and IOM Mental Health: a Comparative Analysis (International Organization for of Four Low Income Countries. Social Migration) (eds.), Migration, Science & Medicine, 64(1):5-20. Development and Poverty Reduction in Asia (pp. 149-170). IOM. Dinh, Thi Phuong Hoa and Merel Mattoush. 2006. A situational analysis of newborn Dang, Nguyen Anh, Bach Duong Le and Hai health and interventions in Vietnam: Van Nguyen. 2005. Youth Employment Towards the development of a newborn in Viet Nam: Characteristics, health action plan. MOH and Save the Determinants and Policy Responses. ILO Children USA, Hanoi. Employment Strategy Papers 2005/9. ILO, Geneva. Do, Quy-Toan and Dang Trung Le. 2007. Natural disaster risk management (I): Dao, Lan Huong and others. 2006. Socio- Household strategies. Processed. Economic Status Inequality and Major Causes of Death in Adults: a 5-Year Dufhues, Thomas. 2007. Accessing Rural Follow-Up Study in Rural Vietnam. Finance: the Rural Financial Market in Public Health, 120(6):497-504. Northern Vietnam. Ph.D. Thesis. Leibniz Institute of Agricultural Development in Dao, Quang Vinh. 2007. Profile of minimum Central and Eastern Europe (IAMO), wage earners and impact on minimum Halle. wage adjustment. Paper Presented at the Workshop on Minimum Wage Policy in Dufhues, Thomas, Ute Lemke and Isabel 127 SOCIAL PROTECTION Fischer. 2004. Constraints and Potential Case of Vietnam. Working Paper 28. of Livestock Insurance Schemes: A Case National University of Singapore, Study from Vietnam. Research in Singapore. Development Economics and Policy Fritzen, Scott. 2006. Vietnam's Health Discussion Paper No. 05/2004. Workforce in Transition: Problems, University of Hohenheim. Policies and Prospects. Hanoi. Duong, Bich Hanh. 2007. Impact of Health Processed. Insurance Policies on Health Seeking Fritzen, Scott A. 2007a. Legacies of Primary Decisions of Users and the Quality of Health Care in An Age of Health Sector Services in Vietnam. Processed. Reform: Vietnam's Commune Clinics in Duong, D. V., C. W. Binns and A. H. Lee. 2004. Transition. Social Science & Medicine, Utilization of Delivery Services at the 64(8):1611-1623. Primary Health Care Level in Rural Fritzen, Scott A. 2007b. Reorienting Health Vietnam. Social Science & Medicine, Ministry Roles in Transition Settings: 59(12):2585-2595. Capacity and Strategy Gaps. Health Duong, Huy Luong, Tang A Shenglan, Tuohong Policy, 83(1):73-83. Zhang and Margaret Whitehead. 2007. Fritzen, Scott and Caroline Brassard. 2005. Vietnam During Economic Transition: a Vietnam Inequality Report 2005: Tracer Study of Health Service Access Assessment and Policy Choices. Mekong and Affordability. International Journal Economics, Hanoi. of Health Services, 37(3):573-588. Gabriele, Alberto. 2006. Social Services Edmonds, Eric V. and Nina Pavcnik. 2006. Policies in a Developing Market Trade Liberalization and the Allocation Economy Oriented Towards Socialism: of Labor Between Households and the Case of Health System Reforms in Markets in a Poor Country. Journal of Vietnam. Review of International International Economics, 69(2):272- Political Economy, 13(2):258-289. 295. GDC (German Development Cooperation). Ekman, Björn, Henrik Axelson, Duc Anh Ha 2006. The Current Salary System in and Liem T. Nguyen. 2007. Use of Vietnam's Public Administration: Need Maternal Health Care Services and for Decentralized Approach. GDC, Ethnicity: a Cross-Sectional Analysis of Hanoi. Vietnam. Processed. Giang, Thanh Long. 2004. The Pension Scheme Evans, Martin and others. 2007a. How in Vietnam: Current Status and Progressive is Social Security in Viet Challenges in an Aging Society. Vietnam Nam? UNDP, Hanoi. Development Forum (VDF), Hanoi. Evans, Martin and others. 2007b. The Giang, Thanh Long. 2006. Pension Liabilities Relationship Between Old Age and and Generational Relations: the Case of Poverty in Viet Nam. UNDP, Hanoi. Vietnam. Oxford Institute of Ageing French Embassy. 2004. Economic Growth and (OIA)Working Paper No. 106. Oxford Poverty Reduction in Vietnam: Stuck in Institute of Ageing (OIA), London. First Gear?: a Case Study of the O Mon Glenday, Graham and Gangadhar P. Shukla. District in the Mekong Delta. French 2006. Revenue Forecasting Techniques Embassy, Hanoi. Processed. and Application. Processed. Fritzen, Scott. 2004. Strengthening Social Glewwe, Paul, Nisha Agrawal and David Dollar Protection in Low Income Countries: the (eds.). 2004. Economic Growth, Poverty, 128 BIBLIOGRAPHY and Household Welfare in Vietnam. (United Nations Population Fund). World Bank, Washington D.C. 2006b. The 2004 Vietnam Migration Government of Vietnam. 2004. Tax Survey: Migration and Health. GSO, Administration Reform Program 2005 - Hanoi. 2010. Hanoi. GSO (General Statistics Office) and UNFPA Government of Vietnam. 2005a. Prime Minister (United Nations Population Fund). Decision No. 170/2005/QD-TTg of July 2006c. The 2004 Vietnam Migration 8, 2005 Promulgating Poverty Lines for Survey: The Quality of Life of Migrants the 2006-2010 Period. in Vietnam. GSO, Hanoi. Government of Vietnam. 2005b. Vietnam Hanai, Kiyohito and Thi Minh Huyen Bach. Managing Public Expenditure for 2006. Revenue Assignment between the Poverty Reduction and Growth: Public Central and Local Budgets in Vietnam. Expenditure Review and Integrated In The Final Report of the Joint Fiduciary Assessment. Volume 1: Cross- Research Program on the Vietnamese Sectoral Issues. Financial Publishing Tax System (pp. 217-260). Tax Policy House, Hanoi. Department, Ministry of Finance, Vietnam and Policy Research Institute, Government of Vietnam. 2005c. Vietnam Ministry of Finance, Japan. Managing Public Expenditure for Poverty Reduction and Growth: Public Haughton, Jonathan, The Quan Nguyen and Expenditure Review and Integrated Hoang Bao Nguyen. 2006. Tax Fiduciary Assessment. Volume 2: Incidence in Vietnam. Asian Economic Sectoral Issues. Financial Publishing Journal, 20(2):217-239. House, Hanoi. Haughton, Jonathan and Viet Phuong Ngo. Government of Vietnam. 2005d. Viet Nam 2007. Tax Incidence in Vietnam. Suffolk Achieving the Millennium Development University, Boston. Processed. Goals. Hanoi. HEMA (Health Care Support for the Poor of the GSO (General Statistics Office). 2006. The Northern Uplands & Central Highlands). Situation of Enterprises through the 2007. Survey on Health Seeking Results of Surveys Conducted in 2004, Behaviour of the Poor and Ethnic 2005. Statistical Publishing House, Minorities in Son La Province, Son La, Hanoi. July 2007. Processed. GSO (General Statistics Office). 2007. The Hoang, Xuan Thanh. 2007. Participatory Situation of Enterprises through the Poverty Monitoring at Phuoc Dai and Results of Surveys Conducted in 2004, Phuoc Thanh Communes, Bac Ai 2005, 2006. Statistical Publishing District, Ninh Thuan Province. Report House, Hanoi. Prepared for Oxfam GB. Processed. GSO (General Statistics Office). Various years. Hoang, Xuan Thanh, Thi Mong Phuong Le, Van Statistical Yearbook. Statistical Hai Ngo and Thi Thu Huong Nguyen. Publishing House, Hanoi. 2006. Subsidies in Agricultural Extension for Poverty Reduction in GSO (General Statistics Office) and UNFPA Vietnam. SNV, Hanoi. (United Nations Population Fund). 2006a. The 2004 Vietnam Migration Hohmann, Jurgen and Konrad Obermann. 2007. Survey: Internal Migration and Related Improving Payment for Health Care in Life Course Events. GSO, Hanoi. Hoa Binh Province: Developing a "Research Protocol for the Follow-up of GSO (General Statistics Office) and UNFPA the Fund for the Poor and the Health 129 SOCIAL PROTECTION Insurances in Vietnam": Mission Report. Industrial Zones to Community Health]. GFA Consulting Group. Processed. Nha Xuat Ban Khoa Hoc Xa Hoi, Hanoi. Hughes, R. forthcoming. Climate Change and Institute of Sociology. 2005. Impacts of Existing Vulnerability in Vietnam: Key Issues and Residence Registration Policy on Urban Future Options. Processed. Poverty Alleviation: Two Case Studies in ILO (International Labor Organization). various Hanoi and Ho Chi Minh City. Institute years. Labour and Social Trends in Asia of Sociology, Hanoi. and the Pacific. ILO, Bangkok. IPCC (Intergovernmental Panel on Climate ILO (International Labor Organization). 2007a. Change). 2007. Summary for Labour and Social Trends in ASEAN Policymakers. In Susan Solomon, Dahe 2007: Integration, Challenges and Qin, Martin Manning, Melinda Marquis, Opportunities. ILO, Bangkok. Kristen Averyt, Melinda M.B. Tignor, Henry LeRoy Jr. Miller and Zhenlin ILO (International Labor Organization). 2007b. Chen (eds.), Climate Change 2007: The Understanding and Minimizing Risk for Physical Science Basis. Contribution of Strikes in Vietnam. ILO, Hanoi. Working Group I to the Fourth Processed. Assessment Report of the ILO (International Labor Organization). 2007c. Intergovernmental Panel on Climate Visions for Asia's Decent Work Decade: Change. Cambridge University Press, Sustainable Growth and Jobs to 2015. Cambridge, UK & and New York, USA. ILO, Bangkok. Isik-Dikmelik, Aylin. 2006. Trade Reforms and ILSSA (Institute of Labour Science and Social Welfare: An Ex-Post Decomposition of Affairs). 2006. Report: Employment Income in Vietnam. World Bank Policy Trends in Vietnam 2006-2015. SIDA- Research Working Paper no. 4049. CIEM Project. Hanoi. Processed. World Bank, Washington, D.C. ILSSA (Institute of Labour Science and Social JBIC (Japan Bank for International Affairs) and ILO (International Labor Cooperation). 2007. Seventh Poverty Organization). 2006a. Collective Reduction Support Credit (PRSC7): Bargaining in Vietnam. Hanoi. Japan's Views on Policy Actions. Processed. Processed. ILSSA (Institute of Labour Science and Social Jenkins, Rhys and Kunal Sen. 2006. Affairs) and ILO (International Labor International Trade and Manufacturing Organization). 