AUSTRALIA-WORLD BANK GROUP STRATEGIC PARTNERSHIP IN VIETNAM Ethnic Minorities Theme Research Report Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam Table of Contents 111 Research Report Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam Copyright © 2019 The International Bank for Reconstruction and Development / The World Bank Group 1818 H Street, NW, Washington, DC 20433 USA All rights reserved. The findings, interpretations, and conclusions expressed in this report are entirely those of the authors and should not be attributed in any manner to the World Bank, or its affiliated organizations, or to members of its board of executive directors or the countries they represent. The World Bank does not guarantee the accuracy of the data included in this publication and accepts no responsibility whatsoever for any consequence of their use. Photo credits: Hung Pham Table of Contents Acronyms and Abbreviations ......................................................................................................................................................................................................................................................................................5 Foreword .................................................................................................................................................................................................................................................................................................................................................................................6 Executive Summary .................................................................................................................................................................................................................................................................................................................................7 1. Introduction ........................................................................................................................................................................................................................................................................................................................................................27 2. Methodology: How the Study Was Conducted? .........................................................................................................................................................................................29 2.1 Quantitative Component ............................................................................................................................................................................................................................................................29 2.2 Qualitative Component ...................................................................................................................................................................................................................................................................30 2.3 Capacity-Building Component .....................................................................................................................................................................................................................................31 3. Who Are the Top- and Bottom-Performing Ethnic Minorities? .......................................................................................................................32 3.1 Ranking Human Development of the Ethnic Minority Groups .................................................................................................32 3.2 Selection of Top- and Bottom-Performing Ethnic Minority Groups .............................................................................35 3.3 Analytical Framework on Driving Forces for Soco-Economic Performances of Different Ethnic Groups ...................................................................................................................................................................................................................................................................37 3.4 Sampling for the Qualitative Research .....................................................................................................................................................................................................40 4. What are the Main Drivers of Socio-Economic Performance Among Ethnic Minority Groups? ........................................................................................................................................................................................................................................................................................................................................................................................44 4.1 Physical and Economic Connectivity .............................................................................................................................................................................................................44 4.2 Basic Household Livelihood Assets ....................................................................................................................................................................................................................47 4.3 Market Linkages ...................................................................................................................................................................................................................................................................................................54 4.4 Labor Market Mobility ..........................................................................................................................................................................................................................................................................57 4.5 Some Aspects of Access to Education 62 .................................................................................................................................................................................................... 4.6 Some Aspects of Healthcare Services ..........................................................................................................................................................................................................65 4.7 Traditional Institutions and Local Governance .................................................................................................................................................................68 4.8 Gender Roles and Intra-Household Power Dynamics .......................................................................................................................................73 4.9 Perceptions of Ethnicity and the Dynamics of Trust .............................................................................................................................................76 4.10 Access to External Support Initiatives ......................................................................................................................................................................................................79 5. Policy Recommendations .............................................................................................................................................................................................................................................................................................85 5.1 How the Existing Policies Have Addressed the Gaps? ......................................................................................................................................85 5.2 Policy Recommendations ...........................................................................................................................................................................................................................................................88 Table of Contents 3 6. Conclusions ...........................................................................................................................................................................................................................................................................................................................................................95 6.1 A Summary of Driving Forces of the Top or Bottom Performers .................................................................................................95 6.2 Main Drivers of Socio-Economic Development Among Ethnic Minority Groups ............................102 6.3 Limitations of the Study and Directions for Further Research ...........................................................................................................103 References ...........................................................................................................................................................................................................................................................................................................................................................................105 Annex. Details of the Study Sites ..............................................................................................................................................................................................................................................................107 TABLES Table 1. Overview of the Selected Top- and Bottom-Performing Ethnic Minority Groups ........................36 Table 2. Brief Profile of the Top and Bottom Performers, 2015 ...........................................................................................................................................37 Table 3. Selection of the Study Sites for the Qualitative Study ............................................................................................................................................42 Table 4. Distances to Main Public Facilities .............................................................................................................................................................................................................................44 Table 5. Household Landholding of the Selected Ethnic Groups 48 ............................................................................................................................. Table 6. Working by Sex and Age Groups of the Selected Ethnic Groups ..............................................................................................51 Table 7. Means of Access to Information .......................................................................................................................................................................................................................................56 Table 8. Ethnic Minority Women with Technical Training and Language Skills ......................................................................59 Table 9. Educational Attainment Levels of the Selected Ethnic Minorities (%) .........................................................................62 Table 10. Access to Healthcare Services ...............................................................................................................................................................................................................................................66 Table 11. Rates of Child Marriage and Consanguineous Marriage 67 ............................................................................................................................. Table 12. Representations of the Selected Ethnic Minority Groups in the Local Administration 70 Table 13. Religion Practices Among Top and Bottom Performers ...................................................................................................................................71 Table 14. Preservation of Traditional Cultures ......................................................................................................................................................................................................................79 Table 15. Summary of How the Factors Affect the Selected Top- and Bottom-Performing Ethnic Minority Groups ...........................................................................................................................................................................................................................................................................96 FIGURES Figure 1. The HDIs of Selected Ethnic Minority Groups in Vietnam ..........................................................................................................................34 Figure 2. The MPIs of Selected Ethnic Minority Groups in Vietnam ............................................................................................................................34 Figure 3. Driving Forces of the Social and Economic Development of Selected Ethnic Minority Groups .........................................................................................................................................................................................................................................................................................................39 Figure 4. Structures of Household Income ..................................................................................................................................................................................................................................57 Figure 5. Income Structures and Educational Attainment of the Household Head .........................................................64 Figure 6. Mapping of How Existing Policies Have Addressed the Gaps ...........................................................................................................86 Research Report 4 Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam Acronyms and Abbreviations ASEAN Association of South East Asian Nations CEMA Central Committee for Ethnic Minority Affairs CIO Commune Investment Ownership CPC Commune People’s Committee DARD Department of Agriculture and Rural Development DoLISA Department of War Invalids, Labour, and Social Affairs DoET Department of Education and