2006b. National Report Employment in the South: Four Country on Labor Disputes and Settlement of Case Studies. Oxford Development Labor Disputes. Hanoi. Processed. Studies, 34(3):299-322. Institute of Ethnology and World Bank. 2004. JETRO (Japan External Trade Organization). Proceedings of Workshop on Poverty 2005. Impact of Socio-Economic Reduction: Issues and Solutions for Changes on the Livehoods of People Ethnic Minorities in Northern Living in Poverty in Vietnam. Institute of Moutainous Areas in Vietnam. Developing Economics - JETRO, Tokyo. Agriculture Publishing House, Hanoi. JETRO (Japan External Trade Organization). Institute of Social Sciences of Vietnam. 2005. 2006a. Actors in Poverty Reduction in Anh Huong Cua O Nhiem Moi Truong O Vietnam. Institute of Developing Mot So Khu Cong Nghiep Phia Bac Toi Economies. Japan External Trade Suc Khoe Cong Dong [Effect of Organization, Chiba. Environmental Pollution in Northern JETRO (Japan External Trade Organization). 130 BIBLIOGRAPHY 2006b. Comparative Survey on the Processed. Labor Environment in ASEAN, China Knowles, James C. 2007b. Some Key Health and India. JETRO, Tokyo. Insurance Policy Issues in Viet Nam. JICA (Japan International Cooperation Agency) Report to the Asian Development Bank and BMH (Bach Mai Hospital). 2004a. RSC-C60848(VIE): "Health Policy in Joint Evaluation Report on the Japanese Viet Nam: Responding to the Call for Technical Cooperation for the Bach Mai Intensified Engagement: Joint Analytical Hospital Project for Functional and Advisory Activity": Draft. Asian Enhancement. Hanoi. Processed. Development Bank, Hanoi. Processed. JICA (Japan International Cooperation Agency) Knowles, James C., Henrik Axelson and Ngoc and MOH (Ministry of Health). 1999. Uyen Vu. 2007a. Data Used in Joint Evaluation Report on the Project Application of SimIns Plus Model to Viet for Cho Ray Hospital Technical Nam. Report to the Asian Development Cooperation in the Socialist Republic of Bank RSC-C60848(VIE): "Health Vietnam. Hanoi. Processed. Policy in Viet Nam: Responding to the Jowett, M., P. Contoyannis and N.D. Vinh. Call for Intensified Engagement: Joint 2003. The Impact of Public Voluntary Analytical and Advisory Activity". Asian Health Insurance on Private Health Development Bank, Hanoi. Processed. Expenditures in Vietnam. Social Science Knowles, James C., Henrik Axelson and Ngoc and Medicine, 56(2):333-342. Uyen Vu. 2007b. Simulations with the Justino, Patricia. 2005. Beyond HEPR (Hunger SimIns Model Applied to Vietnam: Eradication and Poverty Reduction): a Report to the Asian Development Bank Framework for An Integrated National RSC-C60848(VIE): "Health Policy in System of Social Security in Viet Nam. Viet Nam: Responding to the Call for UNDP Viet Nam Policy Dialogue Paper Intensified Engagement: Joint Analytical 2005/1. DFID and UNDP, Hanoi. and Advisory Activity". Asian Development Bank, Hanoi. Processed. Kabeer, Naila and Thi Van Anh Tran. 2006. Globalisation, Gender and Work in the Knowles, James C. and others. 2006. Summary Context of Economic Transition: The of the Findings of Four Inter-related Case of Viet Nam. UNDP Viet Nam Research Studies on Pro-poor Financing Policy Dialogue Paper 2006/2. UNDP, in Viet Nam: Final Report Prepared for Hanoi. the Ministry of Health and the Asian Development Bank TA No. 4331-VIE: Kerbo, H. 2005. Foreign Investment and Support for Pro-Poor Health Policies. Disparities in Economic Development Hanoi. Processed. and Poverty Reduction: Comparative- Historical Analysis of the Buddhist Knowles, James C. and others. 2005. Making Countries of Southeast Asia. Health Care more Affordable for the International Journal of Comparative Poor: Health Financing in Vietnam: Sociology, 46(5-6):425-459. Final Report-TA No. 3877-VIE Ministry of Health/Asian Development Bank. Knowles, James C. 2007a. Review of Ministry of Health and Asian Implementation of the C2 Sub- Development Bank, Hanoi. component (the Swedish grant component) of the Health Care in the Kokko, Ari and Patrik Tingvall. 2007. The Central Highlands Project: Report to the Welfare State. In Ari Kokko (ed.), The Ministry of Health, Sida and the Asian Role of the State in Vietnam. VASS - Doi Development Bank: Draft. Hanoi. Moi Review Project With Support from UNDP and the Embassy of Sweden, 131 SOCIAL PROTECTION Hanoi. Population Council and JICA, Hanoi. Korinek, K. 2004. Maternal Employment during Le, Van Truyen, Hoang Thai Luong, Khanh Northern Vietnam's Era of Market Phuong Nguyen and Trieu Dzung Le. Reform. Social Forces, 83(2):791-822. 2007. Vietnam National Trade and Korinek, Kim M. 2006. The Status Attainment Health Assessment: Draft Research of Young Adults During Market Report. Ministry of Health, Hanoi. Transition: the Case of Vietnam. Processed. Research in Social Stratification and Lee, Chang-Hee. 2006a. Industrial relations and Mobility, 24(1):55-72. dispute settlement in Vietnam. ILO Kung, I. and H. Wang. 2006. Socially Discussion Paper. International Labour Constructed Ethnic Division of Labour: Office, Hanoi. Labour Control in Taiwanese-Owned Lee, Chang-Hee. 2006b. Recent Industrial Firms in Malaysia and Vietnam. Relations Developments in China and International Sociology, 21(4):580-601. Viet Nam: the Transformation of Kunieda, Shigeki and Ngoc Huynh Do. 2006. Industrial Relations in East Asian Vietnamese Personal Income Tax Transition Economies. Journal of Reform. In The Final Report of the Joint Industrial Relations, 48(3):415-429. Research Program on the Vietnamese Lempert, D., Van Le Nguyen and Tan Sinh Tax System (pp. 21-72). Tax Policy Bach. 2004. Assessment of Institutional Department, Ministry of Finance, Capacity for Disaster Risk Management Vietnam and Policy Research Institute, in Viet Nam (NDM-Partnership, Phase 1 Ministry of Finance, Japan. Main Report ). UNDP, Hanoi. Processed. Lambert, Sylvie, Martin Ravallion and Lieberman, S. S., J. J. Capuno and H. Van Dominique van de Walle. 2007. A Micro- Ming. 2005. Decentralizing Health: Decomposition Analysis of the Lessons from Indonesia, the Philippines Macroeconomic Determinants of Human and Vietnam. In R. White and P. Smoke Development. World Bank Policy (eds.), East Asia Decentralizes: Making Research Working Paper no. 4358. Local Government Work (pp. 155-178). World Bank, Washington, DC. World Bank, Washington, DC. Le, Bach Duong and others. 2005. Social Luu, Hoai Chuan. 2004. Danh Gia Hieu Qua Protection for the Most Needy in Hoat Dong Cua Bac Sy Tuyen Vietnam. The Gioi Publishers, Hanoi. Xa/Phuong Tai Mot So Dia Phuong: De Le, Minh Tuan, Minh Duc Pham and Luc de Tai Khoa Hoc Cap Bo 2002-2003 Wulf. 2007. Estimating Economic [Evaluating the Effectiveness of Doctors Benefits for Revenue Administration at Commune/Ward Level]. Institute of Reform Projects. World Bank, Health Strategy and Policy, Hanoi. Washington D.C. Martinez-Vazquez, Jorge. 2005. Making Fiscal Le, Ngoc Thang and others. 2006. Nghien Cuu Decentralization Work in Vietnam. Ve Dinh Canh, Dinh Cu O Viet Nam Georgia State University, Atlanta. (Research on Settled Agriculture and Martinez-Vazquez, Jorge and Juan Luis Gomez. Settlement in Vietnam). National 2005. Effective Fiscal Decentralization Political Publishing House, Hanoi. in Vietnam. In Proceedings of the Annual Le, Thi Phuong Mai and Quy Nhan Vu. 2005. A Conference on Taxation (pp. 356-361). Final Assessment on the Reproductive Matsuo, Ayaka. 2006. Child Domestic Labour Health Project in Nghe An Province. in South-East and East Asia: Emerging 132 BIBLIOGRAPHY Good Practices to Combat It. MOH (Ministry of Health). 2004. Assessment of International Labour Office, Bangkok. 1-Year Implementation of Examination McGee, Robert W. 2006. The Ethics of Tax and Treatment for the Poor (In Evasion: a Survey of Vietnamese Accordance With Decision Opinion. Andreas School of Business 139/2002/QD-TTg of the Prime Working Paper. Barry University, Miami Minister). Hanoi. Processed. Shores, Florida. MOH (Ministry of Health). 2004. HIV/AIDS Mclure, Charles E. 2005. Tax Policy for Estimates and Projections in Vietnam Vietnam. In William T Alpert (ed.), The 2005- 2010. Hanoi. Vietnamese Economy and Its MOH (Ministry of Health). 2005. Quan ly va to Transformation to An Open Market chuc y te (Managing and Organizing System (pp. 72-127). Sharpe, Armonk, Health System). MOH, Hanoi. N.Y. MOH (Ministry of Health). 2007a. Joint Annual Mcnicoll, Geoffrey. 2006. Policy Lessons of the Health Sector Review (JAHR) 2007. East Asian Demographic Transition. Hanoi. Processed. Population Council, New York. MOH (Ministry of Health). 2007b. Study Mekong Economics. 2006. Measuring Report: Review on the Implementation of Grassroots Democracy in Vietnam. Free Healthcare Services for Children Presented to the Embassy of Finland in under Six in Public Healthcare Facilities Hanoi. Mekong Economics, Hanoi. with Regards to Child Mortality and Minford, M., D. Quang, H. T. Nguyen and P. Morbidity Patterns and Available Brooke. 2004. An Expenditure Tracking Treatment. Study Conducted with Study of the Hunger Eradication and UNICEF Support. Hanoi. Processed. Poverty Reduction Programme and MOH (Ministry of Health). 2007c. Vietnam Programme 135: Final Report. UNDP, Health Report 2006. Medical Publishing Hanoi. Processed. House, Hanoi. Ministry of Foreign Affairs in Finland. 2005. MOH (Ministry of Health), UNICEF (United Reducing Poverty in Quang Tri: Success Nations Fund for Children) and GSO Stories in Vietnamese Rural (General Statistics Office). 2005. Survey Development. Ministry of Foreign Assessment of Vietnamese Youth. MOH, Affairs in Finland, Helsinki. Hanoi. Minot, Nicholas, Michael Epprecht, Thi Tram MOLISA (Ministry of Labor Invalids and Anh Tran and Quang Trung Le. 2006. Social Affairs), ILSSA (Institute of Income Diversification and Poverty in Labour Science and Social Affairs) and the Northern Uplands of Vietnam. IFPRI ILO (International Labour Office). 2003. Research Report No. 145. International Equality, Labour and Social Protection Food Policy Research Institute (IFPRI), for Women and Men in the Formal and Washington, D.C. Informal Economy in Viet Nam: Issues MOF (Ministry of Finance). 2007. Report for Advocacy and Policy Development. Medium-Term Fiscal and Expenditure Labour and Social Affairs Publishing Framework Over 2007-2009. Finance House, Hanoi. Publishing House, Hanoi. MOLISA (Ministry of Labor Invalids and MOH (Ministry of Health). various years. Social Affairs) and UNICEF (United Health Statistics Yearbook. MOH, Nations Fund for Children). 