Training DoH Department of Health DoCST Department of Culture, Sport, and Tourism 53EMS Survey on the Socio-Economic Situation of the 53 Ethnic Minority Groups in Viet Nam GoVN Government of Vietnam HDI Human Development Index IFAD International Fund for Agricultural Development ILO International Labour Organization MARD Ministry of Agriculture and Rural Development MdP Multidimensional Poverty MoLISA Ministry of War Invalids, Labour, and Social Affairs MPI Multidimensional Poverty Index NGO Non-governmental Organization NTP NRD National Target Program on New Rural Development NTP SPR National Target Program on Sustainable Poverty Reduction O&M Operation and Maintenance PCEM Provincial Committee for Ethnic Minority Affairs PPC Provincial People’s Committee PrPC Provincial People’s Council SCG Saving and Credit Group SEDP Socio-Economic Development Planning UNDP United Nation Development Programme VBSP Vietnam Bank for Social Policies VND Vietnam Dong Acronyms and Abbreviations 5 Foreword The study was commissioned by the World Bank in Vietnam and the Central Committee on Ethnic Minority Affairs (CEMA) within the Australia-World Bank Partnership II. In addition to the main research questions, a related objective of the study was to build and strengthen the research capacity of CEMA personnel, particularly early- and mid-career researchers from the CEMA Academy for Ethnic Minorities. This study report is a joint effort of a selected group of CEMA researchers with technical guidance and supervision from the two consultants for the World Bank. The CEMA Research Team comprised Phan Văn Cương, Phạm Thị Kim Cương, Lý A Chông, Nguyễn Duy Dũng, Trần Thùy Dương, Đinh Thị Hòa, Lý Thị Thu Hằng, Lê Thị Huyền, Vì Lan Phương, Lê Thị Thu Thanh, Nguyễn Thị Thuận, and Vũ Đăng Truyền. The two consultants from the World Bank leading the joint effort are Phạm Thái Hưng and Nguyễn Thu Hương. The authors would like to thank several individuals and organizations who have contributed to this study. We thank CEMA Minister Đỗ Văn Chiến to grant access to the Socio-Economic Survey on 53 Ethnic Minorities; leader of the CEMA’s International Cooperation Department (Mr. Hà Viết Quân), leaders of the CEMA’s Academy of Ethnic Minorities (Mr(s). Trần Trung, Bế Trung Anh) and of some departments of the Academy (Mr Hà Quang Khuê, Ms. Nguyễn Hồng Hải) for commenting on various stages of the study. We would like to express our gratitude to Mr/s(s) Roxanne Hakim, Martin Henry Lenihan, Obert Pimhidzai, Nguyễn Quý Nghị, Đỗ Thu Trang, Vũ Thùy Dung of the World Bank Office in Vietnam for contributing to all stages of this study, from the start to the end. We would like to thank the two peer reviewers of the World Bank; researchers, representatives of the Government of Vietnam’s ministries, representatives of development partners who have provided useful comments and suggestions to this report at various technical workshops organized during the process of this study. The research team would like to express our gratitude to officials in the areas of ethnic minorities, poverty reduction, agriculture and rural development, healthcare, education, finance, planning, home affairs in the seven provinces, 13 districts, and 16 communes that the team has consulted during the fieldwork. Especially, we would like to extend our special gratitude to the representatives of 832 households from different ethnic groups, who kindly agreed to share their thoughts and information that were very useful for us in formulating the research findings presented in this report. Finally, the findings and recommendations of this report are those of the authors; these are not necessarily the views of the World Bank, CEMA, or any organizations, households, or individuals that the research team has consulted during the process of this study. Research Report 6 Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam Executive Summary Introduction 1. While Vietnam has now reached lower middle-income country status, the gaps between the ethnic minority population and the majority group are evident and widening over time. In addition, ethnic minority groups are different in terms of where they are in these gaps. This study attempts to examine why and how certain ethnic groups have managed to rise to the ‘top’ as ‘best performers’ while the other groups seem to stand on the ‘bottom’ as ‘least performers’. The key study questions are: • What are drivers of the socio-economic development of the different ethnic groups? • Why have some ethnic minority groups successfully managed to escape poverty while others have lagged far behind? • How have such factors have been addressed in the respective policies and designated programs or projects initiated by the Government of Vietnam, development partners, and other stakeholders? • What are the changes needed for future design and implementation of initiatives to support sustainable socio-economic development among ethnic minorities? Who are the Top- and Bottom-Performing Ethnic Minorities? 2. This study adopts a mixed methodological approach, combining both quantitative and qualitative methods. In order to identify the top- and bottom-performing ethnic minorities, the 2015 Ethnic Minorities Socio-Economic Survey of 53 groups (53EMS) dataset was used to calculate the Human Development Index (HDI) and Multidimensional Poverty Index (MPI) as two indicators of socio-economic development of the ethnic minorities. The final selection of the top and bottom performers was made after intensive consultation with numerous scholars and experts in the field and World Bank and CEMA representatives. Accordingly, the following 6 ethnic groups are selected: the Mường and Sán Dìu represent the top performers; the Khơ Mú and Mông represent the bottom performers; and the Xơ Đăng were selected as being close to the bottom of the ranking, and the Khmer were selected close to the top of the ranking. Given this selection, the study covers a spectrum of socio-economic development rankings as shown in the following table. Executive Summary 7 Sán Dìu Mường Khmer Xơ Đăng Khơ Mú Mông Ethno Chinese Vietic Mon-Khmer Mon-Khmer Mon-Khmer Mông-Mien language Population Thái Nguyên, Hòa Bình, Sơn Sóc Trăng, Kon Tum, Nghệ An, Điện Hà Giang, density Vĩnh Phúc, La, Thanh Trà Vinh, An Quảng Nam, Biên, Sơn La, Tuyên Quang, Bắc Giang, Hóa, Phú Thọ, Giang, Cà Quảng Ngãi Lai Châu Lào Cai, Yên Quảng Ninh, Ninh Bình Mau, Bạc Liêu Bái, Lai Châu, Tuyên Quang Sơn La, Cao Bằng Population 163,650 1,378,938 1,279,567 195,673 84,344 1,244,137 HDI 0.5868 0.5679 0.524 0.4494 0.4140 0.3852 MPI 27.09 40.77 34.5 82.83 91.72 88.17 Research Thái Nguyên Sơn La (Phù Sóc Trăng (Mỹ Kon Tum (Đắk Sơn La (Mai Sơn La (Bắc sites (Đồng Hỷ/ Yên/Mường Tú/Phú Mỹ; Tô/Kon Đào; Sơn/Hát Yên/Xím Nam Hòa; Phú Thải), Thanh Trần Đề/Đại Ngọc Hồi/ Lót), Thanh Vàng), Thái Bình/Bàn Đạt) Hóa (Ngọc An 2), Trà Vinh Bờ Y); Quảng Hóa (Mường Nguyên Lạc/Thạch (Trà Cú/ Định Nam (Bắc Trà Lát/Tén Tằn (Đồng Hỷ/Văn Lập), Kon Tum An) My/Trà Đốc) & Mường Lang), Thanh (Ngọc Hồi/ Chanh) Hóa (Mường Bờ Y) Lát/Pù Nhi) 3. The identification of the best- and least-performing group served as the guiding principle for the selection of the qualitative study sites. Ethnographical fieldwork was conducted between April and July 2018 in 7 selected provinces, 13 districts, 16 communes, and 32 villages. In total, the study team consulted 105 stakeholders at the provincial level, 260 at the district level, 192 at the commune level, and 832 at the village and household levels (of which 48.5 percent are female), making a total of around 1,389 persons in the consultation sample. 4. The findings of this study indicate an intricate web of interrelated conditions and factors that significantly contribute to the route out of poverty among selected ethnic minorities in Vietnam. For analysis purpose, these factors are grouped into the following categories: (a) physical and economic connectivity; (b) basic household livelihood assets (productive landholdings, household labor force, vulnerabilities and shocks); (c) market linkages; (d) labor market mobility; (e) access to education; (f) health care services; (g) role of traditional institutions and local governance; (h) perceived gender roles and intra-household power dynamics; (i) perceptions of ethnicity and the dynamics of trust; and (j) access to external support initiatives. What Factors Drive the Social and Economic Development of Different Ethnic Minority Groups? 5. A summary follows of the key factors having an impact on the performances of different ethnic groups. Research Report 8 Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam (a) Physical and Economic Connectivity 6. Top performers are better connected. Physical connectivity is found to attribute to the better performing Mường, Sán Dìu, and Khmer groups, with access to reduced transportation costs, improved access to markets, and other economic opportunities with the mainstream society. This gives weight to economic connectivity to urban areas or industrial parks, where ethnic minority youth can find opportunities in the wage labor market, or ethnic minority farmers can participate in some market value chains. This connectivity to industrial clusters helps enable many ethnic minority women to commute to work, which would otherwise be impossible for them. Since distance is a factor that impacts local perceptions about women taking up paid work outside their homes, it influences how households weigh the options allowing women to seek work away from the area where they live. The further away the jobs, the less likely women are to pursue them. There are differences between the top performers and bottom performers in this regard. Among the top-performing Sán Dìu, Mường, and Khmer, the distance factor does not really play an important role in women’s participation in the labor market. 7. Poor connectivity is associated with the bottom performers. Notably, the Mông appear to be the worst connected compared to all other selected groups. In addition to poor physical connectivity, the bottom-performing groups experience poor economic connectivity. The geographic distance should be seen in a specific regional context. The notion of distance and/or level of remoteness among various groups varies and does not necessarily reflect the realities on the ground, considering factors like weather or the current state of infrastructure. This suggests that one should not consider distance alone in terms of physical geography but should include economic, social and cultural elements as well. (b) Basic Household Livelihood Assets 8. In terms of arable land possession, the bottom-performing groups are better endowed than the top-performing ones. The average landholdings of the Khơ Mú and Xơ Đăng are nearly double those of the Mường or the Khmer. The Sán Dìu possesses the smallest amount of arable land. Earnings from productive lands contribute significantly to poverty reduction. Possession of fertile land is an important condition for some groups to raise to the top. Evidence from the qualitative research generally suggests that the Mường and Sán Dìu appear to be well endowed in terms of fertile landholdings. 9. Among the bottom performers, productive lands are limited. Lack of cultivation land is seen as the main cause of poverty that passes from one generation to the next (also referred to as sustained poverty or inherited poverty). This in turn limits the opportunity or access to livelihood diversification, leaving those with sufficient land to invest in, for example, shrimp farming or raising special crops. Lack of finance is a barrier for poor farmers to make more efficient use of their land. However, there are still ways for better-off groups (within the bottom performing)1 to expand their 1 Implicit here is the recognition that there are a number of better-off households among the bottom performing groups. In this report, ‘better-off’ households refer to those who are found to be wealthier compared to others in the community, regardless whether they belong to the top- or bottom-performing groups. Executive Summary 9 arable land, (i.e., through forest clearance, soil restoration or creating terrace fields, or by leasing the land of their brothers or brothers in law but not from outsiders). Land productivity among the bottom performers also depends on whether they have deliberately learned the cultivation techniques of the more technologically advanced neighboring ethnic groups such as the Thái, Tày, Kinh, or Hoa. 10. Productive landholdings correlate with displacement and resettlement experiences in the past, varying from top performers to the bottom performers. The Mường migrants in Kon Tum not only benefited from considerable state resettlement assistance but were able to utilize large areas of forest land in the early stages due to their cultural affinities with the Kinh and their expertise in paddy and fruit tree cultivation. Whereas the Khơ Mú and Xơ Đăng were resettled in fixed hamlets under the government sedentarization policy and were given land to take up paddy rice cultivation. However, a number of them were unwilling to adapt to the new mode of livelihood and sold their allotted land to the migrant Kinh and resumed their traditional technique of swidden, slash- and-burn cultivation for their livelihood. 11. Adequate labor for the production process is another important factor that helps contribute to some ethnic minority groups becoming top performers. The bottom performers have larger household sizes compared to the top performers. The hardships that the bottom- performing groups have encountered probably force all of the adult family members to seek some outside income-generating work. However, ethnic minorities mainly find employment in simple or manual labor jobs, and limited occupation in the management labor sector. 