2004. Hanoi. Situational Analysis on Children with Disabilities in Viet Nam. Hanoi. 133 SOCIAL PROTECTION Processed. Workshop on Minimum Wage Policy in MONRE (Ministry of Natural Resources and Market Economy, Held by the Institute Environment). 2003. Initial National of Labor Science and Social Affairs with Communication to the IPCC. Ministry of the Support of the World Bank, Da Nang, Natural Resources and Environment, March 9-10, 2007. Processed. Hanoi. Processed. Nguyen, Duy Khe and others. 2005. Health and Moore, Mick. 2004. Taxation and the Political: Health Care: Equity Aspects in Fila Agenda, North and South. Forum for Bavi, Vietnam. In INDEPTH Network Development Studies No. 1-2004. (ed.), Measuring Health Equity in Small Institute of Development Studies (IDS), Areas: Findings from Demographic Brighton, UK. Surveillance Systems (pp. 127-142). Ashgate, Aldershot, UK. Moore, Mick. 2007. How Does Taxation Affect the Quality of Governance? IDS Working Nguyen, Huu Dzung. 2005. Research on Paper 280. Institute of Development Labour Market and Career Guidance for Studies, Brighton, UK. Vietnamese Youth at Current Time: General Report. (Independent Research MPI (Ministry of Planning and Investment) and Theme at National Level, Research SIDA (Swedish Agency for International Theme Code DTDL-2003/11). Labour Development Cooperation). 2005. Chia and Social Affairs Publishing House, Se - a Different Approach to Poverty Hanoi. Alleviation: Chia Se Vietnam/Sweden Poverty Alleviation Programme. MPI Nguyen, Huu Dzung. 2007. Minimum Wage and SIDA, Hanoi. Setting in Market Economy. Paper Presented at the Workshop on Minimum MPI (Ministry of Planning and Investment) and Wage Policy in Market Economy, Held UNDP (United Nations Development by the Institute of Labor Science and Programme). 2006. Study on Social Affairs with the Support of the Competitiveness and Impacts of World Bank, Da Nang, March 9-10, Liberalization: the Case of Insurance 2007. Processed. Services. UNDP, Hanoi. Nguyen, Huu Hai. 2007. Develop a Reliable Mu, Ren and Dominique van de Walle. 2007. Social Security System to Correspond to Rural Road and Poor Area Development the Socialist-Oriented Market Economy in Vietnam. World Bank Policy Research (in Vietnamese only). Research Paper Working Paper no. 4340. World Bank, No. 8. VASS - Doi Moi Review Project Washington, DC. With Support from UNDP and the National Institute of Nutrition. 2005. Nutrition Embassy of Sweden, Hanoi. Processed. Status of Children and Their Mother in Nguyen, Khanh Phuong. 2006. Factors Viet Nam 2005. Hanoi. Processed. Affecting Household Out-Of-Pocket (UNICEF contribution ­ no file) Expenditure. Ministry of Health and Navigos Group. 2007. Vietnam Salary Survey Asian Development Bank, Hanoi. Report 2007. Hanoi. Processed. Neefjes, Koos. 2004. Participation and Nguyen, Minh Thang, I. Bhushan, E. Bloom Empowerment in the HEPR Programme: and S. Bonu. 2007. Child Immunization Report of a Desk Study. UNDP, Hanoi. in Viet Nam: Situation and Barriers to Coverage. Journal of Biosocial Science, Nguyen, Duc Hung. 2007. Industrial Relation 39(1):41-58. and Wage Bargaining in Industrial Relation. Paper Presented at the Nguyen, Nguyet Nga and Martin Rama. 2007. A 134 BIBLIOGRAPHY Comparison of Quantitative and programme on The Social Challenge of Qualitative Poverty Targeting Methods Development. UNRISD, Geneva. in Vietnam. Q-Squared Working Paper Nguyen, Thi Lan Huong. 2007a. Basis of No. 32. Centre For International Studies regional defferentiation of minimum - University Of Toronto. wages. Paper Presented at the Workshop Nguyen, Quynh Hoa, Ann Ohman, Cecilia on Minimum Wage Policy in Market Stalsby Lundborg and Thi Kim Chuc Economy, Held by the Institute of Labor Nguyen. 2007. Drug Use and Health- Science and Social Affairs with the seeking Behavior for Childhood Illness Support of the World Bank, Da Nang, in Vietnam: A Qualitative Study. Health March 9-10, 2007. Processed. Policy, 82(3):320-329. Nguyen, Thi Lan Huong. 2007b. Nguyen, T. Binh, James W. Albrecht, Susan B. Recommendations from the research and Vroman and M. Daniel Westbrook. 2007. proposed follow-up plan. Paper A Quantile Regression Decomposition of Presented at the Workshop on Minimum Urban-Rural Inequality in Vietnam. Wage Policy in Market Economy, Held Journal of Development Economics, by the Institute of Labor Science and 83(2):466-490. Social Affairs with the Support of the Nguyen, Thanh Liem. 2004. Migration and World Bank, Da Nang, March 9-10, Health in Areas of Vietnam. Ph.D. 2007. Processed. Thesis. Brown University. Nguyen, Thi Lan Huong. 2007c. Scenario for Nguyen, Thanh Liem and Michael J. White. integration of minimum wage levels. 2007. Health Status of Temporary Paper Presented at the Workshop on Migrants in Urban Areas in Vietnam. Minimum Wage Policy in Market International Migration, 45(4):101-134. Economy, Held by the Institute of Labor Science and Social Affairs with the Nguyen, Thi Hang. 2007. Opening Speech on Support of the World Bank, Da Nang, the Workshop. Workshop on Minimum March 9-10, 2007. Processed. Wage Policy in Market Economy, Held by the Institute of Labor Science and Nguyen, Thi Minh Hoa and Martin Rama. 2007. Social Affairs with the Support of the Community Effects and Poverty among World Bank, Da Nang, March 9-10, the Ethnic Minorities of Vietnam. 2007. Processed. Processed. Nguyen, Thi Hong Ha. 2006. The Demand for Nguyen, Tien Dung and Misuo Ezaki. 2005. Student Health Insurance in Vietnam: Regional Economic Integration and Its Final Report Submitted to Asian Impacts on Growth, Poverty and Income Development Bank TA No. 4331-VIE. Distribution: the Case of Vietnam. Asian Development Bank, Hanoi. Review of Urban & Regional Processed. Development Studies, 17(3):197-215. Nguyen, Thi Hong Tu, Thi Lien Huong Nguyen Nguyen, Tran Hien. 2006. The Expanded and Bich Diep Nguyen. 2004. Program on Immunization: Twenty Years Globalization and Its Effects on Health of Achievements in Vietnam. National Care and Occupational Health in Viet Expanded Program on Immunization and Nam: Draft paper prepared for the National Institute of Hygiene and RUIG/UNRISD project on Epidemiology, Hanoi. Globalization, Inequality and Health, a Nguyen, Viet Cuong (ed.). 2004. 72 vu an tranh collaborative international project chap lao dong dien hinh: tom tat va binh forming part of the RUIG research luan (72 Typical Labor Disputes Cases: 135 SOCIAL PROTECTION Summary and Comments). Nha xuat ban Lao dong, Hanoi. Lao dong Xa hoi, Hanoi. Pham, Thai-Hung. 2007. Rural Nonfarm Nguyen, Xuan Thanh and others. 2006. Does Employment Under Trade Reform: "The Injury Poverty Trap" Exist?: a Evidence from Vietnam, 1993-2002. Longitudinal Study in Bavi, Vietnam. PRUS Working Paper No.35. Health Policy, 78(2-3):249-257. Department of Economics. University of Nose, Sakuya and Yasuyuki Sawada. 2007. Sussex. Consumption Insurance against Phan, Thanh Thuy. 2006. Quality of Private Unforeseen Epidemics: The Case of Health Services in Bac Giang City, Avian Influenza in Vietnam. Processed. Vietnam. Master Thesis. Umea O'Donnell, Owen A., Eddy Van Doorslaer and University. Department of Public Health Adam Wagstaff. 2006. Decomposition of and Clinical Medicine. Inequalities in Health and Health Care. Plan in Vietnam. 2004a. Best Practices and In Andrew M. Jones (ed.), Elgar Lessons Learnt From Lang Giang Companion to Health Economics (pp. Health Project. Plan in Vietnam, Hanoi. 179-192). Elgar, Cheltenham, U.K. Plan in Vietnam. 2004b. Report on School O'Donnell, Owen A., Furio C. Rosati and Eddy Based Child Health Promotion van Doorslaer. 2004. Health Effects of Research. Plan in Vietnam, Hanoi. Child Work: Evidence from Rural Plan in Vietnam. 2004c. Thuc Trang Y Te Vietnam. Research Paper Series 53. Tor Truong Hoc Va Dinh Huong Xay Dung Vergata University, CEIS. Mo Hinh Can Thiep Nang Cao Suc Khoe Oostendorp, Remco H., Gian Nguyen Cong, Truong Hoc [Report on School Based Hong Quang Doan and Thi Lan Huong Child Health Promotion Research]. Nguyen. 2006. Trade Liberalization, the Plan, Hanoi. Gender Wage Gap and Returns to Porto, Guido. 2007. Vietnamese Catfish: Education in Vietnam: Working Paper. Impacts of U.S. Antidumping. World Free University. Processed. Bank, Washington, DC. O'Reilly, Sheelagh. 2005. The Utility of Forest Qi, Li, Bill Taylor and Stephen Frost. 2003. Ethics: a Proposed Approach Based on Labour Relations and Regulation in Field Work in Vietnam: Discussion Vietnam: Theory and Practice. City Paper. International Institute for University of Hong Kong, Hong Kong. Environment and Development (IIED), London. Razafindrakoto, Mireille and Francois Roubaud. 2007. The Informal Sector in Oxfam International. 2006. Public Health at Viet Nam: What do Know and How to Risk: A US Free Trade Agreement Could Go Further? A Statistical and Economic Threaten Access to Medicines in Perspective. Statistical Scientific Thailand. Oxfam Briefing Paper No. 86. Information, CS-06. Oxfam International, Oxford, UK. Salemink, O. 2006. Changing Rights and Painter, Martin. 2006. Sequencing Civil Service Wrongs: the Transnational Construction Pay Reforms in Vietnam: Transition or of Indigenous and Human Rights Among Leapfrog? Governance, 19(2):325-346. Vietnam's Central Highlanders. Focaal - Pham, Quy Tho. 2006. Chuyen dich co cau lao European Journal of Anthropology, dong trong xu huong hoi nhap quoc te 47:32-47. (Moving labor structure towards the Sato, Motohiro and Cong Phi Tong. 2006. Land- international integration). Nha xuat ban Related Taxes and Fees in Vietnam. In 136 BIBLIOGRAPHY The Final Report of the Joint Research Assessment of the Potential. Forest Program on the Vietnamese Tax System Policy and Economics, 8(4):386-396. (pp. 123-166). Tax Policy Department, Sunderlin, William D. and Thu Ba Huynh. 2005. Ministry of Finance, Vietnam and Policy Poverty Alleviation and Forests in Research Institute, Ministry of Finance, Vietnam. Center for International Japan. Forestry Research (CIFOR), Jakarta. SEI (Stockholm Environment Institute). 