12. Economic deprivation among households seems to be proportionate to the number of children having not yet reached working age. This situation is often observed in some ethnic groups having a high birth rate due to the absence of birth control for religious reasons like the Xơ Đăng in Kon Tum, or because of preference for sons over daughters among the Mông in Sơn La, Thái Nguyên, and Thanh Hóa. This helps explain why households considered to be categorized as ‘poor’ are often those having children still of school age and thus below legal working age and illegible for paid jobs. 13. Market price fluctuations represent a major type of shock that ethnic minorities have encountered in their major livelihood activities. This has been seen with decreased prices in maize among the Mông, price fluctuation among the Khơ Mú farmers, decreased prices in tea leaves among the Sán Dìu, and decreased price of sweet potatoes reported by the Khmer reported. In terms of livestock, many of the households from all 6 studied ethnic minority groups reported that pig raising, which was one of the most popular livestock options a few years ago, was no longer the choice (even for some local varieties that could still find niche markets) due to sharp reduction in prices in recent years. 14. The market shock exerts negative influence on all studied ethnic groups but the top and bottom performers responded and coped with the shock differently. For the best- performing Mường, decreasing prices of maize, pigs, or tea induced the switch to other crops such as orange and pomelo. Whereas the interviewed Khơ Mú and Mông households had difficulties in switching to other crops given their poor land quality and their lack of technical knowledge. Among Research Report 10 Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam the least performers, the practice of exchanging labor is common. In the context of the market economy, the mobilization of supplementary labor force from kinship may help keep production costs down but may not help increase productivity since these family members may lack the skills provided by labor force procured from the labor market. 15. The bottom-performing groups are more prone to suffer hardships from natural disasters. Natural disasters have been found not to be as such a major risk to the top-performing Sán Dìu or Mường. At many Mông locations, the risk of landslides under heavy rain is prominent. Serious drought was also reported by both the Mông and Khơ Mú, and to some extent by Xơ Đăng in Kon Tum. The bottom-performing groups appear to have less capacity to cope with natural disasters compared to the top-performing groups that rely on their generally solid social capital built with family and other relations, which can be of great importance in post-disaster relief and reconstruction. Suffering from natural disaster, many bottom-performing households fell into indebtedness to local money lenders. Unable to pay back their debts at very high interest rates, these farmers often had to sell their production resources, including land, thereby depriving themselves of their very livelihoods. The high dependency on natural resources as a major livelihood strategy also makes some bottom-performing groups more vulnerable economically. (c) Market Linkages 16. The best-performing groups appear to be better linked to the market and vice versa. Connectivity clearly contributes to market linkages. That said, the degree of market power exercised by traders or shop owners, especially those who happen to be private money lenders, appears to be a major impediment to market linkages and livelihood strategies of the least-performing groups. 17. From a gender perspective, males generally assume the task of dealing with outsiders when marketing household products due to their ability to speak the Kinh language. Another perceived advantage is that males have the physical endurance and mobility for riding motorbikes over long distances. This is observed in both top- and bottom-performing groups. Nevertheless, one of the reasons attributed to the lack of success of bottom-performing groups in gaining access to the market — mostly controlled by lowland Kinh — is the pervasive prejudice toward them by the majority Kinh people. In this regard, males among top-performing groups might be less subject to such ethnic prejudices. 18. Cash crop agriculture is one of the main routes out of poverty in all studied provinces. The top-performing groups and the better-off households appear to be capable to diversify into most cash crops available. Nevertheless, even among the top performers, participating in the key agriculture value chains remains limited. For instance, the top-performing Mường and Sán Dìu have tried to ‘upgrade’ their livelihood strategies by engaging in some highly potential value chains. For the bottom performers, whether they could engage in some potential value chains depends partly on the availability of some valued indigenous crops. 19. Informal credits, both in the form of production inputs and cash provided by traders and retailers, could lead to unexpected impediments to fair competition. The traders, retailers, Executive Summary 11 or ‘investors’ usually impose produce prices to their advantage; this is possible due to the pre-existing loans that tie the indebted farmers to them. There seems to be a vicious circle of indebtedness of poor households in all provinces visited. Emergency loans provided by these actors often charge high interest rates. It is noted that the market power of shop owners and traders is generally weaker in areas with better connectivity, physically and economically. In the areas with good connectivity conditions, the top performers usually have competing choices in terms of where to buy inputs and sell their outputs. Such competition keeps the margins and interest rates charged by shop owners and traders relatively competitive. 20. High costs and seasonal availability of transportation make the bottom-performing groups even more dependent on traders (mainly dominated by the lowland Kinh). The issue of manipulation of market and prices is reported to be more significant in Thanh Hóa and among the Khơ Mú in Sơn La. In a related note, it was found that collective buying or selling was rarely observed in either the top or bottom performers. Local farmers seem to keep doing their livelihood activities on individual basis; this tendency, in turn, makes them more vulnerable to market power exercised by traders or agri-input suppliers. (d) Labor Market Mobility 21. Labor market opportunities have become increasingly important for poverty reduction in the ethnic minority areas over the past five years. The top performers are likely to earn the majority of their household income outside agriculture while the bottom performers remain heavily reliant on agriculture as their main source of income. For the top performers, there are usually available options for wage employment as workers either in nearby locations or other provinces. The Mường, Sán Dìu, and Khmer youth, who have completed lower secondary school or higher educational level, could easily find a job in factories located in their home districts or nearby. Jobs within village reach are usually lower paying and less diverse as compared to the opportunities offered in other provinces. These local job positions seem to be suitable to many ethnic minority women, especially if married, as these women could daily commute to work and come home at the end of the day to fulfill their role as housewives. 22. Remittances from household members working far from home play a vital role in changing the physiognomy of the most villages. In visited sites, migration was said to be increasingly popular but is highly gendered: migration was almost exclusive to men, especially among the Mông in Sơn La, Thái Nguyên, and Thanh Hóa, and to a lesser extent for the Sán Dìu (Thái Nguyên) and the Khmer in the Mekong Delta. 23. Wage employment opportunities are increasingly available for the bottom performers, mainly in low-end jobs. Language barriers and low educational attainment are factors inhibiting these ethnic minorities to attain decent job opportunities. Physical connectivity is less likely an issue when it comes to labor market opportunities because there are job-opening advertisement boards and brochures posted all over public spaces, even in the remotest communes or villages. Research Report 12 Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam 24. The informal labor market is an increasingly popular choice among the bottom- performing groups. While accessing the formal labor market or labor export appears to be out of reach to most middle-age villagers or those enduring language barriers or low education attainment, cross-border migration has been seen as an alternative, and increasingly important source of household income for some ethnic minority groups such as the Mông, Ca Dong,2 Khmer, and Khơ Mú. The lack of education reduces access to better employment or productive opportunities among the poor, resulting in their low earnings and risk-taking. 25. Becoming an increasing important pathway out of poverty, wage employment opportunities also present some challenges to ethnic minorities. With respect to formal wage employment, criteria limiting working age between 18 and 35 and requiring a secondary education level often rule out the participation of middle-age women and men in these factory jobs. There have been concerns raised by both workers and state officials on the lack of benefits that wage workers are eligible for, including social insurance, workplace protection and safety rules, and other non-wage benefits. Higher paid jobs are available in other provinces but migrating to other provinces also incurs additional costs. 26. Engaging in the wage labor market also raises a number of family-related issues. Pre- school-age children often are left in the care of their grandparents or other relatives while their parents are at work, especially for migrant parents. Teachers point out this often causes some children to lag behind in class, especially among groups with high rates of non-proficiency in the Kinh language like the Mông, Khơ Mú, and Xơ Đăng. In addition, during parents’ long absences for migrant work results in lack of concern for children’ schooling, thus leading ultimately to dropping out, as observed among the Mường and Khơ Mú in Sơn La and the Khmer in the Mekong Delta. 27. There exists an ethnic differentiation in job structure and share of labor export market. The labor markets such as Japan and South Korea, which demand workers with certain technical skills and language fluency, attract mainly Kinh applicants and some ethnic minority groups considered to be socially and economically ‘advanced’ like the Mường. These returnee workers are generally reintegrated with good jobs in the local labor market due to their working experiences abroad. Whereas the Middle East countries attract female workers, mainly for domestic work, among them women from ethnic groups in the remote areas like the Ca Dong, Mông and Khơ Mú. These migrants have little education and their weak social position make them highly susceptible to exploitation and abuse. Their employment prospect hardly improves after their return. (e) Some Aspects of Access to Education 28. A high prevalence of school dropout in middle or high school is reported in all studied provinces. Leaving at middle-school level is found to be more prevalent among the bottom- performing groups, while the high school dropout rate is reported to be more common among the top performers. Lower rates of educational enrollment are partially explained by household poverty itself, and they correlate to the lack of connectivity among the bottom performers. The prevalence 2 Ca Dong is a local name for a subgroup of the Xơ Đăng in Quảng Nam. Executive Summary 13 of school dropout is reportedly due to geographical distances, lack of social contacts with the Kinh leading to language barriers, and limited contacts with outsiders in the national language. 29. There is a widespread perception that the least performers like Mông, Khơ Mú, and Xơ Đăng are generally taciturn and timid. Conversely, a lack of cultural sensitivity coupled with popular discourses about backwardness, self-deficiency, and superstition as well language barriers among Kinh teachers contribute to misunderstandings and an inability to communicate on educational issues particularly with regard to the Khơ Mú, Xơ Đăng, Ca Dong, and especially the Mông. These factors may sharpen existing perceptions/prejudices of the majority Kinh toward those ethnic minority groups. 30. Those who do not enjoy privileged relationships with power-holders have little chance to benefit from current job opportunities, whether among the best performers or least performers. There were some Mường and Sán Dìu students, for example, who managed to attend universities or colleges but could not find jobs in the state sector. In such cases, the lack of ‘political connectedness’ rather than just a ‘physical connectivity’ became a problem. This factor raises a sense of disillusionment among local people, across all ethnic minority groups, about the benefits of pursuing higher education. While seemingly less significant among the top-performing groups, a combination of political and physical connectivity could contribute to the lower academic achievement at high school level, which, in turn, determines less access to better paying jobs for the least-performing groups. 31. The least-performing groups’ lower tertiary education attainment excludes them from the most rewarding jobs. This in turn will lead them down to the challenge of increased automation and the resulting job loss. Whereas those from the best-performing groups are more likely to have a secondary school education or higher, then usually have ended up being employed in jobs that are likely at lower risks of automation. From gender perspectives, women are more likely than men to be employed in an occupation at high risk of automation. Thus, women tend to face more risks of automation-related job loss than their male fellows. At household level, the technological unemployment will have a more disruptive impact on household welfare given that most ethnic minority families now rely on wages for the bulk of their income. 