2005a. Sunoo, Jan Jung-Min. 2007. 24 Frequently Final Report of Joint Government of Viet Asked Questions about Industrial Nam Donor Review of Rural Water Relations in Vietnam. Hanoi. Processed. Supply, Sanitation and Health in Viet Nam: Building strengths to change Swinkels, Rob and Carrie Turk. 2006. directions. Hanoi. Explaining Ethnic Minority Poverty in Vietnam: A Summary of Recent Trends SEI (Stockholm Environment Institute). 2005b. and Current Challenges. Background Sector Status Report of Joint Paper for CEM/MPI Meeting on Ethnic Government - Donor Sector Review of Minority Poverty, Hanoi, 28 September Rural Water Supply, Sanitation and 2006. Processed. Health in Viet Nam. Hanoi. Takahashi, Kazushi. 2007. Sources of Regional Sepehri, Ardeshir, Sisira Sarma and Wayne Income: Disparity in Rural Vietnam: Simpson. 2006. Does Non-Profit Health Oaxaca-Blinder Decomposition. IDE Insurance Reduce Financial Burden? Discussion Paper no. 95. Institute of Evidence from the Vietnam Living Developing Economies (IDE), Tokyo. Standards Survey Panel. Health Economics, 15(6):603-616. Tanaka, Tomomi, Colin F. Camerer and Quang Nguyen. 2005. Poverty, Politics, and Sepehri, Ardeshir, Wayne Simpsona and Sisira Preferences: Field Experiments and Sarmab. 2006. The Influence of Health Survey Data from Vietnam. New York Insurance on Hospital Admission and University, New York. Length of Stay--The Case of Vietnam. Social Science & Medicine, 63(7):1757- Tat, S. and D. Barr. 2006. Healthcare in the New 1770. Vietnam: Comparing Patients' Satisfaction with Outpatient Care in a Sepehri, Ardeshir, Cuong Vu and Thanh Tam Traditional Neighborhood Clinic and a Le. 2005. User Fees, Financial New, Western-Style Clinic in Ho Chi Autonomy and Access to Social Services Minh City. Social Science & Medicine, in Viet Nam. United Nations Discussion 62(5):1229-1236. Paper No. 6. United Nations, Hanoi. Thai, Hung Cam. 2006. Money and Masculinity Sikor, Thomas and Quang Tan Nguyen. 2007. among Low Wage Vietnamese Why May Forest Devolution Not Benefit Immigrants in Transnational Families. the Rural Poor? Forest Entitlements in International Journal of Sociology of the Vietnam's Central Highlands. World Family, 32(2):247-271. Development, 35(11):2010-2025. Tipple, G. 2006. Employment and Work Stern, Nicholas. 2007. The Economics of Conditions in Home-Based Enterprises Climate Changes: The Stern Review. in Four Developing Countries: Do They Cambridge University Press, Constitute `Decent Work'? Work, Cambridge, UK. Employment and Society, 20(1):167-179. Sunderlin, William D. 2006. Poverty Alleviation Tran, Anh Vu, H. Lee Seldon, Duc Hoang Chu Through Community Forestry in and Phan Kien Nguyen. 2006. Electronic Cambodia, Laos, and Vietnam: An 137 SOCIAL PROTECTION Healthcare Communications in Vietnam Taking Stock, Planning Ahead: in 2004. International Journal of Evaluation of the National Targeted Medical Informatics, 75(10-11):764- Programme on Hunger Eradication and 770. Poverty Reduction and Programme 135. Tran, Duc Vien, Vinh Quang Nguyen and Van UNDP and MOLISA, Hanoi. Thanh Mai. 2005. Decentralization UNDP (United Nations Development Process and Its Impacts on Livelihoods Programme) and MPI (Ministry of of Ethnic Minority Groups: A Study on Planning and Investment). 2006. Impacts Decentralization Process in Forest of Basic Public Services Liberalization Management in Northern and North on the Poor and Marginalized People: Central Uplands of Vietnam. Agriculture the Case of Health, Education and Publishing House, Hanoi. English Electricity in Viet Nam. UNDP and MPI, Tran, Quang Huy, Annika Johansson and Hoang Hanoi. Long Nguyen. 2007. Reasons for not UNESCO (United Nations Educational, Reporting Deaths: A Qualitative Study in Scientific and Cultural Organization). Rural Vietnam. World Health and 2005. Community-Based Literacy Population, Feb.(2007). Programmes for Minority Language Tran, Thi Mai Oanh. 2006. Report on Health Contexts in Asia. UNESCO, Bangkok. and Poverty Study (TA No. 4331-VIE: UNFPA (United Nations Population Fund). Support for Pro-Poor Health Policies). 2006. Provision and Utilisation of Asian Development Bank, Hanoi. Reproductive Health Care Services in Processed. Seven UNFPA-Supported Provinces in Tran, Tuan and others. 2004. Maternal Social the 7th Country Programme: Baseline Capital and Child Health in Vietnam. Survey Report. UNFPA, Hanoi. Young Lives, Hanoi. UNICEF (United Nations Fund for Children). Tran, Tuan, Thi Mai Dung Van, Ingo Neu and 2006. Out-Patient Visits and Hospital Michael J. Dibley. 2005. Comparative Admissions at Three Referral Pediatric Quality of Private and Public Health Hospitals in Vietnam, 1995-2004: Services in Rural Vietnam. Health Policy Preliminary Report. Hanoi. Processed. and Planning, 20(5):319-327. UNICEF (United Nations Fund for Children) Trinh, Hoa Binh, Van Hai Tran, Thi Thuy Nga and CERWASS (Center for Rural Water Nguyen and Thi Hiep Vu. 2007. Supply and Environmental Sanitation). Application of Economic Modeling to 2007. Water Environment and Sanitation Measure the Likely Impact of Stronger Knowledge, Attitude and Practices Intellectual Property Rights (IPR) Survey among H'mong and Gia Rai. Protection on Medicine Prices in Processed. Vietnam: Research Report. Ministry of UNICEF (United Nations Fund for Children) Health and World Health Organization, and Institute of Sociology. 2007. Impact Hanoi. Processed. of the Domestic Migration on the Family UNDP (United Nations Development and Children Left Behind. On-going Programme). 2003. Health Care Research Project. Hanoi. Processed. Financing for Viet Nam. UNDP UNICEF (United Nations Fund for Children) Discussion Paper no. 2. UNDP, Hanoi. and WHO (World Health Organization). UNDP (United Nations Development 2007. Child Survival Profile: Vietnam. Programme) and MOLISA (Ministry of UNICEF, Hanoi. Labor Invalids and Social Affairs). 2004. United Nations. 2007. World Population 138 BIBLIOGRAPHY Prospects: The 2006 Revision. United VNCI (Vietnam Competitiveness Initiative) and Nations, New York. VCCI (Vietnam Chamber of Commerce UNODC (United Nations Office on Drugs and and Industry). 2007. The Vietnam Crime). 2003. Ethnic Minorities, Drug Provincial Competitiveness Index 2007: Use and Harm in the Highlands of Measuring Economic Governance for Northern Vietnam: a Contextual Private Sector Development. VNCI Analysis of the Situation in Six Policy Paper No. 12. With Contribution Communes from Son La, Lai Chau, and by The Asia Foundation (TAF) and Lao Cai. UNODC, Hanoi. Development Alternatives Inc. (DAI), Hanoi. Valdelin, Jan, Thanh Huyen Dao and Gunilla Krantz. 2006. Health Cooperation at the Vo, Thi Kieu and Wendy Poussard. 2006. Who is Crossroads: More of the Same or at Risk? An Analysis of Deaths Caused Making Difference: Vietnam-Sweden by Natural Disasters in Quang Ngai Health Cooperation on Health Policy Province from 1999 To 2005. AusAID, and Systems Development 2001-2005. Hanoi. Processed. SIDA, Stockholm. Wagstaff, Adam. 2005a. Decomposing Changes van de Walle, Dominique. 2004. Testing in Income Inequality into Vertical and Vietnam's Public Safety Net. Journal of Horizontal Redistribution and Comparative Economics, 32(4):661-679. Reranking, with Applications to China and Vietnam. World Bank Policy van de Walle, Dominique and Dileni Research Working Paper No. 3559. Gunewardena. 2000. Sources of Ethnic World Bank, Washington, DC. Inequality in Vietnam. World Bank Policy Research Working Paper no. Wagstaff, Adam. 2005b. Inequality 2297. World Bank, Washington, DC. Decomposition and Geographic Targeting with Applications to China and van de Walle, Dominique and Ren Mu. 2007. Vietnam. Health Economics, 14(6):649- Fungibility and the Flypaper Effect of 653. Project Aid: Micro-evidence for Vietnam. Journal of Development Wagstaff, Adam. 2007a. The Economic Economics, 84(2):667-685. Consequences of Health Shocks: Evidence from Vietnam. Journal of van Doorslaer, Eddy and others. 2006. Effect of Health Economics, 26(1):82-100. Payments for Health Care on Poverty Estimates in 11 Countries in Asia: An Wagstaff, Adam. 2007b. Health Insurance for Analysis of Household Survey Data. The the Poor: Initial Impacts of Vietnam's Lancet, 368(9544):1357-1367. Health Care Fund for the Poor. World Bank Policy Research Working Paper VASS (Vietnamese Academy of Social no. 4134. World Bank, Washington D.C. Sciences). 2007. Vietnam Poverty Update Report 2006: Poverty and Wagstaff, Adam. 2007c. Social Health Poverty Reduction in Vietnam 1993- Insurance Reexamined. World Bank 2004. National Political Publishers, Policy Research Working Paper No. Hanoi. 4111. World Bank, Washington, DC. Vittas, Dimitri, Gregorio Impavido and Ronan Wagstaff, Adam and Menno Pradhan. 2005. O'Connor. 2007. Upgrading the Health Insurance Impacts on Health and Investment Policy Framework of Public Non-Medical Consumption in a Pension Funds. World Bank, Developing Country. World Bank Policy Washington, DC. Research Working Paper no. 3563. World Bank, Washington D.C. 139 SOCIAL PROTECTION Wagstaff, Adam and Eddy van Doorslaer. 2001. South-East Asia Regions (2006-2010). Paying for Health Care: Quantifying WHO, Geneva. Fairness, Catastrophe, and World Bank. 2003. Vietnam Development Impoverishment, With Applications to Report 2004: Poverty. World Bank, Vietnam, 1993-98. World Bank Policy Hanoi. Research Working Paper No. 2715. World Bank, Washington, DC. World Bank. 2004. Vietnam Development Report 2005: Governance. World Bank, Wagstaff, Adam, Eddy van Doorslaer and Hanoi. Naoko Watanabe. 2001. On Decomposing the Causes of Health World Bank. 2005a. Comments on the Draft of Sector Inequalities With An Application the New Social Security Law. World to Malnutrition Inequalities in Vietnam. Bank, Hanoi. Processed. World Bank Policy Research Working World Bank. 2005b. Health Analysis 2004 - Paper no. 2714. World Bank, VHLSS 2004. World Bank, Hanoi. Washington, DC. Processed. Waldman, Linda and others. 2005. World Bank. 2005c. Vietnam Development Environment, Politics, and Poverty: Report 2006: Business. World Bank, Lessons from a Review of PRSP Hanoi. Stakeholder Perspectives: Synthesis Review. Study Initiated under the World Bank. 2006a. Accelerating Vietnam's Poverty Environment Partnership (PEP), Rural Development: Growth, Equity and and jointly funded and managed by the Deversification. World Bank, Hanoi. Canadian International Development World Bank. 2006b. Explaining Ethnic Minority Agency (CIDA), Department for Poverty in Vietnam: a Summary of International Development (DFID) of Recent Trends and Current Challenges. the United Kingdom, and German World Bank, Hanoi. Processed. Technical Cooperation Agency (GTZ). Institute of Development Studies, World Bank. 2006c. Social Protection in London. Vietnam: Background Note. World Bank, Hanoi. Processed. Waseem, Saba and Carmen de Paz Nieves. 2006. Social Protection in Vietnam: World Bank. 2006d. Urban Development Background Note. World Bank, Hanoi. Strategy: Meeting the Challenges of Processed. Rapid Urbanization and the Transition to a Market Oriented Economy. Vietnam's WHO (World Health Organization). 1998. CHD Infrastructure Challenge. World Bank, (Child Health and Development) Hanoi. 1996­1997 Report. WHO, Geneva. World Bank. 2006e. Vietnam Development WHO (World Health Organization). 2003. Report 2007: Aiming High. World Bank, Health and Ethnic Minorities in Viet Hanoi. Nam. WHO, Hanoi. World Bank. 2006f. Water Supply and WHO (World Health Organization). 2005a. Sanitation Strategy: Building on a Solid Maternal Mortality in Viet Nam 2000- Foundation. Vietnam's Infrastructure 2001: An In-Depth Analysis of Causes Challenge. World Bank, Hanoi. and Determinants. WHO, Geneva. World Bank. 2007a. Vietnam - First Program WHO (World Health Organization). 2005b. 135 Second Phase Support Operation for Strategy on health care financing for Communes: Facing Extreme Hardship in countries of the Western Pacific and Ethnic Minority and Mountainous Areas. 140 BIBLIOGRAPHY World Bank, Washington D.C. Development Bank TA 4480 VIE. Hanoi. World Bank. 2007b. Vietnam: Higher Education Processed. and Skills for Growth: Draft. Hanoi. Xu, Ke and others. 2003. Household Processed. catastrophic health expenditure: A World Bank. 2007c. Vietnam Interim Health multicountry analysis. The Lancet, Sector Issues Note. World Bank, Hanoi. 362(9378):111-117. Processed. Zezulin, Lena S. and Jean-Noel Martineau. World Bank. Forthcoming-a. Country Social 2003. Financial Markets Development Analysis: Ethnicity and Development in Program and the Social Security System: Vietnam. World Bank, Hanoi. Report Presented to Asian Development Bank under T.A. No. 3147-VIE. Asian World Bank. Forthcoming-b. Tax Policy Study: Development Bank, Hanoi. Processed. Work in Progress. Hanoi. Processed. Zhang, H. X. and others. 2006. Migration in a WPI (World Perspectives, Incorporated). 2007. Transitional Economy: Beyond the Developing Agricultural Insurance in Planned and Spontaneous Dichotomy in Vietnam: Phase I Final Report for Asian Vietnam. Geoforum, 37(6):1066-1081. 141 STATISTICAL APPENDIX 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 2.2 A GDP by industrial origin and by economic sector in constant prices Table 2.2 B GDP by industrial origin -- 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 Merchandise exports by commodities Table 3.3 Merchandise imports by commodities Monetary Survey Table 4.1 Monetary survey Budget Table 5.1 State budget revenues (VND billdion) Table 5.2 State budget revenues (share of GDP) Table 5.3 State budget expenditures (VND billion) Table 5.4 State budget expenditures (share of GDP) Table 5.5 External Debt Prices Table 6.1 A Monthly change in consumer retail prices Table 6.1 B Monthly comsumer price index Table 6.2 A Price index by commodity groups (monthly change) Table 6.2 B Price index by commodity groups: monthly growth rates (December 2005 = 100) 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 production output Table 8.2 Major industrial products Table 1.1: 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 40,311 41,721 21,737 60,295 2005 83,106 1.31 40,846 42,260 22,337 60,770 2006 84,156 1.26 41,355 42,801 22,824 61,332 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 (2007) Table 1.2: POPULATION BY SEX AND LOCALITY IN 2006 (thousand persons) Total By sex By locality Provinces/ Cities Male Female Urban Rural Ha Noi 3,217 1,612 1,604 2,102 1,115 Hai Phong 1,803 891 912 826 978 Ha Giang 684 338 345 76 608 Tuyen Quang 732 363 370 69 664 Cao Bang 519 255 264 70 449 Lang Son 746 371 375 150 596 Lai Chau 320 161 159 50 270 Dien bien 459 231 229 77 382 Lao Cai 586 393 193 108 478 Yen Bai 741 367 374 147 594 Bac Can 302 151 151 45 256 Thai nguyen 1,127 563 564 257 870 Son La 1,008 507 501 110 897 Hoa Binh 820 409 412 126 694 Vinh Phuc 1,180 572 609 165 1,015 Phu Tho 1,337 657 680 210 1,127 Bac ninh 1,010 492 518 133 877 Bac Giang 1,594 788 806 147 1,448 Quang Ninh 1,091 553 539 509 582 Ha Tay 2,544 1,226 1,318 261 2,282 Hai Duong 1,723 835 888 268 1,454 Hung Yen 1,143 553 590 127 1,016 Thai Binh 1,865 896 969 136 1,730 Nam Dinh 1,974 963 1,011 309 1,666 Ha Nam 827 400 426 80 747 Ninh Binh 923 450 473 141 781 Thanh Hoa 3,680 1,803 1,878 360 3,320 Nghe An 3,064 1,503 1,562 326 2,739 Ha Tinh 1,306 649 658 144 1,162 Quang Binh 848 419 429 119 729 Quang Tri 626 309 316 154 472 Thua Thien - Hue 1,144 562 582 360 784 Quang Nam 1,473 713 760 252 1,221 Da Nang 789 381 407 680 109 Quang Ngai 1,296 629 667 186 1,110 Binh Dinh 1,566 763 803 394 1,172 Phu Yen 873 433 440 183 690 Khanh Hoa 1,135 564 572 453 682 Ninh Thuan 568 281 287 185 383 Binh Thuan 1,163 580 583 437 726 Gia Lai 1,162 591 570 350 812 Kon Tum 383 192 191 134 249 Dac Lac 1,738 876 861 389 1,349 Dac Nong 407 202 205 53 355 Lam Dong 1,179 590 589 443 736 Ho Chi Minh City 6,106 2,946 3,160 5,245 861 Binh Duong 964 462 502 282 682 Tay Ninh 1,047 515 532 178 869 Binh Phuoc 810 411 398 123 686 Dong Nai 2,215 1,099 1,116 686 1,529 Baria - Vung Tau 926 464 462 414 513 Long An 1,423 700 723 236 1,188 Dong Thap 1,668 814 854 252 1,416 An Giang 2,210 1,087 1,124 621 1,590 Tien Giang 1,717 834 884 256 1,461 Ben Tre 1,353 657 696 132 1,221 Vinh Long 1,057 514 543 158 899 Tra Vinh 1,037 505 532 150 887 Can Tho 1,140 561 579 572 568 Hau giang 797 392 405 124 673 Soc Trang 1,276 622 654 235 1,041 Kien Giang 1,685 831 854 405 1,279 Bac Lieu 820 399 421 209 611 Ca Mau 1,232 608 624 248 984 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 (2007) Table 1.3: TOTAL EMPLOYMENT BY SECTOR (thousand of persons) Rev Rev Prel 2000 2001 2002 2003 2004 2005 2006 Total Employment 37,610 38,563 39,508 40,574 41,586 42,527 43,347 State 3,501 3,604 3,751 4,035 4,142 4,039 4,008 Non-state 34,109 34,959 35,757 36,538 37,445 38,488 39,339 State Sector Employment 3,501 3,604 3,751 4,035 4,142 4,039 4,008 Central 1,442 1,499 1,569 1,628 1,678 1,613 1,572 Local 2,059 2,105 2,181 2,407 2,464 2,426 2,435 Employment by Sector Agriculture, forestry and fisheries 24,481 24,468 24,456 24,443 24,431 24,342 24,123 Industry and Construction 4,930 5,552 6,085 6,671 7,217 7,782 8,280 Services 8,199 8,542 8,967 9,460 9,939 10,402 10,945 Note : Figures are rounded Source : GSO (2007) Table 2.1: GDP BY INDUSTRIAL ORIGIN AND BY ECONOMIC SECTOR (VND billion at current prices) Rev Prel 2000 2001 2002 2003 2004 2005 2006 Total 441,646 481,295 535,762 613,443 715,307 839,211 973,790 State 170,141 184,836 205,652 239,736 279,704 322,241 363,449 Non-State 271,505 296,459 330,110 373,707 435,603 516,970 610,341 Agriculture, Forestry and Fisheries 108,356 111,858 123,383 138,285 155,993 175,984 198,266 Agriculture 87,537 87,861 96,543 106,385 119,107 132,985 149,234 Forestry 5,913 6,093 6,500 7,775 9,412 10,052 10,780 Fisheries 14,906 17,904 20,340 24,125 27,474 32,947 38,252 Industry and Construction 162,220 183,515 206,197 242,126 287,616 344,224 404,753 Mining 42,606 44,345 46,153 57,326 72,492 88,897 99,919 Manufacturing 81,979 95,211 110,285 125,476 145,475 173,122 206,945 Electricity and Water 13,993 16,028 18,201 22,224 25,091 28,929 33,386 Construction 23,642 27,931 31,558 37,100 44,558 53,276 64,503 Services 171,070 185,922 206,182 233,032 271,698 319,003 370,771 Trade 62,836 67,788 75,617 83,297 96,995 113,768 132,794 Hotel and Restaurant 14,343 15,412 17,154 18,472 22,529 29,329 35,861 Transportation and Communication 17,341 19,431 21,095 24,725 30,402 36,629 43,825 Finance, Banking and Insurance 8,148 8,762 9,763 10,858 12,737 15,072 17,607 Science and Technology 2,345 2,646 3,009 3,694 4,315 5,247 6,059 Real Estate and Renting 19,173 21,589 24,452 27,287 31,304 33,635 36,814 Public Administration 12,066 12,784 13,816 16,676 19,061 23,037 26,737 Education and Training 14,841 16,245 18,071 21,403 23,335 26,948 30,718 Healthcare and social welfare 5,999 6,417 7,057 8,865 10,851 12,412 14,093 Culture and Recreation 2,558 2,800 2,987 3,376 3,693 4,158 4,617 Party and Association 614 651 712 774 885 1,054 1,217 Community and Social Service 9,853 10,412 11,412 12,497 14,354 16,293 18,789 Private Household Employment 953 985 1,037 1,108 1,237 1,421 1,640 Source : GSO (2007) Table 2.2: GDP BY INDUSTRIAL ORIGIN AND BY ECONOMIC SECTOR (VND billion at constant 1994 prices) Rev Prel 2000 2001 2002 2003 2004 2005 2006 Total 273,666 292,535 313,247 336,242 362,435 393,031 425,135 State 111,522 119,824 128,343 138,160 148,865 159,836 170,001 Non-State 162,144 172,711 184,904 198,082 213,570 233,195 255,134 Agriculture, Forestry and Fisheries 63,717 65,618 68,352 70,827 73,917 76,888 79,505 Agriculture 54,493 55,613 57,912 59,761 62,107 64,072 65,892 Forestry 2,544 2,556 2,568 2,589 2,610 2,635 2,665 Fisheries 6,680 7,449 7,872 8,477 9,200 10,181 10,948 Industry and Construction 96,913 106,986 117,126 129,399 142,621 157,867 174,239 Mining 18,430 19,185 19,396 20,611 22,437 22,854 23,037 Manufacturing 51,492 57,335 63,983 71,363 79,116 89,338 100,396 Electricity and Water 6,337 7,173 7,992 8,944 10,015 11,247 12,574 Construction 20,654 23,293 25,755 28,481 31,053 34,428 38,232 Services 113,036 119,931 127,769 136,016 