32. Provision of vocational training for the poor and ethnic minorities is offered free of charge, but in fact the training is not really demand-driven. The training, primarily focused on agriculture and livestock, is usually organized in the national language, using some traditional classroom approaches. It is not easy to get sufficient numbers of young participants registered for the free training. Most of the ethnic minority youth, even among the top performers, are not interested in vocational training in agriculture or livestock production. (f) Some Aspects of Healthcare Services 33. There is an increase in the usage of the healthcare services among ethnic minorities. The top-performing ethnic groups use health care services more than the bottom-performing groups. Utilization of maternal health service by ethnic minority women remains especially low among the Research Report 14 Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam bottom-performing Mông, Khơ Mú and Xơ Đăng. The incidence of Mông and Khơ Mú women who undertook pregnancy check-ups are considerably lower than those in the top-performing groups. However, a high frequency of homebirths within the Mông, Khơ Mú and Xơ Đăng communities was confirmed in the study sites and attributed to the speed of delivery. It is likely that the lack of cultural sensitivity and language barriers among health care workers contributes to misunderstandings and a poor ability to communicate regarding health issues. 34. With regards to child marriage, the bottom-performing Mông have experienced high incidences of child marriage and high fertility rate, but no other studied bottom- performing group was listed with very high rates of underage marriage rate. Meanwhile, the two top-performing Mường and Khmer appear to have a very low fertility rate. Early marriage, early childbirth, and multiple children are sure signs of inherent poverty. In such cases, immature mothers are more likely to have left school early and are thus less likely to know about reproductive health and contraceptive measures. 35. There are high rates of consanguineous marriages among the Mông and Khơ Mú. Whereas, qualitative research indicates the practice of consanguineous marriage is uncommon among local residents. This is also consistent with the information from direct consultation with the Mông and Xơ Đăng. The issue of consanguineous marriage has been subject to many policy remedies and intensive propaganda and the ethnic minority households interviewed was under pressure of revealing its actual prevalence. (g) Traditional Institutions and Local Governance 36. Official and traditional systems of village local governance co-exist and play a role in implementing government and development policies at grassroots level. All studied ethnic groups had traditional leaders such as the village elderly (già làng); head of village (trưởng bản); head of family clans (trưởng họ); and more recently the person with prestige (người có uy tín), who was elected by village residents (and not hereditary) as well as religious leaders (especially the protestant leaders in many Mông villages and monks in the Khmer villages). These people are key figures in managing village life. 37. At the village-level, the village head is a key player in conveying government policies to villagers, along with village mass organizations. As observed in all study sites, the village head seems to take an active role in providing people with alternatives to develop their livelihood, especially in terms of livestock, loans, and employment. It is usually the case that the village head represents the only or the more dominant ethnic group in the village. In the top-performing groups, the roles of village heads are generally more pronounced than in the remote villages of bottom- performing groups, where community leadership is more influenced by traditional persons such as village elderly or clan chiefs. 38. Quantitative analysis indicates a relatively high presence of ethnic minorities in the commune administration system. In reality, most key positions like the Chairman of the People’s Committee and the Commune’s Secretary of the Party are held by Kinh cadres whose qualifications Executive Summary 15 are perceived far better than the local ethnic minorities. With current government policies aimed at reinforcing grassroots structures by moving young Kinh cadres to remote areas, there is little chance for ethnic minority locals to break into these positions, even for the best performers such as the Sán Dìu or the Mường in Kon Tum. 39. Beside key positions in the commune administration system, ethnic minorities are found to hold deputy or non-key positions as well as leadership in mass organizations. Regarding these non-key positions, ethnic minority participation seems to be strongest from the larger and indigenous ethnic groups. Small ethnic minority kinsmen are rarely represented even in these non-key positions. This might put these ethnic groups into more disadvantaged positions in terms of access to resources available from policies and other external support in the commune or villages. 40. Village elders also play an important role in village management but seemingly common only among Central Highlanders. In reality, the village elderly committee has no actual power but merely symbolic. A village elderly however is often a senior member of the leading clans, having vast experience in agricultural production, and well versed in local customs. He advises the head of the village on important matters in day-to-day management and mediates conflicts among clans and households. 41. For most of the studied ethnic groups, clan membership (dòng họ) is the primary source of identity. It serves a management function and is a binding force among local residents. Help in terms of capital, tips on job openings, information on development schemes from household members and other relatives, especially those who work for (local) government often give these households an advantage. Nevertheless, kinship does not seem to play an important role in the process of assessing household poverty and gaining access to resources at grassroots level in the Mekong Delta. 42. For the Mông and the Mường (in Kon Tum), unwritten customary provisions have been handed down from generation to generation, specifying the proper conduct of social behavior and family relations. With exception of the Mường migrated community in the Central Highlands, the top-performing groups are less subject to customary provisions compared to the bottom performing. This tendency might be partly attributed to the fact that the role of traditional institutions, specifically the eroding of the power of the chieftainship, seems to likely decrease among these top performers as a result of their perceivably successful integration into the mainstream society of the lowland Kinh. 43. Apart from their spiritual functions, cultural activities and traditional festivities serve to strengthen ethnic and community cohesion. These are also convenient occasions for locals to share information and experiences in agricultural production such as Lễ cơm mới (festival of worshipping the new harvest) among the Ca Dong in Quang Nam, Ok Om Bok (festival of worshipping the moon), Chol Chnam Thmay (traditional new year festival), and Pchum Banh (thanksgiving to the family ancestors) among the Khmer in the Mekong Delta. It seems that the bottom performers such as the Mông and Khơ Mú have done relatively well in preserving their traditional cultural activities. Research Report 16 Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam 44. Women’s Union and Farmers’ Union are two mass organizations with branches operating at the grassroots (village) level. In addition to their main mandates related to propaganda, managing the preferential loans from the Vietnam Bank for Social Policies (VBSP) is the primary channel for these organizations to contribute to economic wellbeing of ethnic minorities. In development partners-supported projects, Women’s Union (and Farmers’ Union in some cases) have been involved as an active implementing agency of subcomponents that are designated for women as the primary beneficiaries. In most cases, this has proven to be an effective implementation arrangement for women’s economic empowerment. (h) Gender Roles and Intra-Household Power Dynamics 45. Among some highly patrilineal groups like the Mông, Mường, and Sán Dìu, women, particularly married women, are expected to play a subordinate role, with men making most decisions in the household and community. In the seven studied provinces, some ethnic minority groups are patrilineal (the Mường, Sán Dìu, Mông, and Khơ Mú), and some are bilineal (the Xơ Đăng and Khmer). On the one hand, a woman’s position is rather important in societies considered to be patrilineal as indicated by the Sán Dìu in Thái Nguyên. Among bilineal groups like the Khmer and Xơ Đăng, while gender relations are more or less equal with both sons and daughters having equal inheritance rights, the position of men/husbands over women/wives can be more dominant due to their role in agricultural production and other off- farm activities, which bring ready cash for day-to-day-expenses as reported among the Khmer and the Xơ Đăng. 46. Gender roles in livelihood activities are more pronounced, with men involved in harvesting, construction, and animal trading, and women leading in handicraft and clothing production, planting, and animal raising. Men used to be the primary participants in agricultural extension training and community meetings, but officials and NGOs now report greater participation of women. It should also be noted that women of the top-performing groups are actively involved as livelihood habits change as shown among the Mường and the Sán Dìu. 47. Women face social and economic inequalities and barriers of access to resources and livelihood opportunities, yet with a varying degree among the best- and least-performing ethnic communities. For instance, among the top-performing Sán Dìu, Mường, and Khmer where women often have had some schooling and are fluent in the Kinh language, they tend to have more say on agricultural matters. Among the least-performing Mông, Xơ Đăng, and Khơ Mú, women, especially those who did not go to school, rarely or never take part in community activities, leaving production decisions to their menfolk. At another level, while men have the final say in decisions concerning household cultivation and livestock, these decisions are affected by the way they interact with local cadres. This is partly coupled with the pre-existing discourses about backwardness, self- deficiency, and superstition, which are more likely associated with the least-performing groups, rather than the best- performing groups (or perhaps to a lesser degree). 48. At the same time, new labor dynamics for women are prompting shifts in customary gender-ascribed divisions of labor in young households. There do exist factory jobs in the Executive Summary 17 vicinity, which are particularly attractive to local women who want to be near home to do house chores and look after their children after work. There is evidence that in some Mông and Khơ Mú communes in Sơn La and Thanh Hóa, and Sán Dìu in Thái Nguyên, traditions favoring men over women still prevail but are beginning to change such as how local men engage directly in domestic chores and share the household burden with female household members. 49. Moreover, unpaid care work appears to be a critical impediment to women’s economic opportunity, as observed in both the top- and bottom-performing groups. Societal prejudices that have been internalized by women themselves — particularly regarding their traditional housewife role — impact the participation of women in non-agricultural economic sector activities; they instead, for example, seek work in the cities or abroad. As a result, burdened with childcare, the bulk of agricultural labor, and lack of transportation, women are not able to establish and maintain social networks, especially outside the village, at least not at the same level as men. Furthermore, young couples working far from home often leave their children under the care of their parents. This situation creates a kind of unpaid care work often observed among women above age 45, thus depriving women in this age group from non- agricultural work opportunities in the labor market outside their localities. 50. The sex of the household head seems to be a misleading proxy for gender relations. While male-headed households have both spouses present, female-headed households have an absent husband, either due to divorce, migration, or death. A woman could head a household if she does not marry (as among the Khmer and Ca Dong). In such cases, it is important to distinguish between de jure female-headed households (headed by divorced or widowed women) and de facto female-headed households (in which the husband is absent, but may contribute to household finances). The last situation seems to be applicable to the Khmer as seasonal migration is typical in the economic life of the Khmer. 51. Considering only gender of a head of household renders women living in male- headed households invisible. In male-headed households, particularly when the husband’s parents make all decisions on expenditure, a young wife who does not have any access to the household resources is in a vulnerable situation. This situation is quite pronounced among the Mông in Sơn La where the traditional practice of cohabiting with the in-laws prevails. It reveals the intra-family gender and generational dynamics of control between husband and wife, between parents themselves, and between parents and young couples. Therefore, research findings shed light on the complexities inherent in livelihood strategies of minority ethnic communities. (i) Perceptions of Ethnicity and the Dynamics of Trust 52. In all the studied provinces, attitudes of inter-ethnic discrimination and stigma are reported among top performers toward least performers. A commonly held perception by ethnic minority residents is that the Kinh are much more skillful and resourceful than most of the ethnic minorities. Eventually the lowland Kinh appear to dominate and monopolize the whole village trade and business, advantaged by their extensive social networks and greater mobility. This seems to further widen social distancing between the Kinh and the ethnic minority residents in the uplands. Research Report 18 Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam The situation seems less pronounced among the top performing Sán Dìu, Mường, and Khmer given their bigger share in local trading activities at hamlet level. 53. Recent evidence indicates that many upland groups tend to judge themselves by lowland standards and internalize their inferiority. This sense of inferiority and timidity is perceived as a matter of degree when it involves cross-ethnic group comparisons (Nguyen 2016, Well-Dang 2012). This sense of inferiority and timidity, which has been reported to be more common among the least-performing Khơ Mú, Mông, and Ca Dong, particularly impacts participation of young adults who seek work in the cities rather than in non-agricultural economic sector activities. This sense of inferiority and timidity seem to be changing for those with adequate representation in local leadership as well greater mobility as migrant workers in urban contexts such as the Khmer in the Mekong River Delta. 