145,897 158,276 171,391 Trade 44,644 47,779 51,245 54,747 59,027 63,950 69,418 Hotel and Restaurant 8,863 9,458 10,125 10,646 11,511 13,472 15,145 Transportation and Telecom 10,729 11,441 12,252 12,925 13,975 15,318 16,870 Finance, Banking and Insurance 5,650 6,005 6,424 6,935 7,495 8,197 8,867 Science and Technology 1,571 1,749 1,909 2,044 2,196 2,368 2,543 Real Estate and Renting 12,231 12,631 13,106 13,796 14,396 14,816 15,252 Public Administration 8,021 8,439 8,768 9,228 9,773 10,477 11,270 Education and Training 9,162 9,687 10,475 11,260 12,125 13,127 14,231 Healthcare and social welfare 3,946 4,151 4,464 4,853 5,234 5,640 6,082 Culture and Recreation 1,601 1,648 1,706 1,857 1,997 2,163 2,329 Party and Association 317 334 353 372 395 423 454 Community and Social Service 5,734 6,026 6,353 6,743 7,141 7,655 8,210 Private Household Employment 567 583 589 610 632 670 720 Source : GSO (2007) Table 2.2B: GDP BY INDUSTRIAL ORIGIN -- GROWTH RATE (in percent) Rev Prel 2000 2001 2002 2003 2004 2005 2006 Total 6.8 6.9 7.1 7.3 7.8 8.4 8.2 State 7.7 7.4 7.1 7.6 7.7 7.4 6.4 Non-State 6.2 6.5 7.1 7.1 7.8 9.2 9.4 Agriculture, Forestry and Fisheries 4.6 3.0 4.2 3.6 4.4 4.0 3.4 Agriculture 4.0 2.1 4.1 3.2 3.9 3.2 2.8 Forestry 0.4 0.5 0.5 0.8 0.8 1.0 1.1 Fisheries 11.6 11.5 5.7 7.7 8.5 10.7 7.5 Industry and Construction 10.1 10.4 9.5 10.5 10.2 10.7 10.4 Mining 7.2 4.1 1.1 6.3 8.9 1.9 0.8 Manufacturing 11.7 11.3 11.6 11.5 10.9 12.9 12.4 Electricity and Water 14.6 13.2 11.4 11.9 12.0 12.3 11.8 Construction 7.5 12.8 10.6 10.6 9.0 10.9 11.0 Services 5.3 6.1 6.5 6.5 7.3 8.5 8.3 Trade 6.3 7.0 7.3 6.8 7.8 8.3 8.6 Hotel and Restaurant 4.1 6.7 7.1 5.1 8.1 17.0 12.4 Transportation and Telecom 5.8 6.6 7.1 5.5 8.1 9.6 10.1 Finance, Banking and Insurance 6.1 6.3 7.0 8.0 8.1 9.4 8.2 Science and Technology 24.0 11.3 9.1 7.1 7.4 7.8 7.4 Real Estate and Renting 2.6 3.3 3.8 5.3 4.3 2.9 2.9 Public Administration 3.9 5.2 3.9 5.2 5.9 7.2 7.6 Education and Training 4.0 5.7 8.1 7.5 7.7 8.3 8.4 Healthcare and social welfare 6.4 5.2 7.5 8.7 7.9 7.8 7.8 Culture and Recreation 6.4 2.9 3.5 8.9 7.5 8.3 7.7 Party and Association 5.7 5.4 5.7 5.4 6.2 7.1 7.3 Community and Social Service 3.1 5.1 5.4 6.1 5.9 7.2 7.3 Private Household Employment 3.1 2.8 1.0 3.6 3.6 6.0 7.5 Source : GSO (2007) Table 2.3A: GDP DEFLATOR Rev Prel 2000 2001 2002 2003 2004 2005 2006 Total 1.6 1.6 1.7 1.8 2.0 2.1 2.3 State 1.5 1.5 1.6 1.7 1.9 2.0 2.1 Non-State 1.7 1.7 1.8 1.9 2.0 2.2 2.4 Agriculture, Forestry and Fisheries 1.7 1.7 1.8 2.0 2.1 2.3 2.5 Agriculture 1.6 1.6 1.7 1.8 1.9 2.1 2.3 Forestry 2.3 2.4 2.5 3.0 3.6 3.8 4.0 Fisheries 2.2 2.4 2.6 2.8 3.0 3.2 3.5 Industry and Construction 1.7 1.7 1.8 1.9 2.0 2.2 2.3 Mining 2.3 2.3 2.4 2.8 3.2 3.9 4.3 Manufacturing 1.6 1.7 1.7 1.8 1.8 1.9 2.1 Electricity and Water 2.2 2.2 2.3 2.5 2.5 2.6 2.7 Construction 1.1 1.2 1.2 1.3 1.4 1.5 1.7 Services 1.5 1.6 1.6 1.7 1.9 2.0 2.0 Trade 1.4 1.4 1.5 1.5 1.6 1.8 1.9 Hotel and Restaurant 1.6 1.6 1.7 1.7 2.0 2.2 2.4 Transportation and Telecom 1.6 1.7 1.7 1.9 2.2 2.4 2.6 Finance, Banking and Insurance 1.4 1.5 1.5 1.6 1.7 1.8 2.0 Science and Technology 1.5 1.5 1.6 1.8 2.0 2.2 2.4 Real Estate and Renting 1.6 1.7 1.9 2.0 2.2 2.3 2.4 Public Administration 1.5 1.5 1.6 1.8 2.0 2.2 2.4 Education and Training 1.6 1.7 1.7 1.9 1.9 2.1 2.2 Healthcare and social welfare 1.5 1.5 1.6 1.8 2.1 2.2 2.3 Culture and Recreation 1.6 1.7 1.8 1.8 1.8 1.9 2.0 Party and Association 1.9 1.9 2.0 2.1 2.2 2.5 2.7 Community and Social Service 1.7 1.7 1.8 1.9 2.0 2.1 2.3 Private Household Employment 1.7 1.7 1.8 1.8 2.0 2.1 2.3 Source : GSO (2007) Table 2.3B: CHANGE IN GDP DEFLATOR (in percent) Rev Prel 2000 2001 2002 2003 2004 2005 2006 Total 3.4 1.9 4.0 6.7 8.2 8.2 7.3 State 2.0 1.1 3.9 8.3 8.3 7.3 6.0 Non-State 4.4 2.5 4.0 5.7 8.1 8.7 7.9 Agriculture, Forestry and Fisheries 1.8 0.2 5.9 8.2 8.1 8.5 9.0 Agriculture 1.0 -1.7 5.5 6.8 7.7 8.2 9.1 Forestry 2.7 2.6 6.2 18.6 20.1 5.8 6.0 Fisheries Industry and Construction 6.8 2.5 2.6 6.3 7.8 8.1 6.5 Mining 18.0 0.0 2.9 16.9 16.2 20.4 11.5 Manufacturing 3.7 4.3 3.8 2.0 4.6 5.4 6.4 Electricity and Water 4.2 1.2 1.9 9.1 0.8 2.7 3.2 Construction 1.0 4.8 2.2 6.3 10.2 7.8 9.0 Services 1.4 2.4 4.1 6.2 8.7 7.4 0.0 Trade -0.5 0.8 4.0 3.1 8.0 8.3 7.5 Hotel and Restaurant 2.8 0.7 4.0 2.4 12.8 11.2 8.8 Transportation and Telecom 5.4 5.1 1.4 11.1 13.7 9.9 8.6 Finance, Banking and Insurance 2.6 1.2 4.2 3.0 8.5 8.2 8.0 Science and Technology -0.6 1.4 4.2 14.7 8.7 12.8 7.5 Real Estate and Renting 2.4 9.0 9.2 6.0 9.9 4.4 6.3 Public Administration -0.6 0.7 4.0 14.7 7.9 12.7 7.9 Education and Training 1.9 3.5 2.9 10.2 1.2 6.7 5.1 Healthcare and social welfare 4.3 1.7 2.3 15.6 13.5 6.2 5.3 Culture and Recreation 1.1 6.3 3.1 3.8 1.7 4.0 3.1 Party and Association -0.5 0.6 3.5 3.2 7.7 11.2 7.6 Community and Social Service 2.6 0.6 4.0 3.2 8.5 5.9 7.5 Private Household Employment 3.3 0.5 4.2 3.2 7.8 8.4 7.4 Source : GSO (2007) Table 2.4A: NATIONAL ACCOUNTS: SOURCES AND USES (VND billion at current prices) Rev Prel 2000 2001 2002 2003 2004 2005 2006 Sources 452,524 492,277 563,446 664,671 769,307 874,299 1,006,261 GDP 441,646 481,295 535,762 613,443 715,307 839,211 973,790 Trade Balance 10,878 10,982 27,684 51,228 54,000 35,088 32,471 Uses 452,524 492,277 563,446 664,731 769,307 874,299 1,006,261 Total Consumption 321,853 342,607 382,137 445,221 511,221 584,793 668,540 Gross Capital Formation 130,771 150,033 177,983 217,434 253,686 298,543 347,900 Statistical Discrepancy -100 -363 3,326 2,076 4,400 -9,037 -10,179 Source : GSO (2007) Table 2.4B: NATIONAL ACCOUNTS: SOURCES AND USES (VND billion at constant 1994 prices) Rev Prel 2000 2001 2002 2003 2004 2005 2006 Sources 283,751 304,230 334,640 367,691 392,558 417,469 449,966 GDP 273,666 292,535 313,247 336,243 362,435 393,031 425,135 Trade Balance 10,085 11,695 21,393 31,448 30,123 24,438 24,831 Uses 283,751 304,232 334,640 367,691 392,558 417,469 449,966 Total Consumption 200,665 210,029 225,610 243,515 260,940 280,104 301,382 Gross Capital Formation 83,496 92,487 104,256 116,623 128,916 143,291 156,645 Statistical Discrepancy -410 1,716 4,774 7,553 2,702 -5,926 -8,061 Source : GSO (2007) Table 3.1: BALANCE OF PAYMENTS (US$ million, unless otherwise indicated) Rev Est 2000 2001 2002 2003 2004 2005 2006 Exports (fob) 14,449 15,027 16,706 20,150 26,485 32,447 39,826 Imports (fob) 14,071 14,546 17,760 22,730 28,772 34,886 42,602 Trade Balance 378 481 -1,054 -2,580 -2,287 -2,439 -2,776 Non-Factor Services -615 -572 -750 -780 -870 -220 -8 Services (net) -595 -635 -790 -815 -890 -1,220 -1,430 Transfers (net) 1,476 1,250 1,920 2,240 2,485 3,380 4,050 - Official Transfers 136 150 150 140 175 230 250 - Private Transfers (net) 1,340 1,100 1,770 2,100 2,310 3,150 3,800 Current Account Balance (incl. Grants) 644 524 -674 -1,940 -1,565 -500 -165 Capital Account -525 -330 1,137 4,017 2,448 2,628 4,486 Medium and Long-Term 730 625 520 974 1,396 1,360 1,140 - Disbursements 1,410 990 1,100 1,540 1,920 1,915 1,690 ODA Loans 1,360 960 1,070 1,260 1,395 1,430 1,380 Non-concessional LT Borrowing 50 30 30 280 525 485 310 - Amortizations 680 365 580 566 524 555 550 Portfolio investment 865 1,315 Short-Term (net) (1) -1,755 -1,380 -993 1,738 -293 -1,030 276 Direct investment 1,100 1,245 2,025 1,895 1,880 1,960 2,400 FDI loan repayments 600 820 415 590 535 525 645 Overall Balance 119 194 463 2,077 883 2,131 4,321 Financing -119 -194 -463 -2,077 -883 -2,131 -4,321 Memo item: Current account as % of GDP 0.0 0.0 0.0 57.5 63.2 -0.9 -0.3 Note : Figures are rounded (1) Including errors and omissions Source : SBV, World Bank and IMF Table 3.2: MAJOR EXPORTS BY COMMODITY (US$ million) Prel 2000 2001 2002 2003 2004 2005 2006 Total Exports 14,448 15,027 16,706 20,176 26,485 32,442 39,826 Rice 667 625 726 721 950 1,047 1,276 Quantity (000 tons) 3,477 3,729 3,241 3,813 4,060 5,250 4,643 Average Unit Value (US$/ton) 192 168 224 189 234 199 275 Crude oil 3,503 3,126 3,270 3,812 5,671 7,373 8,265 Quantity (000 tons) 15,424 16,732 16,879 17,143 19,501 17,967 16,419 Average Unit Value (US$/ton) 227 187 194 222 291 410 503 Coal 94 113 156 184 355 669 915 Quantity (000 tons) 3,251 4,290 6,049 7,246 11,624 17,986 29,307 Average Unit Value (US$/ton) 29 26 26 25 31 37 31 Rubber 166 166 268 378 641 804 1,286 Quantity (000 tons) 273 308 449 433 975 587 708 Average Unit Value (US$/ton) 607 539 597 872 658 1,370 1,817 Tea 70 78 83 60 96 97 110 Quantity (000 tons) 56 68 75 60 99 88 106 Average Unit Value (US$/ton) 1,250 1,150 1,103 1,002 961 1,103 1,045 Coffee 501 391 322 505 641 735 1,217 Quantity (000 tons) 734 931 719 749 975 892 981 Average Unit Value (US$/ton) 683 420 449 674 658 824 1,241 Cashew Nut 167 152 209 284 436 502 504 Quantity (000 tons) 34 44 62 84 105 109 127 Average Unit Value (US$/ton) 4,892 3,474 3,358 3,390 4,150 4,610 3,973 Black Pepper 146 91 107 105 152 150 190 Quantity (000 tons) 37 57 77 74 112 109 117 Average Unit Value (US$/ton) 3,943 1,601 1,399 1,416 1,362 1,381 1,632 Marine Products 1,479 1,778 2,023 2,200 2,401 2,739 3,358 Vegetable & Fruits 214 330 201 151 179 235 259 Textiles and Garments 1,892 1,975 2,752 3,687 4,386 4,838 5,834 Footwear 1,465 1,559 1,867 2,268 2,692 3,040 3,592 Handicraft 237 235 331 367 426 569 630 Wood products 294 335 435 567 1,139 1,563 1,933 Electronic and Computer parts 788 709 605 855 1,075 1,427 1,708 Electric cables and wires 130 154 188 292 389 523 705 Plastic products 100 134 153 154 261 350 480 Source : GSO (2007) Table 3.3: MAJOR IMPORTS BY COMMODITY (US$ million) Prel 2000 2001 2002 2003 2004 2005 2006 Total Imports 15,637 16,162 19,733 25,227 31,954 36,978 44,891 Petroleum products 2,058 1,828 2,017 2,433 3,574 5,024 5,970 Quantity (000 tons) 8,777 8,998 9,966 9,995 11,050 11,478 11,213 Average Unit Value (US$/ton) 234 203 202 243 323 438 532 Fertilizers 509 404 477 628 824 641 687 Quantity (000 tons) 3,973 3,189 3,824 4,119 4,079 2,877 3,189 Average Unit Value (US$/ton) 128 127 125 152 202 223 216 Steel and Irons 812 965 1,334 1,657 2,573 2,931 2,936 Quantity (000 tons) 2,868 3,938 4,951 4,574 5,186 5,524 5,707 Average Unit Value (US$/ton) 283 245 269 362 496 