54. When it comes to poverty reduction, the top performers tend to attribute the ‘risk avoidance mentality’ to the bottom performers of other ethnic groups. In this line of thinking, the studied least-performing ethnic groups adopt a ‘play safe’ attitude in agricultural production as well non-farm livelihood activities. Nevertheless, from the insider point of view, most poor households among the least-performing groups are slow in absorbing new techniques or responding to changes in context, more conservative, and do not trust outsiders, such as the lowland Kinh because of past experiences of unfulfilled promises by Kinh traders of numerous trading companies. Moreover, all studied ethnic minority groups often adopt a ‘wait-and-see’ attitude: if someone shows they can do it, the rest will follow suit. This eventually leads to oversupply and depresses prices. Reasons for the poverty experienced by the poor households are often attributed to their timid nature and aversion to change, and shying away from other livelihood options and/or new production methods. 55. Perceptions of ethnicity and cultural elements form the identity of each ethnic minority in Vietnam against the backdrop of their promoted integration into mainstream society and the globalization process. Among the studied ethnic groups, the best-performing Mường and Khmer and the least-performing Mông have preserved their traditions and identity. Whereas the Sán Dìu in the best-performing group were reported to have no households with members knowing their ethnic traditional dance. (j) Access to External Support Initiatives 56. Ethnic minorities in general have been targeted by a number of GoVN-issued policies and programs and development partner-supported projects. These external support initiatives have resulted in important outcomes in many aspects of living standards in ethnic minority areas. Most notably, improvements in physical connectivity, access to public services (education and healthcare), and housing conditions and public utilities (electricity, drinking water) are evident. . 57. While improvements in infrastructure and public services are evident, the outcomes in ‘soft investments’ such as production support, capacity development, promoting voice, and participation of ethnic minorities in community development are less evident. The top- performing groups tend to benefit from these non-infrastructure investments more that the bottom- Executive Summary 19 performing groups. On the one hand, this might be related to better endowment of top performers such as the Mường or Sán Dìu in terms of basic livelihood and market linkages compared to those of the bottom performers. On the other hand, this might reflect the willingness to change as in the Mường and Sán Dìu. In contrast, aversion to change was observed among the Mông and Khơ Mú groups in this study. 58. One important area of external support is to improve access to credit for the poor and ethnic minorities. For GoVN-led policies, the preferential lending through VBSP is the major channel of access to credit for ethnic minorities. Top-performing Mường and Sán Dìu tend to use their loans more effectively in investing in crops and livestock. Most of the Mông and Khơ Mú households were found to use VBSP loans for cattle-raising. It was reported that the bottom-performing households, with more limited ability to save, were less likely to benefit from these micro-finance arrangements. Therefore, in areas with bottom-performing groups, the saving and credit group (SCG) sustainability after completion of the external support initiative was found to be less likely as compared to areas with the top-performing groups. 59. For poverty-targeted programs, households need to be on the official poor list to be considered eligible. Statistics on poverty and poverty reduction bear limited resemblance to the realities on the ground and are often heavily manipulated to suit local political expediency. Field contacts indicate that villagers are frustrated by the way they are classified as ‘poor’ one day and become ‘no longer poor’ the next cycle. Moreover, the distinction between ‘poor’ household and ‘near poor’ household is quite arbitrary (as the official poverty assessment criteria are interpreted by the local officials) and heavily influenced by the power dynamics displayed throughout the poverty assessment process at the hamlet or village level. In this regard, representation in the local administration system or village authorities appears to be a factor that could influence, in one way or the other, the result of poverty assessment. 60. Related to poverty assessment and targeting, there is a power dynamic at play between different groups within the community. Policies are not simply imposed top down, implemented by local cadres, and obediently obeyed but often meet with resistance from locals when they see a threat to their interests. This process crosses all the studied groups. In the presence of reliable poverty assessment data, the figures reported on poverty reduction could be a potentially misleading indication of the development results of policies and projects for poverty reduction of ethnic minorities. 61. There has been a large gap between basic principles in policy designs and actions in practice. Consistent and enforced guidelines are still lacking for local-level participatory planning and having the prioritization processes of the poverty reduction programs as part of participatory planning. Furthermore, the quality of participation, especially of ethnic minority women, was found to be limited. The lack of participation in the socio-economic development planning process impedes the voices of many ethnic minorities in the local decision-making process. In this regard, the top performers seem to be most ‘active’ while participation of the bottom performers such as the Khơ Mú is ‘extremely limited’. Research Report 20 Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam 62. Another mechanism of decentralization is ‘commune investment ownership’ (CIO) at the commune level under authority of the Commune Peoples Committee (CPC). The qualitative study indicates that CIO has been stronger in areas with top-performing groups compared to that of bottom-performing groups. This is partly attributed to capacity of CPCs in the areas with good connectivity. 63. Gender mainstreaming is another principle that is reflected in most current GoVN- led policies and programs. Most notably, the largest poverty reduction initiative, the National Target Program on Sustainable Poverty Reduction (NTP SPR) 2016-2020 was designed with the principle of giving priority to women, especially women in poor households and ethnic minority women, for participation in and benefit from program activities. Recent evidence indicates that the implementation of this gender principle remains very limited, reportedly due to lack of enforced guidelines and local capacity to implement gender mainstreaming. 64. According to some interviewed local officials, development schemes sponsored by the state or NGOs seem to create a mentality of dependency. This might be a result of having GoVN-sponsored policies and programs provide production inputs and other support for free while such provision of inputs is not a result of a participatory process where the support reflects the actual needs of the beneficiaries. In other words, the poor ethnic groups are targeted by policies, but decentralization and participation were not sufficiently receptive to the poor groups in seeking their input as agents of change. It is noted that such dependency appears to be more pronounced in the bottom-performing ethnic groups. Addressing the Drivers of Differing Levels of Social and Economic Development Ethnic Minority Groups 65. How have gaps between the top- and bottom-performing ethnic minorities been addressed by GoVN in collaboration with other development partners who are interested in poverty reduction for ethnic minorities? Because there are recent reviews on policies and programs on poverty reduction in general and for ethnic minorities in particular, this study does not aim to review the existing policies on ethnic minority development. Instead, the research team has chosen to use existing reviews to map the analytical framework developed for this study. The purpose of this ‘mapping’ is to find out how the existing policies and programs have addressed the factors that determine relative socio-economic development of different ethnic groups. 66. The mapping results indicate that the main focus of the existing policies have been placed on physical connectivity, access to public services, access to credit. There are areas of moderate focus such as production support and local governance. There are factors that have been partly addressed by the existing policies but with weak or limited focus such as support to labor market, landholding, traditional institutions market linkages, vulnerability and shocks, and other direct support to the poor such as housing, water and sanitation. In addition, gender roles and power are also partly addressed by adaptation of participatory socio-economic development planning. Executive Summary 21 67. Notably, there are areas that are arguably ‘blank’ in the existing policy agenda. The factors such as ‘culture, spiritual belief, religion’ and ‘misperception and stigma’ are generally not addressed by existing policies. In addition, ‘social network and kinship’, ‘political connectedness’, and ‘remittances’ are not subject to the existing policies for poverty reduction. Most importantly, the existing policies have not addressed the gaps across different ethnic minorities groups. Instead, the ultimate policy target aims at poverty reduction for ethnic minorities and hence closing the gaps between the ethnic minority groups and the majority group. The findings from this study indicate that the top-performing groups are generally subject to better access and endowments compared to the bottom-performing groups. This could be taken to suggest that the gaps will widen in the future. 68. It is important to ‘revisit’ existing policies to ensure the different factors that determine the wellbeing of ethnic minorities could be addressed. In addition, to close the gaps across the best- and least-performing groups, the bottom performers should benefit more than or equal to the top performers. 69. Some policy recommendations from the study follow. (a) Refocusing Ethnic Minority Development 70. Infrastructure investments should be refocused equally to maintenance of the existing infrastructures in most areas. This extensive focus on infrastructure investment was justified in the early stages of policies and programs in the ethnic minority areas. It was the stages when geographical locations of ethnic groups might be most important. After more than two decades of intensive investment, physical connectivity has been improved substantially. The focus should be now re-directed to (a) only the areas with very poor physical connectivity where the bottom- performing groups reside and (b) organization and management (O&M) investment, both in physical terms and in O&M capacity, in the remaining areas where physical connectivity is already high. 71. Soft investment, especially in production support, capacity development, and participation should be the main focus of polices for poverty reduction in ethnic minority areas. This is also a way to make better usages of the investment in improving physical connectivity. Of all these areas, giving voice and participation opportunities to the ethnic minorities, especially among the bottom-performing groups, should be the core focus for interventions. 72. Access to labor market should be the main focus of future policies. To make labor market work for the poor ethnic minorities (especially women and those in the bottom-performing groups), requires a holistic approach. Equal opportunity legislation needs to be consolidated with affirmative action with regards to non-discrimination for ethnic minorities at workplace, especially in the foreign-invested sector. 73. Women’s economic empowerment should be strengthened in response to the many disadvantages faced by ethnic minority women. Ethnic minority women face ‘conventional’ gender inequality and the inequality just by being from an ethnic minority. It was found that ethnic Research Report 22 Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam minority women, especially those in the bottom-performing groups, are at disadvantaged positions to almost all type of important access that contributes to their economic empowerment. 74. The top performers and bottom performers respond differently to factors that contribute to socio-economic development of ethnic minorities. External interventions that work for the majority of ethnic minorities might not work for the bottom performers, and these interventions might widen the gap across ethnic groups. The disadvantages encountered by bottom performers warrant the focus for future poverty reduction efforts. 75. Addressing misperceptions and social stigmas should be a priority of government policy. To date, misperceptions and social stigma have been topics of discussions in studies and various consultations between GoVN and development partners. Given this significance, these issues must be recognized as an impediment to socio-economic development of ethnic minorities and hence be considered as targets of concrete policy actions. 76. Rationalizing and systemizing the policies and programs on socio-economic development should require a restructuring process of poverty reduction mandates across these key players. Out of 130 policies, there are only nine policies that were under the direct mandate of CEMA; the other policies and programs for ethnic minorities that are under the mandate of other ministries, of which CEMA have exercised little influence. Rationalizing the existing policies on ethnic minority development requires is not possible without the joint effort of CEMA and other key ministries. (b) Enhancing the Existing Policies and Mechanisms 77. Market access, including efforts to address informal credit, should be improved. More than improving physical connectivity, supporting market institutions and actors that support doing business in ethnic minority areas is important. Advanced approaches such as blockchain and information and communications technology could ease many of the constraints encountered by ethnic minorities to be part of the local, national, or even international value chains. A formula for supporting production capacity needs a breakthrough to generate value added for ethnic minority farmers. In a related note, the practice of ‘informal credits’ should be addressed, eliminating the problem of many of the poorest and bottom-performing ethnic minorities getting locked in a cycle of borrowing from informal lenders at high cost and selling at low then ending catastrophically with the transfer of land and other assets in repayment for high costs charged by private money lenders. 78. Local governance and traditional institutions should be strengthened. Capacity building for GoVN officials at the local level should be a focus to ensure cultural and ethnic sensitivity in addressing poverty for ethnic minorities. In addition, representation of ethnic minorities in local government should be further strengthened. In this regard, until ethnic minorities could assume key positions in the local government, the interpretation and implementation of policies will continue to reflect the views and experiences of the Kinh cadres who now mostly fill these key positions. 79. The initiatives to involve community leaders in implementing policies and projects for poverty reduction on the ground should be continued and expanded. To make this useful Executive Summary 23 for changes, it is important to put this effort in the wider context of decentralization and participation. These community leaders should also be the focus of future capacity development. In addition, the roles of religious leaders in local economic development should be recognized to avoid the hesitation perceived in some cases by the authorities in dealing with the religious leaders. 80. Capacity development approach and content should be completely revisited and renewed. Capacity development is recognized in almost all policies and projects for ethnic minority development. However, capacity development is usually seen as on-and-off short training courses or study tours in GoVN-led policies and programs. There has been a somewhat misleading impression that capacity development activities are already sufficient. In fact, existing capacity development activities are more or less redundant at places. What is needed is capacity development that is viewed as a process, with focus on approaches and thinking (rather than to articulate the policies and principles) in relevant capacity-building modality. 81. Vulnerability and shocks should be targeted in all policies. Bottom performers are more vulnerable to external shocks and less capable in coping with these shocks. An approach for building up resilient needs to be developed to fit the context of ethnic minorities in the country. That resilient framework should be simplified to facilitate policy intervention planning and targeting. Based on that framework, resilience should be built into policies on ethnic minorities. Given the complexity of this focus, it might be useful to pilot interventions with support from development partners. Experiences and lessons learned should be then shared for future policy dialogue in this regard. 82. The role of provincial government in preparing ethnically responsive measures should be exercised. One principle in almost all GoVN-issued policies on ethnic minorities introduced at the central level is to decentralize power to provincial authorities to develop specific mechanisms and guidelines to make the policies responsive to local context and socio-economic characteristics of the beneficiaries. In reality, this depends on the ability and willingness of the provincial line departments to submit the relevant mechanisms for appraisal and approval of the provincial authorities. However, the provincial line departments usually mirror the policies developed at the central level and therefore the intention of decentralizing to the provincial authorities to make the central-level policies more responsive to local context has been hardly materialized. 83. Decentralization should be strengthened in ‘real’ terms. In relation to making policies more locally and ethnically responsive, decentralization is expected to place the decision-making power at the local level, especially the community. To continue this decentralization requires strong enforcement that can only be ensured by clear instruction and strong determination from the central and provincial levels. 84. Participatory socio-economic development planning (SEDP) should be institutionalized at the central level to guide the local prioritization process. The lack of a uniform regulation or guideline on the local SEDP from the central level represents a major impediment to the adaptation of participatory SEDP in practice. Therefore, it is recommended that the participatory SEDP should be institutionalized consistently from the central to the provincial level. Research Report 24 Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam Conclusions (a) Pathway out of Poverty • Residing in the areas that are easily connected with basic infrastructures and economic clusters that offer job opportunities. Given the recent improvements in physical infrastructures, connectivity to economic cluster have become increasingly important. • Engaging in cash crops that are not already ‘controlled’ by the lowland Kinh in the locations such as the key agriculture value chains of the country. These might be the cash crops that serve to local niche markets or the ones that are indigenous to the area. • Actively looking for labor market opportunities, especially as wage employment. This is adopted as a household-level strategy so that within-household division of labor may change accordingly to give opportunities for the ‘employable’ adults, including women, to engage in the labor market opportunities; • Finding ways to interact actively with opportunities and stakeholders that determine allocation of resources; and, on that basis, make good use of the resources available from different policies and projects, including access to credit. The ability of finding these ways depends on political connectedness, participation in the local administration or local mass organizations. • Interacting with key community members (hamlet cadres, representatives of traditional institutions, and local residents) who are given proper conditions to develop self-reliance and relational autonomy in partaking in community development initiatives toward poverty sustainable reduction. • Adapting selectively to certain aspects of Kinh culture. Factors constituting and enabling connectedness depends on the ability of ethnic minorities to adapt in such ways in order to be less subjected to misperceptions or social stigma inserted by the Kinh-dominated society. This contributes to closing the gap and creating ‘equal footing’ between these ethnic groups and the Kinh majority in accessing opportunities as well as the returns from these opportunities. (b) Trapped at the Bottom • Being poorly connected to physical infrastructure. These are mainly attributed to distance or quality of access rather than availability of access to infrastructures, which has been substantially improved over the past two decades. Being poorly connected to economic opportunities, especially labor market opportunities, is a major constraining factor for escaping poverty. • Lack of availability of cash crop opportunities which are caused either by poor livelihood potentials in the locations or external factors such as being resettled or the loss/interruption of access to natural resources (such as forest) that contribute to the household livelihoods. More importantly, lack of capabilities (e.g., labor, productive landholdings) and risk-taking to engage in (existing or emerging) cash crop opportunities. Executive Summary 25 • Constraints to access to labor market opportunities, especially wage employment in factories. These constraints are linked to low educational attainment levels, language barriers, other cultural barriers caused by lack of exposure to new environment, and gender-biased household division of labor that traps women in house-chores and agriculture work in the local areas. • Inability to develop an active, proper role for various community members (hamlet cadres, representatives of traditional institutions and local residents) as well as an inability to create opportunities to develop self-reliance and relational autonomy in partaking in community development initiatives toward poverty sustainable reduction. This is intensified by the absence of effective mechanisms to facilitate voices and participation; and as a consequence, the bottom-performing groups tend to share little in local decision-making process. • Inability and/or reluctance to establish connectedness and to adapt to Kinh culture and society thus perpetuating prejudices especially among least-performing groups. These prejudices are widespread among the Kinh cardes, teachers, healthcare staff, and employers, and therefore undermine access to opportunities as well as public services. Limitations and Future Research Agenda 85. In these final remarks, some limitations of the study should be noted when interpreting the findings in this report. A limitation to this study was that research sites were selected based on direct consultation with authorities dealing with ethnic minority affairs at the provincial and district levels based on the pre-determined criteria. Moreover, the fact that participants for in-depth interviews and group discussions were selected via village cadres (even there were general criteria for this selection available). This approach inherently lacks objectivity. Additionally, the nature of qualitative research data collection could result in a social-desirability bias. These issues were noted and partly addressed by the study team but some of its effects remain. 86. For future research agenda, there are are a few issues that the current study does not address (or only briefly discuss) and and which could be the contents for future research agenda. First, it is not possible in this study to examine the living and working conditions at the destinations, especially workplace features. These aspects are important to reflect the ‘quality’ of these opportunities and hence to understand the dynamics of labor market participation by ethnic minorities. Second, the issue of religious practices was reported to be an increasing important aspect of the wellbeing. However, this issue has not been captured properly in this study. Finally, although the selection of the study sites was made to ensure that one selected ethnic group could be studied in different locations, these differences were not pronounced from the fieldwork; and, as a result, the report is quite salient in informing such within-group differences. Research Report 26 Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam Introduction 1 1. While Vietnam has now reached lower middle-income country status, the gaps among the ethnic minority populations and the majority groups have become more evident and widening over time. Also, ethnic minority groups differ in terms of where they fall in these gaps. Poverty rates among these minority groups differ widely, ranging from 27.09 percent for the Sán Dìu to 88.17 percent for the Mông using the multidimensional poverty line of the Government of Vietnam (GoVN) and the 53 Ethnic Minorities Socio-Economic Survey (53EMS). Recent evidence indicates that some groups have “out-performed” other groups in terms of poverty reduction as well human capital development (CEMA, UNDP and Irish Aid 2017). There have been only a handful of studies explaining why some ethnic minority groups tend to perform better than others. For instance, Andrew Wells- Dang (2012) studied those ethnic minorities who succeeded in escaping poverty, pointing out that this success is attributed to government programs providing credit, agricultural training, and infrastructure; and, above all, to the efforts made by ethnic minority communities themselves. These efforts involve a shift to commercial agricultural production as well as a willingness to “re-imagine themselves as modern individuals” away from old stereotypes. 2. This study aims to empirically examine why and how certain ethnic groups have managed to rise to the “top” as “best performers” while the other groups seem become trapped at the “bottom” as “least performers”.3 The study set out to answer these key questions: (a) What are the factors contributing to different socio-economic performances of ethnic minorities? (b) Why have some ethnic minority groups successfully managed to escape poverty while others have lagged far behind? (c) How have such factors been addressed in the respective policies and designated programs or projects initiated by the government, development partners? (d) What are the changes needed for future design and implementation of initiatives to support sustainable socio-economic development among ethnic minorities? 3. This study adopted a mixed approach, combining both quantitative and qualitative methods. For the quantitative analysis, access to the raw data of Survey on Socio-Economic Conditions of 53 Ethnic Minorities (hereafter ‘53EMS’) was approved by the CEMA Minister. The Human Development Index (HDI) and Multidimensional Poverty Index (MPI) are used as proxy for 3 The term top and best performing are used in the current study to refer to the ethnic minority groups (performers) that are at the top-end of socio-economic development rankings (measured by the HDI and MPI proxies). Similarly, the terms bottom and, least performing are used to refer to the ethnic minority groups/performers that are at the bottom-end of these rankings. It is acknowledged that these terms do not encompass the qualitative meanings expressed here but are used as a shorthand for brevity and distinction of characteristics in this study report. Introduction 27 socio-economic development of different groups. Along with the collected qualitative data and expert opinions on ethnic minority groups, the 53EMS dataset was analyzed to identify the ‘top’ and ‘bottom’ performers among the ethnic minority groups. Once the top and bottom performers were identified, ethnographic fieldwork was conducted among the selected ethnic minority groups. 