531 515 Others Machinery and Equipment 2,571 2,741 3,793 5,359 5,249 5,281 6,628 Textile fiber and yarn 231 247 314 298 339 340 544 Cotton 101 132 97 106 190 167 219 Material for garment & footwear 1,422 1,590 1,711 2,034 2,253 2,282 1,951 Motor vehicles and pats 238 433 604 834 904 1,193 672 Motorbikes 787 670 422 329 452 541 557 Pharmaceutical material 62 69 83 76 100 116 133 Medicine 325 296 320 374 410 502 548 Paper of all kinds 142 159 193 230 248 362 475 Chemicals 307 352 406 510 683 865 1,042 Chemical products 304 361 482 582 706 841 1,007 Plastic materials 480 495 617 785 1,191 1,456 1,866 Computer and Electronic components 882 666 664 975 1,342 1,706 2,048 Wood - sawn and log 158 163 179 274 539 651 775 Milk and dairy products 141 247 122 164 206 311 321 Animal feed and materials 159 179 234 421 475 594 737 Source : GSO (2007) Table 4.1: MONETARY SURVEY ACCOUNT Rev Prel 2000 2001 2002 2003 2004 2005 2006 (in VND trillion, end of period) Net Foreign Assets 95.7 117.6 117.4 131.4 145.8 191.1 287.9 Foreign assets 112.7 135.9 135.9 150.5 172.3 220.5 327.0 Foreign liabilities -17.0 -18.3 -18.4 -19.1 -26.4 -29.4 -39.1 Net Domestic Assets 127.2 162.2 211.7 279.8 390.3 499.6 634.7 Domestic credit 155.2 191.2 239.9 316.9 435.2 585.6 730.3 Net claims on government -0.5 2.1 8.8 20.1 14.9 32.5 36.5 Credit to the economy 155.7 189.1 231.1 296.7 420.3 553.1 693.8 Claims on state enterprises 69.9 79.7 89.5 105.4 142.9 181.3 218.5 Claims on other sectors 85.8 109.4 141.6 191.3 277.4 371.8 475.3 Other items, net -28.0 -29.0 -28.2 -37.0 -44.9 -86.0 -95.6 Broad money (M2) 222.9 279.8 329.1 411.2 536.2 690.7 922.7 of which: total deposit 170.7 213.5 254.9 320.6 427.1 559.5 763.9 Dong liquidity 152.5 191.1 235.5 314.1 408.1 531.5 723.2 Currency outside banks 52.2 66.3 74.3 90.6 109.1 131.2 158.8 Deposits 100.3 124.8 161.2 223.6 299.0 400.3 564.4 Foreign currency deposits 70.4 88.7 93.6 97.1 128.1 159.2 199.5 (Annual change in percent ) Net Foreign Assets 56.4 22.9 -0.2 11.9 11.0 31.0 50.7 Net Domestic Assets 28.2 27.5 30.5 32.2 39.5 28.0 27.0 Domestic credit 34.1 23.2 25.5 32.1 37.4 34.5 24.7 Credit to the economy 38.2 21.5 22.2 28.4 41.7 31.6 25.4 Claims on state enterprises 28.7 14.0 12.3 17.8 35.6 26.9 20.5 Claims on other sectors 46.9 27.5 29.4 35.1 45.0 34.0 27.8 Total liquidity 38.9 25.5 17.6 25.0 30.4 28.8 33.6 of which: total deposits 43.3 25.1 19.4 25.8 33.2 31.0 36.5 Dong liquidity 30.7 25.3 23.2 33.4 29.9 30.2 36.1 Currency outside banks 26.1 27.0 12.0 22.0 20.4 20.2 21.1 Deposits 33.2 24.4 29.2 38.7 33.7 33.9 41.0 Foreign currency deposits 60.7 26.0 5.6 3.7 32.0 24.3 25.3 Note : Data from 1999 onwards comprise the SBV, six SOCBs and 83 non-state banks Source : SBV and IMF TABLE 5.1: STATE BUDGET REVENUES (VND billion) final account rev est 2000 2001 2002 2003 2004 2005 2006 A Total revenues and grants 90,794 103,888 121,716 158,057 198,614 217,080 264,261 I Current revenues 87,928 100,918 118,346 145,823 180,197 199,953 244,043 II Taxes 79,497 91,688 106,154 127,948 155,579 188,119 230,565 1 Corporate income tax 28,950 33,298 36,826 47,410 56,987 71,738 100,820 2 Individual income tax 1,831 2,058 2,338 2,951 3,521 4,238 5,181 3 Land and housing tax 366 330 336 359 438 514 592 4 License tax 381 400 407 778 657 724 794 5 Tax on the transfer of properties 934 1,191 1,332 1,817 2,607 2,797 3,363 6 Tax on land use right transfer 213 298 327 408 640 985 1,250 7 Value added tax 17,072 19,327 25,916 33,130 38,814 45,728 54,773 8 Special consumption tax 5,250 6,229 7,272 8,851 12,773 15,702 17,144 9 Natural resources tax 7,487 8,416 8,543 9,719 17,398 21,916 20,232 10 Agricultural tax 1,776 814 772 151 130 132 120 11 Export & import tax 13,437 17,458 21,915 22,374 21,614 23,645 26,296 12 Other taxes 158 158 170 III Fees, charges and non-tax 8,431 9,230 12,192 17,875 24,618 11,834 13,478 13 Revenue from discrepancy of import prices 131 116 168 133 40 1 0 14 Fees and charges (include gasoline fee) 4,950 5,120 6,016 6,483 7,765 7,055 8,008 15 Rental of land 568 570 459 513 1,035 1,003 1,596 16 Others 2,782 3,424 5,549 10,746 15,778 3,775 3,874 IV Capital revenues 838 959 1,120 9,265 15,540 14,785 16,600 VIII Grants 2,028 2,011 2,250 2,969 2,877 2,342 3,618 B Carry-over 3,400 2,145 19,353 26,162 7,030 8,510 Source : MOF TABLE 5.2: STATE BUDGET REVENUES (share of GDP) final account rev est 2000 2001 2002 2003 2004 2005 2006 A Total revenues and grants 20.6 21.6 22.7 25.8 27.8 25.9 27.1 I Current revenues 19.9 21.0 22.1 23.8 25.2 23.8 25.1 I.1 Taxes 18.0 19.1 19.8 20.9 21.7 22.4 23.7 1 Corporate income tax 6.6 6.9 6.9 7.7 8.0 8.5 10.4 2 Individual income tax 0.4 0.4 0.4 0.5 0.5 0.5 0.5 3 Land and housing tax 0.1 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 0.1 5 Tax on the transfer of properties 0.2 0.2 0.2 0.3 0.4 0.3 0.3 6 Tax on land use right transfer 0.0 0.1 0.1 0.1 0.1 0.1 0.1 7 Value added tax 3.9 4.0 4.8 5.4 5.4 5.4 5.6 8 Special consumption tax 1.2 1.3 1.4 1.4 1.8 1.9 1.8 9 Natural resources tax 1.7 1.7 1.6 1.6 2.4 2.6 2.1 10 Agricultural tax 0.4 0.2 0.1 0.0 0.0 0.0 0.0 11 Export & import tax 3.0 3.6 4.1 3.6 3.0 2.8 2.7 12 Other taxes 0.0 0.0 0.0 0.0 0.0 0.0 0.0 I.2 Fees, charges and non-tax 1.9 1.9 2.3 2.9 3.4 1.4 1.4 13 Revenue from discrepancy of import prices 0.0 0.0 0.0 0.0 0.0 0.0 0.0 14 Fees and charges 1.1 1.1 1.1 1.1 1.1 0.8 0.8 15 Rental of land 0.1 0.1 0.1 0.1 0.1 0.1 0.2 16 Others 0.6 0.7 1.0 1.8 2.2 0.4 0.4 II Capital revenues 0.2 0.2 0.2 1.5 2.2 1.8 1.7 III Grants 0.5 0.4 0.4 0.5 0.4 0.3 0.4 B Carry-over 0.0 0.7 0.4 3.2 3.7 0.8 0.9 Source : MOF Table 5.3: STATE BUDGET EXPENDITURES (VND billion) final account rev est 2000 2001 2002 2003 2004 2005 2006 A Total expenditures 99,751 117,285 129,434 162,150 187,353 226,935 267,575 I Current expenditures 70,127 77,049 84,216 102,521 121,238 154,978 181,491 1 Administration expenditure 8,089 8,734 8,599 11,359 15,901 16,770 18,994 2 Expenditure on economic affairs & services 5,796 6,288 7,987 8,164 10,301 12,843 15,010 3 Social expenditures 30,694 37,369 40,747 50,185 55,185 77,312 91,409 3.1 Education 9,910 12,006 13,758 17,390 20,401 29,120 33,822 3.2 Training 2,767 3,426 4,086 5,491 4,942 7,280 8,376 3.3 Health 3,453 4,211 4,656 5,372 6,009 10,749 12,685 3.4 Science, technology & environment 1,243 1,625 1,852 1,853 2,362 2,620 3,235 3.5 Culture 919 921 1,066 1,258 1,584 1,718 2,024 3.6 Radio and television 717 838 681 1,056 1,325 1,127 1,140 3.7 Sports 387 483 586 648 883 675 943 3.8 Population and family planning 559 434 841 666 397 410 533 3.9 Social subsidies 10,739 13,425 13,221 16,451 17,282 23,613 28,651 4 Interest payment 3,514 4,485 5,330 6,395 7,217 7,000 8,913 5 Defence 13,058 14,409 18,463 22,892 6 Public security 5,745 6,576 8,432 11,150 7 Others 22,034 20,173 21,553 7,615 11,649 14,158 13,123 II Investment expenditure 29,624 40,236 45,218 59,629 66,115 71,957 86,084 1 Capital expenditure 26,211 36,139 40,740 54,430 61,746 66,799 81,730 2 Others 3,413 4,097 4,478 5,199 4,369 5,158 4,354 B Carry-over 3,400 2,145 4,443 16,390 34,439 10,475 22,515 Source : MOF Table 5.4: STATE BUDGET EXPENDITURES (share of GDP) final account rev est 2000 2001 2002 2003 2004 2005 2006 A Total expenditures 22.6 24.4 24.2 26.4 26.2 27.0 27.5 I Current expenditures 15.9 16.0 15.7 16.7 16.9 18.5 18.6 1 Administration expenditure 1.8 1.8 1.6 1.9 2.2 2.0 2.0 2 Expenditure on economic affairs & services 1.3 1.3 1.5 1.3 1.4 1.5 1.5 3 Social expenditures 6.9 7.8 7.6 8.2 7.7 9.2 9.4 3.1 Education 2.2 2.5 2.6 2.8 2.9 3.5 3.5 3.2 Training 0.6 0.7 0.8 0.9 0.7 0.9 0.9 3.3 Health 0.8 0.9 0.9 0.9 0.8 1.3 1.3 3.4 Science, technology & environment 0.3 0.3 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 0.2 3.6 Radio and television 0.2 0.2 0.1 0.2 0.2 0.1 0.1 3.7 Sports 0.1 0.1 0.1 0.1 0.1 0.1 0.1 3.8 Population and family planning 0.1 0.1 0.2 0.1 0.1 0.0 0.1 3.9 Social subsidies 2.4 2.8 2.5 2.7 2.4 2.8 2.9 4 Interest payment 0.8 0.9 1.0 1.0 1.0 0.8 0.9 5 Defence 2.1 2.0 2.2 2.4 6 Public security 0.9 0.9 1.0 1.1 7 Others (including salary increase in 2007) 5.0 4.2 4.0 1.2 1.6 1.7 1.3 II Investment expenditure 6.7 8.4 8.4 9.7 9.2 8.6 8.8 1 Capital expenditure 5.9 7.5 7.6 8.9 8.6 8.0 8.4 2 Others 0.8 0.9 0.8 0.8 0.6 0.6 0.4 B Carry-over 0.8 0.4 0.8 2.7 4.8 1.2 2.3 Source : MOF Table 5.5: EXTERNAL DEBT (US$ million, unless otherwise indicated) rev est 2002 2003 2004 2005 2006 Public and Publicily Guaranteed 9,413 11,383 13,505 14,208 15,641 Official Creditors Multilaterals 3,256 4,510 5,323 5,540 6,149 Of Which IDA 1,100 1,476 1,744 1,780 2,010 Bilaterals 5,427 6,142 7,294 7,070 7,772 Private Creditors 730 731 888 1,598 1,721 Bonds 382 382 382 1,113 1,095 Commercial Banks 184 184 350 362 516 Other Private 165 165 156 122 110 Total Long-Term DOD 9,413 11,383 13,505 14,208 15,641 Disbursement 986 1,749 1,839 2,246 1,477 Payment 849 776 612 698 765 Principal 642 573 327 435 436 Interest 207 202 285 263 329 Source : MOF Table 6.1A: MONTHLY CHANGE IN CONSUMER RETAIL PRICES (in percent) Month/ Year 1999 2000 2001 2002 2003 2004 2005 January 1.7 0.4 0.3 1.1 0.9 1.1 1.1 February 1.9 1.6 0.4 2.2 2.2 3.0 2.5 March -0.7 -1.1 -0.7 -0.8 -0.6 0.8 0.1 April -0.6 -0.7 -0.5 0.0 0.0 0.5 0.6 May -0.4 -0.6 -0.2 0.3 -0.1 0.9 0.5 June -0.3 -0.5 0.0 0.1 -0.3 0.8 0.4 July -0.4 -0.6 -0.2 -0.1 -0.3 0.5 0.4 August -0.4 0.1 0.0 0.0 -0.1 0.6 0.4 September -0.6 -0.2 0.5 0.2 0.1 0.3 0.8 October -1.0 0.1 0.0 0.3 -0.2 0.0 0.4 November 0.4 0.9 0.2 0.3 0.6 0.2 0.4 December 0.5 0.1 1.0 0.3 0.8 0.6 0.8 Source : GSO (2007) Table 6.1B: MONTHLY CONSUMER RETAIL PRICE INDEX (Index, January 1995 = 100) Month/ Year 1999 2000 2001 2002 2003 2004 2005 January 130.7 129.1 128.3 130.4 135.3 139.6 152.9 February 133.2 131.2 128.8 133.2 138.2 143.8 156.7 March 132.3 129.8 127.9 132.2 137.4 144.9 156.8 April 131.5 128.9 127.3 132.2 137.4 145.6 157.7 May 131.0 128.1 127.0 132.6 137.2 146.9 158.5 June 130.6 127.4 127.0 132.7 136.8 148.1 159.1 July 130.1 126.7 126.8 132.6 136.4 148.8 159.8 August 129.5 126.8 126.8 132.6 136.3 149.6 160.3 September 128.8 126.6 127.4 132.8 136.4 150.1 161.5 October 127.5 126.7 127.4 133.2 136.1 150.1 162.2 November 128.0 127.8 127.7 133.7 137.0 150.4 162.7 December 128.6 127.9 129.0 134.1 138.0 151.2 164.0 Annual Index 130.2 128.1 127.6 132.7 136.9 147.4 159.3 Annual Growth Rate 4.3 -1.6 -0.4 4.0 3.2 7.7 8.1 Dec/ Dec Growth Rate 0.1 -0.5 0.8 4.0 3.0 9.5 8.5 Source: GSO (2007) Table 6.2A: PRICE INDEX BY COMMODITY GROUPS (monthly change) GOODS and SERVICES Jan-06 Feb-06 Mar-06 Apr-06 May-06 Jun-06 Jul-06 Aug-06 Sep-06 Oct-06 Nov-06 Dec-06 General Index 1.2 2.1 -0.5 0.2 0.6 0.4 0.4 0.4 0.3 0.2 0.6 0.5 Food & foodstuff 1.7 3.2 -0.9 0.2 0.2 0.4 0.4 0.2 -0.2 0.5 1.2 0.7 of which: Food 2.1 1.7 0.2 0.3 0.3 0.2 0.4 0.3 0.4 1.3 3.8 2.4 Foodstuff 1.4 3.7 -1.1 0.2 0.1 0.4 0.5 0.1 -0.7 0.2 0.5 0.1 Beverage & tobacco 1.2 1.5 -0.5 0.1 0.4 0.3 0.6 0.2 0.3 0.3 0.8 -0.1 Garment, hats, footwear 0.9 0.8 0.3 0.2 0.4 0.4 0.3 0.5 0.7 0.4 0.0 0.7 Housing & construction materials 0.8 0.7 0.1 -0.3 0.9 1.0 0.8 1.1 0.9 -0.5 -0.2 0.5 Household appliancies 0.6 0.6 0.4 0.4 0.5 0.7 0.3 0.3 0.4 0.5 0.7 0.7 Healthcare, pharmaceutical items 0.5 0.5 0.5 0.1 0.4 0.5 0.3 0.3 0.2 0.6 0.2 0.2 Transport & Telecommunication 0.1 0.8 -0.6 0.1 2.9 0.1 0.2 0.9 0.7 -1.8 -0.5 0.1 Educational items 0.1 0.1 0.1 0.1 0.3 -0.2 0.3 0.3 0.7 1.6 0.1 0.1 Cultural and recreation items 0.2 1.9 -0.7 -0.2 0.9 0.0 0.0 0.3 0.4 0.2 0.4 0.2 Goods and other services 0.8 1.7 -0.4 0.2 1.0 -0.3 0.5 0.2 0.0 0.9 1.0 0.8 Gold 4.0 5.4 1.8 4.8 17.6 -5.6 -3.1 2.3 -2.9 -3.1 1.7 3.2 US Dollar 0.0 0.1 -0.1 0.1 0.8 -0.3 -0.2 0.1 0.1 0.2 0.2 0.0 Source : GSO (2007) TABLE 6.2B: PRICE INDEX BY COMMODITY GROUPS (Index, December 2005 = 100) GOODS and SERVICES Jan-06 Feb-06 Mar-06 Apr-06 May-06 Jun-06 Jul-06 Aug-06 Sep-06 Oct-06 Nov-06 Dec-06 General Index 101.2 103.3 102.8 103.0 103.6 104.0 104.5 104.9 105.2 105.4 106.0 106.6 Food & foodstuff 101.7 105.0 104.0 104.2 104.4 104.8 105.3 105.5 105.3 105.8 107.1 107.8 of which: Food 102.1 103.8 104.0 104.4 104.7 104.9 105.3 105.6 106.0 107.4 111.5 114.2 Foodstuff 101.4 105.2 104.0 104.2 104.3 104.7 105.2 105.4 104.6 104.8 105.3 105.5 Beverage & tobacco 101.2 102.7 102.2 102.3 102.7 103.0 103.6 103.8 104.2 104.5 105.3 106.0 Garment, hats, footwear 100.9 101.7 102.0 102.2 102.6 103.0 103.3 103.9 104.6 105.0 105.0 105.5 Housing & construction materials 100.8 101.5 101.6 101.3 102.2 103.2 104.1 105.2 106.2 105.6 105.4 106.1 Household appliancies 100.6 101.2 101.6 102.0 102.5 103.2 103.6 103.9 104.3 104.8 105.5 105.7 Healthcare, pharmaceutical items 100.5 101.0 101.5 101.6 102.0 102.5 102.8 103.1 103.3 104.0 104.2 104.3 Transport & Telecommunication 100.1 100.9 100.3 100.4 103.3 103.4 103.6 104.5 105.3 103.4 102.9 103.0 Educational items 100.1 100.2 100.3 100.4 100.7 100.5 100.8 101.1 101.8 103.4 103.5 103.8 Cultural and recreation items 100.2 102.1 101.4 101.2 102.1 102.1 102.1 102.4 102.8 103.0 103.4 104.3 Goods and other services 100.8 102.5 102.1 102.3 103.3 103.0 103.5 103.7 103.7 104.7 105.7 105.7 Gold 104.0 109.6 111.6 116.9 137.5 129.8 125.8 128.7 125.0 121.1 123.1 127.1 US Dollar 100.0 100.1 100.0 100.1 100.9 100.6 100.4 100.5 100.6 100.8 101.0 101.0 Source : GSO (2007) Table 7.1: AGRICULTURAL PRODUCTION (VND billion at current prices) Rev 2000 2001 2002 2003 2004 2005 Gross Output 129,141 130,178 145,021 153,955 172,495 183,342 Crop Cultivation 101,044 101,403 111,172 116,066 131,552 134,755 Livestock 24,960 25,501 30,575 34,457 37,344 45,226 Services 3,137 3,273 3,275 3,433 3,599 3,362 Source : GSO (2007) Table 7.2: AGRICULTURAL PRODUCTION (VND billion at constant 1994 prices) Rev 2000 2001 2002 2003 2004 2005 Gross Output 112,112 118,990 122,150 127,628 132,888 137,112 Crop Cultivation 90,858 92,907 98,061 101,763 106,423 107,898 Food Crops 55,163 55,066 59,619 61,029 63,621 63,853 Industrial Crops 21,782 23,109 22,247 24,175 25,612 8,928 Livestock 18,505 19,283 21,200 22,907 23,439 26,108 Services 2,748 2,800 2,890 2,958 3,027 3,107 Memorandum Items: Paddy Ouput (000 tons) 32,530 32,108 34,447 34,569 39,581 39,622 Cultivated Area (000 ha) 7,666 7,493 7,504 7,452 7,445 7,329 Yield (ton/ ha) 4.24 4.29 4.59 4.64 5.32 5.41 Source : GSO (2007) Table 7.3: INDUSTRIAL CROP PRODUCTION AND YIELDS Rev Prel 2000 2001 2002 2003 2004 2005 2006 Production (000 metric tons) Cotton 19 34 40 35 28 34 26 Jute 11 15 20 12 13 13 11 Sedge 61 65 88 96 90 81 93 Sugar cane 15,044 14,657 17,120 16,855 15,649 14,949 15,679 Peanut 355 363 400 406 469 489 465 Soya-beans 149 174 206 220 246 293 258 Tobacco 27 32 33 32 23 26 43 Tea - raw and fresh 315 340 424 449 514 570 612 Coffee 803 841 700 794 836 752 854 Rubber 291 313 298 364 419 482 546 Black pepper 39 44 47 69 73 80 83 Coconut 885 892 915 893 960 977 982 Area Cultivated (000 ha) Cotton 19 28 34 28 28 26 21 Jute 6 8 10 5 5 6 6 Sedge 9 10 12 14 13 13 13 Sugar cane 302 291 320 313 286 266 285 Peanut 245 245 247 244 264 270 249 Soya-beans 124 140 159 166 184 204 186 Tobacco 24 24 27 23 16 17 27 Tea 88 98 109 116 121 123 123 Coffee 562 565 522 510 497 497.4 488.6 Rubber 412 416 429 441 454 482.7 511.9 Black pepper 28 36 48 51 51 49.1 48.5 Coconut 161 156 140 134 133 132 133 Average Yield (metric ton/ ha) Cotton 1.0 1.2 1.2 1.3 1.0 1.3 1.3 Jute 2.1 1.9 2.1 2.6 2.6 2.3 1.8 Sedge 6.6 6.6 7.2 6.8 6.9 6.4 7.4 Sugar cane 49.8 50.4 53.5 53.8 54.7 56.1 55.0 Peanut 1.5 1.5 1.6 1.7 1.8 1.8 1.9 Soya-beans 1.2 1.2 1.3 1.3 1.3 1.4 1.4 Tobacco 1.1 1.3 1.2 1.4 1.4 1.5 1.6 Tea - raw and fresh 3.6 3.5 3.9 3.9 4.3 4.7 5.0 Coffee 1.4 1.5 1.3 1.6 1.7 1.5 1.7 Rubber 0.7 0.8 0.7 0.8 0.9 1.0 1.1 Black pepper 1.4 1.2 1.0 1.4 1.4 1.6 1.7 Coconut 5.5 5.7 6.5 6.7 7.2 7.4 7.4 Source : GSO (2007) Table 8.1: INDUSTRIAL PRODUCTION OUTPUT (VND billion at constant 1994 prices) Rev Prel 2000 2001 2002 2003 2004 2005 2006 Gross Industrial Output 198,326 227,342 261,092 305,080 355,624 416,563 487,492 State sector 82,897 93,434 105,119 117,637 131,655 141,117 154,231 Central 54,962 62,119 69,640 80,917 92,896 104,372 116,751 Local 27,935 31,316 35,479 36,720 38,759 36,745 37,480 Non-state sector 44,144 53,647 63,474 78,292 95,785 120,127 148,783 Collectives 1,334 1,575 1,668 1,770 1,893 1,969 2,032 Private, households and mixed 42,810 52,072 61,807 76,522 93,892 118,158 146,750 Foreign-invested sector 71,285 80,261 92,499 109,152 128,184 155,319 184,479 Key Industries Coal 2,366 2,695 3,189 3,689 4,752 6,111 7,798 Oil and gas 22,746 23,766 23,817 25,132 28,403 27,410 25,768 Mining and metal ores 209 239 281 344 467 476 384 Stones and other mining 2,015 2,398 3,039 3,597 3,842 4,354 4,900 Food and beverage 43,634 50,373 56,061 64,585 74,694 84,482 99,452 Cigarettes and tobacco 5,744 6,690 7,658 9,189 10,160 11,234 12,316 Textile products 10,046 10,641 12,338 14,214 16,626 19,079 22,176 Garment - apparel 6,042 6,862 8,182 10,466 12,792 15,304 18,463 Leather tanning and processing 8,851 9,529 11,096 13,535 16,018 18,920 22,190 Wood and wood products 3,598 3,903 4,488 5,485 6,570 8,120 10,282 Paper and paper products 3,930 4,562 4,877 5,655 7,140 8,311 9,416 Printing and publishing 2,274 2,453 2,876 3,515 3,774 4,626 5,945 Chemicals 11,123 12,852 14,714 16,323 19,029 23,848 28,426 Rubber products and plastic 6,456 8,128 9,706 11,291 15,169 18,237 21,940 Non-metallic products 18,259 21,625 25,913 29,855 33,483 37,055 41,713 Metal production 5,914 6,842 8,516 10,430 11,226 13,949 16,882 Metallic products 5,768 7,063 8,506 10,646 12,963 17,595 22,159 Machinery and equipment 2,761 3,421 3,711 4,612 5,371 5,495 5,718 Computer and office equipment 1,295 977 1,003 1,538 1,846 3,206 4,645 Electric and electronic equipments 3,622 5,172 6,520 7,462 9,050 11,992 15,638 Radio, TV and telecom 4,395 5,407 6,169 7,162 7,956 9,137 10,628 Production & repairing motor vehicles 3,232 4,265 5,774 8,306 8,692 9,753 12,030 Production & repairing other transport means 6,414 7,090 8,534 9,676 12,172 15,834 20,077 Furnitures 3,931 4,759 6,057 7,846 10,179 13,411 18,155 Recycled products 150 151 174 204 261 267 287 Electricity and gas 11,828 13,551 15,741 18,071 20,385 23,427 26,798 Water supply 1,066 1,152 1,328 1,361 1,409 1,570 1,701 Source : GSO (2007) Table 8.2: MAJOR INDUSTRIAL PRODUCTS (VND billion at constant 1994 prices) Rev Prel Product Unit 2000 2001 2002 2003 2004 2005 2006 Assembled automobiles unit 13,547 20,526 29,536 47,701 50,954 59,152 41,557 Assembled motorbikes 000' unit 643 610 1,052 1,180 1,828 1,982 2,094 Assembled tivi sets 1,000 1,013 1,126 1,597 2,188 2,660 2,515 2,282 Beverage mil. liters 779 971 940 1,119 1,343 1,461 1,548 Bicycle tires 000 Pieces 20,675 21,656 22,778 26,686 26,008 20,387 23,455 Bicycle tubes 000 pieces 21,917 22,997 24,032 36,083 32,386 26,848 31,625 Bricks mil. pieces 9,087 9,811 11,365 12,810 14,661 16,530 19,893 Cement 000 tons 13,298 16,073 21,121 24,127 26,153 30,808 32,690 Chemical fertilizers 000 tons 1,210 1,270 1,158 1,294 1,714 2,190 2,176 Cigarettes mil. packs 2,836 3,075 3,375 3,871 4,192 4,485 4,030 Coal mil. tons 11.6 13.4 16.4 19.3 27.3 34.1 38.9 Crude oil mil. tons 16.3 16.8 16.9 17.7 20.1 18.5 17.2 Diesel engines Pieces 15,623 18,721 32,570 184,418 182,443 201,593 171,508 Electric engines Pieces 45,855 53,442 64,085 95,779 132,320 194,374 215,010 Electricity mil. kWh. 26,682 30,673 35,888 40,546 46,202 52,078 59,050 Fabrics of all kinds mil. meters 356 410 470 496 502 561 575 Glass products 000 tons 113 115 115 147 154 163 168 Insecticides 000 tons 20.1 20.0 20.7 40.9 54,523 45,877 50,285 Paper and paper products 000 tons 408 445 490 687 809 901 997 Porcelain mil. pieces 247 314 284 524 404 514 551 Rice mill equipment pieces 12,484 18,298 13,433 10,112 5,749 2,734 2,960 Salt 000 tons 590 699 974 909 906 898 955 Sawn wood 000 m3 1,744 2,036 2,667 3,291 3,009 3,232 3,414 Soap and detergent 000 tons 247 326 361 377 401 421 493 Steel 000 tons 1,583 1,914 2,503 2,954 3,280 3,403 3,827 Sugar (refine) 000 tons 790 739 790 1,073 1,191 1,102 1,032 Tea 000 tons 70 82 100 85 122 127 133 Textile yans 000 tons 130 162 227 235 241 259 265 Tin (billet) Tons 1,803 1,728 1,565 1,915 2,356 1,766 1,830 Transformers pieces 13,535 15,664 18,633 33,364 50,146 45,540 52,430 Water pumps for agriculture pieces 3,496 4,238 3,578 7,787 10,038 8,298 9,730 Source: GSO (2007)