4. Another aspect of the research that went into this study was the capacity building for several early- and mid-career researchers from the CEMA Academy for Ethnic Minorities. Before the start of qualitative data collection, around 20 CEMA research staff were recruited to attend two technical trainings on quantitative and qualitative research methods. From the course participants, 12 researchers were selected to join the team for qualitative data collection (April to July 2018). Some of these researchers were from Mông, Sán Dìu, Thái and Mường groups; thus, their background, knowledge, and insight were invaluable in conducting the study and contributing to this final report. 5. The report is organized as follows. Section 2 describes the methodology used in conducting the report. Section 3 presents preliminary findings from the quantitative analysis of 53EMS pertaining to ‘who are the top- and bottom-performing ethnic minority groups?’. The selection of the top- and bottom-performing ethnic minorities using the 53EMS provides the basis to make selection criteria of ethnic groups for the qualitative study. The research findings, informed by both the 53EMS quantitative analysis and the qualitative study, are presented in Section 4 and discussed in accordance with the key factors determining socio-economic performances of different ethnic groups. Section 5 discusses implications and recommendations for strategies, policies, and projects by GoVN and development partners for continuing efforts on poverty reduction for ethnic minorities, especially the bottom performers. Section 6 summarizes the driving forces that lead to the top or bottom, in terms of socio- economic performances. Dance of the Umbrellas Research Report 28 Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam Methodology: How the Study was Conducted? 2 Given the complex nature of the research questions, this study used a combination of quantitative and qualitative components that benefited from the enhanced capacity building of the CEMA team researchers who were integral to data collection and contribution to this final report. The integrative approach to the research that combined qualitative and quantitative components along with the research capacity building component is further discussed. 2.1 Quantitative Component 6. For the quantitative part of this study, the 53EMS was adopted as the primary source of information in order to identify the best- and least-performing ethnic minority groups; and to describe the basic profiles of these groups relative to the average of ethnic minorities. 7. The 53EMS is the most comprehensive database on socio-economic development of Vietnam ethnic minorities date. The 2015 survey was the first in a series, which will be undertaken every 5 years. The master sample was identified from the 2009 Population and Housing Census. The 53EMS sample consists of 341,142 ethnic minority households in 5,402 communes located in 489 districts in 51 provinces of the country. This 53EMS 2015 was the first in the series, which will be undertaken every 5 years. A household questionnaire was applied to all the sampled households. This 15-page questionnaire was designed with questions on (a) basic household information (demography, labor market participation); (b) information on women ages 12-49 (re-productive health); (c) information on the deceased persons (number and causes of death); (d) housing conditions and assets; and (e) socio-cultural aspects and access to public services). Out of this sample, a sub-sample of 52,159 households was selected (around 15 percent) to complete an income questionnaire with queries on different sources of household income (from agriculture, livestock, forestry, aquaculture, agricultural services, and non-farm activities). 8. The HDI and MPI were applied to identify the best- and least-performing ethnic minority groups. The 53EMS provided sufficient information to calculate the HDI for different ethnic minority groups. Following the United Nation’s standardized approach in calculating HDI, information from 53EMS was used to calculate the gross national index, life expectancy index as well as education index. To supplement HDI, the MPI approach was adopted as an alternative proxy for socio-economic development of ethnic minorities. Accordingly, MPIs for all ethnic minority groups were calculated by using the GoVN multidimensional poverty (MdP) line.4 This MdP line helps to capture income and 4 Refer to Decision No. 59/2015/QD-TTg on the official multidimensional poverty (MdP) line for Vietnam. Methodology: How the Study was Conducted? 29 five dimensions of public services, including housing, water and sanitation, healthcare, education, and information. Ideally, given the focus of this study, an additional dimension on ethnicity should be added to augment this MdP line. However, the 53EMS provided only limited information on cultural and ethnicity aspects of ethnic minorities (there were three questions on whether the household members know the traditional songs, dances, and music instruments); hence, the proxy for this ethnicity dimension would not be too informative. Therefore, this dimension is not included in the MdP measurement. 9. Using the data available from 53EMS, the ethnic minorities groups were ranked either by the HDI or MPI calculation. In order to select the top- and bottom-performing ethnic minority groups, a shortlist was drawn from the two calculations and used to consult with social scientists and development practitioners who are experts in ethnic minority-related issues in Vietnam. The final list of top and bottom performers was subsequently informed from the quantitative process, which was eventually based on the HDI and MPI measures as well insights from the expert consultation activities. 2.2 Qualitative Component 10. With the top and bottom performers selected, qualitative approaches were adopted to shed light on the driving forces underlying their differing performances. It was expected that each ethnic group should be examined in at least two respective provinces with different geo- graphical conditions. In most general terms, the qualitative methods utilized in this study include the following research instruments: • Key informant interviews were conducted with representatives from the government agencies, and donor-supported projects at all levels. At the provincial level, key informant interviews were held with representatives from Department of Ethnic Minority Affairs (DEMA), Department of War Invalids, Labour and Social Affairs (DoLISA), Department of Agriculture and Rural Development (DARD), Department of Health (DoH), Department of Education and Training (DoET), Department of Planning and Investment (DPI), Department of Finance (DoF), Women’s Union and Farmer’s Association, provincial vocational training college, donor- supported provincial management units, and some local enterprises that link to livelihood activities of ethnic minorities in the provinces. At the district level, district affiliates of the respective organizations were involved as key informants. In addition, representatives of the District’s People Committee and District Vietnam Bank for Social Policy were also consulted. At the commune level, the Chairman or Vice Chairman of the Commune Peoples Committee (CPC) and key staffs were consulted. In addition, these intereviews were also carried out at the village levels with poor and non-poor households, and household businesses (such as shop owners) of different ethnic groups in the village. • Focus group discussions were carried out with numerous stakeholders at the commune and village levels. At the commune level, these discussions were conducted with key CPC staff, representatives of mass organizations, head of the commune health station, and head masters of the schools in the commune. At the village level, focus group discussions were held with village representatives such as village heads, village mass organizations, family clan chiefs, and village elders. Research Report 30 Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam 11. Participation in this study was completely voluntary. Data collection sessions, either through in-depth interviews or focus group discussions, began with a fully informed consent procedure. Furthermore, participants could choose to stop participating at any point during the data collection session and/or could decline to answer specific items without any penalty. Participants’ confidentiality was protected by virtue of the team not recording any names or identifying information or specific locations in the study report. 2.3 Capacity-building Component 12. The capacity-building component was specifically set to re-enforce research capacities of early- and mid-career researchers from the CEMA Academy for Ethnic Minorities. Accordingly, this component consisted of two training courses using the adult-learning approach on quantitative and qualitative research methods.5 The contents of these courses were tailored to supplement the study. An acute objective of the training was to prepare the participants for undertaking fieldworks among the top- and bottom-performing ethnic groups. 13. A group of 20 researchers from the CEMA Academy for Ethnic Minorities and other CEMA departments were selected to attend the training courses. A consultation meeting with the prospective researchers was organized to make a general assessment of their background and training expectations. The training courses were then organized between late December 2017 and early January 2018 (each course was held in a week). Since the courses were designed for those with some experience in doing empirical research, the training was delivered as a platform for CEMA researchers and the consultants to share experiences and reflections on the use of different research instruments in the study of ethnic minority development. 14. Among the course participants, 12 researchers were then recruited to form the CEMA research team in this study. The selection was made in consultation between the CEMA Academy for Ethnic Minorities and the World Bank consultants, taking into consideration the participants’ eventual performance during the respective training courses as well as their research interests and experience. Some study team members have ethnic ties to Mông, Sán Dìu, Thái and Mường. Their background, knowledge, and insight of these ethnic groups were invaluable for the study during the field visits. The two World Bank consultants developed a set of study instruments in close consultation with the CEMA research team.6 Before the start of data collection, the team was trained on the study instruments as part of fieldwork preparation. The team carried out the qualitative data collection from April to July 2018 and to a certain extent its analysis under the technical supervision of the two World Bank consultants. After completion of fieldwork activities, the CEMA researchers submitted a short report on selected thematic issues. These drafts subsequently helped to inform this final study report. 5 The contents of these training courses are available from the authors upon request. 6 The research team has developed a complete set of data collection tools for different types of stakeholders. These tools are available (in both English and Vietnamese language) upon request. Methodology: How the Study was Conducted? 31 Who are the Top- 3 and Bottom-Performing Ethnic Minorities? As described in the study methodology in section 2, in order to identify the top- and bottom- performing ethnic minorities, the 53EMS is referenced to calculate HDI and MPI as two proxies for socio-economic development of ethnic minorities. 3.1 Ranking Human Development of the Ethnic Minority Groups 15. The Human Development Index. The calculation of HDI follows the United Nations’ standardized approach. The HDI is in the family of focus measures of human development that gives a composite index to the key dimensions of human development.7 This HDI is not a comprehensive measure of human development. Instead, it consists of three dimensions including income, life expectancy and education index. According to the UN Human Development Office, these essential aspects are prerequisite for people to “lead a long and healthy life, to acquire knowledge and to have access to resources needed for a decent standard of living”. Literature on the subject has debated the advantages and disadvantages of the HDI approach in measuring human development. Among common critiques are that HDI masks a series of disparities within a country and that the medium- or long-term nature of HDI disables any reflection of short-term inputs to human development such as policies. In regard to the former point, it was not a serious issue in this study as the HDIs of each ethnic group could be used and would be utilized to rank the level of human development across ethnic minority groups. With respect to the latter, this shortcoming is addressed by the usage of a Multidimensional Poverty Index. 16. The Multidimensional Poverty Index. The MPI for each ethnic group helps to supplement the HDI approach. The MPI definition varies from one country to another, depending on the policy priority agenda for addressing poverty reduction.8 Accordingly, the MPI consists of income and access to basic public services, including housing, water and sanitation, healthcare, education, and information. In that sense, a rural household is classified to be ‘poor’ if the household happens to fall 7 In contrast to the focus measures, there are breath measures of human development that encompasses all aspects of human development. While these breath measures are more comprehensive, quantifying human development using these breath measures is not always available. Source: http://hdr.undp.org/en/hdi-what-it-is, accessed No- vember 29, 2018. 8 In Vietnam, the first MdP line was institutionalized by the Decision No. 59/QD-TTg, dated November 2015. Research Report 32 Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam into either the following situations: (a) having income per capita of less than VND700,000 per month; or (b) having income per capita between VND700,000 and one million VND per month and lack of access to at least 3 out of the 10 indicators that reflect the access to basic public services.9 Compared to HDI, MPI reflects additional measures such as water and sanitation, housing, and information. Moreover, as MPI is based on accesses to these basic services, it could reflect short-term inputs to human development (instead of only focusing on outcomes such as life expectancy and educational index). Therefore, MPI is considered to supplement HDI in order to identify the top- and bottom- performing ethnic groups in this study. 17. Small-populated ethnic groups were not considered for this study. In the beginning, it was intended that the study would not address ethnic groups with very small population of less than 15,000 in population. In addition to general policies and programs that target ethnic minorities, these groups are also subjected to numerous designated supporting policies (mainly to provide special social assistance). Due to their small population and isolated locations (and lifestyles), these small- populated ethnic groups have generally not been exposed to many of the driving forces that are examined in this study, such as the labor market for wage employment and market opportunities for many key cash crops. For a more practical reason, the exclusion of these small populations is a result of data constraints. Data on income are required to calculate under HDI and MPI approaches. This level of information is only available on 15 percent of the total sample of the 53EMS and unreliable for the HDI and MPI on the small groups. As a result, the ranking of HDI and MPI in this study is made for only 34 ethnic minority groups. 10 18. Using the HDI approach, Figure 1 ranks the HDIs of selected ethnic minority groups. Accordingly, the top human development rankings of ethnic minority groups are the Hoa, Tày, Sán Dìu, Mường, Giáy, and Nùng. The bottom-ranked groups are the Mông, Raglay, Khơ Mú, Xinh Mun, Xtieng, and Hà Nhì. The HDI gap is quite large between the top ranked and the bottom ranked. The top-performing groups appear have HDIs ranging between 0.546 (the Nùng) and 0.646 (the Hoa); the bottom-performing groups have HDIs hovering from 0.385 (the Mông) to 0.430 (the Hà Nhì). 9 In the Decision No. 59, access to these public services is measured by 10 indicators, including access to health services, having health insurance, educational attainment of adults, education enrollment of children, quality of housing, housing space per capita, access to safe drinking water, having hygiene toilet facilities, usage of telecom- munication services, and having access to receiving information. 10 Out of 53 ethnic minority groups, the followings are not reported in this table, including (i) 16 groups with the population of less than 10.000 persons. This includes the following groups (with the population calculated from the 53EMS provided in parentheses): La Hủ (10,686), La Ha (9,253), Pà Thẻn (7,271), Lự (6,162), Ngái (858), Chứt (5,751), Lô Lô (4,254), Mảng (4,002), Cờ Lao (2,615), Bố Y (2,309), Cống (2,395), Si La (672), Pu Péo (722), Rơ Măm (451), Brâu (423), and Ơ Đu (367); and (ii) four groups with the population of between 11.000 and 16.000 persons. This includes the following groups (with the population calculated from the 53EMS provided in parentheses): Lào (16,138), La Chí (13,712), Kháng (15,123), and Phù Lá (11,777). For these groups, calculation of HDIs and MPIs are subjected to small sample sizes and thus potentially not reliable. Who are the Top-and Bottom-Performing Ethnic Minorities? 33 FIGURE 1: The HDIs of Selected Ethnic Minority Groups in Vietnam 0.646 0.592 0.587 0.568 0.558 0.546 0.543 0.541 0.530 0.524 0.519 0.509 0.501 0.497 0.490 0.488 0.486 0.484 0.474 0.469 0.467 0.465 0.451 0.449 0.449 0.445 0.437 0.430 0.425 0.414 0.414 0.389 0.385 Sá ày M ìu ng N y n g ay Ch ổ ăm Kh ái Tà r Ôi Cơ ê C Tu Tr o Cơ g Ho Ch ạ Ru ê o Gi o Đă i M ng Br B g Vâ Na Hà u X ì nh g Kh un Ra ú M y g a Xơ Ra Nh e á a M Ho ÊĐ Hr Th Gi hơ R Da C ề M Sá ùn n n Xi tiên ôn Th D m T Gi Ch gl Ki M ườ iê nô u u a a ơ n n é Source: Authors’ calculation from 53EMS 19. With regards to the rankings based on the MPI approach, Figure 2 indicates that the most top-level ethnic groups are the Hoa, Cho Ro, Sán Dìu, Khmer, Chăm, and Mường, while the bottom-level performers are the Khơ Mú, Bru Vân Kiều, Mông, Xinh Mun, Cor, and Xơ Đăng. Notably, the gap between the lower bottom and the upper top in MPIs is huge (even if the Hoa represent an exception and could be taken out of this comparison). For instance, the MPI of the Sán Dìu was 27.1 percent, compared to the Khơ Mú at 91.7 percent. If the average MPIs of the upper top levels and the lower bottom levels are averaged, the value for the former is 32 percent while that of the latter is 88.1 percent. FIGURE 2: The MPIs of Selected Ethnic Minority Groups in Vietnam 91.7 89.0 88.2 86.1 84.2 82.8 79.8 79.0 76.1 75.7 72.0 68.9 68.5 66.2 64.0 63.1 62.6 56.2 53.0 51.1 49.5 48.6 48.2 46.1 43.5 42.0 41.7 40.8 35.5 34.5 27.1 14.5 5.9 Ch a Sá Ro Kh ìu Ch r M ăm ng Ch ạ Xt Ru ng y ê áy n g Cơ y Ho Gi hổ ai Ôi Tu Da i o a Hà ng M hì ng Ra ê Xơ glay ng nh Co Br M un Vâ ông Kh iều ú á e Tà a Ho M Gi a N ÊĐ Hr M Sá n aR Th D N m Gi Ch T M ườ iê Nù iê nô Đă Tà Cơ K ơ u ơ n B Tr n é Xi u Source: Authors’ calculation from 53EMS Research Report 34 Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam 3.2 Selection of Top- and Bottom-Performing Ethnic Minority Groups 20. Based on the derived shortlist of the top- and bottom-performing ethnic groups described in Section 3.2, final selection was made after intensive consultation with numerous scholars and experts in ethnicity and poverty, and with input from the World Bank and CEMA representatives. The consultations were held between December 2017 and January 2018.11 Another issue considered in the selection process is to ensure that the scope of the qualtitative study remain managable within the time and resources available to the study. Accordingly, the following groups were selected: • The Mường and Sán Dìu represent ‘top’ performers. These two groups, together with Hoa (or the Chinese) are consistently in the top-performers leagues in both the HDI or MPI rankings. The Hoa was however not selected as the top performers for this study. Given that the majority of the Hoa resides in urban areas (i.e., nearly half of Hoa population concentrating in Chợ Lớn area of Ho Chi Minh City),12 the socio-economic context of the Hoa is considerably different from that encountered by the remaining ethnic minority groups in Vietnam. A summary of these groups is provided in Table 1. • The Khơ Mú and Mông represent ‘bottom’ performers. The Mông and Khơ Mú, together with Xinh Mun, are consistently in the bottom of the HDI and MPI rankings. However, the Xinh Mun is a relatively small group compared to the selected Khơ Mú. Using the 53EMS update, it is estimated that the Xinh Mun population by 2015 was 27,361 and that of the Khơ Mú was 84,525. In this regard, the Khơ Mú was selected over the Xinh Mun in the bottom- performing groups. • The Xơ Đăng and the Khmer were also selected. The Khmer groups ranks near the top- performing groups when the MPI approach is used and in the upper part of the HDI ranking. The Xơ Đăng group is near the bottom-performing group when identified by the MPI approach and in the lower part of the HDI ranking (but not at the bottom). In this context, the Xơ Đăng – known as one of the indigenous groups in the Central Highlands – was selected being closed to the bottom of the ranks; and the Khmer – being predominantly in the Mekong River Delta – was also selected as the ethnic group that is close to the top of the ranks. 21. With this selection, the study covers a spectrum of socio-economic development rankings with the Mường and Sán Dìu at the top, the Mông and Khơ Mú at the bottom and the Khmer and the Xơ Đăng near the top and bottom, respectively (Table 1). These selected ethnic groups account for approximately 34 percent of the overall ethnic minority population of the country. In terms of geographical distribution, the main distribution areas of the selected groups expand from the Northwest (Sơn La) and Northeast (Thái Nguyên), to the North Central Coast (Thanh Hóa) and South Central Cost (Quảng Nam), to the Central Highlands (Kon Tum) and the Southeast (Sóc Trăng and Trà Vinh). The coverage of such a proportion of ethnic minorities across the country 11 A workshop with experts in the field was organized in December 21, 2017. Further consultation was carried on continuously during study process. 12 The 53EMS indicates that the 62% of the Hoa population resides in urban areas (by 2015). Who are the Top-and Bottom-Performing Ethnic Minorities? 35 makes this study rather challenging. Consequently, it provides more diversified and comprehensive evidence that is important to the study’s credibility and the subsequent recommendations. TABLE 1: Overview of the Selected Top- and Bottom-Performing Ethnic Minority Groups Sán Dìu Mường Khmer Xơ Đăng Khơ Mú Mông Ethno Chinese Vietic Mon-Khmer Mon-Khmer Mon-Khmer Mông-Mien language Geographical Thái Nguyên, Hòa Bình, Sơn Sóc Trăng, Kon Tum, Nghệ An, Điện Hà Giang, distribution Vĩnh Phúc, La, Thanh Trà Vinh, An Quảng Nam, Biên, Sơn La, Tuyên Quang, of ethnic Bắc Giang, Hóa, Phú Thọ, Giang, Cà Quảng Ngãi Lai Châu Lào Cai, Yên groups Quảng Ninh, Ninh Bình Mau, Bạc Liêu Bái, Lai Châu, Tuyên Quang Sơn La, Cao Bằng Population 167,659 1,395,101 1,283,405 195,618 84,525 1,251,040 HDI 0.5868 0.5679 0.524 0.4494 0.4140 0.3852 MPI 27.09 40.77 34.5 82.83 91.72 88.17 Source: Authors’ compilation from different sources for ‘ethno language’, ‘main residing areas’; calculations from the 53EMS for population, HDI, and MPI 22. Using the 53EMS data available, a brief profile of these selected groups is reported in Table 2. 23. To contextualize the numbers, figures on the selected groups were reported with the average figures of all 53 ethnic minorities. The short profiles of the top performers represent a diverging story from those of the bottom performers. In terms of income per capita, for instance, the monthly income per capita of the Sán Dìu and Khmer is higher than the average of all ethnic minorities by nearly 30 percent. In contrast, the monthly income per capita of the Mông and Khơ Mú is less than half of the average income among all ethnic minorities. As a result, the poverty headcounts (using the GoVN income-based poverty applied to 53EMS) of the top performers are considerably lower than those of the bottom performers. With regard to other non-monetary aspects of the profiles, the best-performing groups appear to outperform the least-performing groups. This is especially evident when comparing the housing conditions, usage of electricity, access to safe drinking water, hygiene- improved toilets, and other selected household assets (e.g., television, motorbike, etc.).13 13 This study does not provide a separate (sub)section on descriptive statistics of the selected ethnic groups (though 53EMS is used to provide additional evidence to supplement the qualitative research when relevant). Instead, a brief profile is discussed in this section. Figures on socio-economic characteristics of the selected groups will be used to supplement the findings from the qualitative study, which are discussed in Section 4 of the report. A detailed narrative of different ethnic groups using 53EMS is available (CEMA, UNDP, and Irish Aid 2017; CEMA, UN Women, and Irish Aid 2017). Research Report 36 Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam TABLE 2: Brief Profile of the Top and Bottom Performers, 2015 53 EMS Sán Dìu Mường Khmer Xơ Khơ Mú Mông average Đăng Monthly income pc (‘000 VND) 1,161.4 1,504.3 1,188.9 1,529.4 687.3 511.7 575.2 % around the average income pc 0.0 29.5 2.4 31.7 -40.8 -55.9 -50.5 % Poverty (2015, income based) 23.1 8.5 18.6 14.8 44.6 59.4 45.7 % Near poverty (2015, income based) 13.6 11 19.7 10.3 11.4 13.5 13.4 Household size (# persons) 4.4 4.1 4.2 4.1 4.4 4.9 5.6 Life expectancy (# years) 72.1 73.22 72.37 72.86 70.5 69.24 68.97 % Households w/ semi-concrete house 70.2 72.6 65.4 54.2 80.9 60.1 81.4 % Households w/ temporary house 15.3 2.8 10.3 38.3 16.3 37.3 14.1 Using national grid electricity (%) 93.9 99.9 98.3 98 88.9 58.8 69.9 Having safe drinking water (%) 73.3 90.7 72.5 93.9 51.1 36.3 53.1 Having hygiene toilet (%) 27.9 29.6 27.9 36.7 10.3 4.4 7.0 Having motorbike (%) 80.7 89.6 82.3 68.6 59.8 50.9 75.3 Having refrigerator (%) 32.2 70.6 44.7 15.5 4.1 4.1 3.6 Having TV (%) 84.9 95.7 92 87.2 64.5 53.6 50.1 Source: Authors’ compilation from the 53EMS 3.3 Analytical Framework on Driving Forces for Soco-Economic Performances of Different Ethnic Groups 24. Given the selected top and bottom performers, the following questions are the focus of the qualitative component of the study: What are the reasons contributing to the varying performances of different ethnic groups? Why some ethnic minority groups have successfully managed to escape poverty while others have lagged far behind? To guide the data collection, an analytical framework for the study was developed. Accordingly, wellbeing of an ethnic minority household in Vietnam is conceptualized in the study as the outcome of interactions across a large number of forces, which could be classified into two layers of the so-called ‘direct’ and ‘indirect’ factors affecting wellbeing. These factors are illustrated in Figure 3.14 25. Layer 1: Direct factors of ethnic minority wellbeing. Being direct, the following factors contribute directly to different aspects of living standards of ethnic minority households. Some of these factors could be seen as ‘internal’ or ‘endogenous’ causes of the ethnic wellbeing. (a) Landholding (physical geography and quality of the landholding) is arugably one of the most important livelihood assets of a rural household. Being landless or lacking productive land are usually found as a major cause of chronic poverty among ethnic minorities. 14 It is noted that the classification into ‘direct’ and ‘indirect’ factors is rather arbitrary. As highlighted, wellbeing of a household is a final outcome of complex interactions between all of these factors and making clear-cut distinction between ‘direct’ or ‘indirect’ is not straightforward. However, this classification is useful for ‘mapping’ the factors that determine the living standards of the ethnic minority households and hence used as an analytical framework in this study. Who are the Top-and Bottom-Performing Ethnic Minorities? 37 (b) Labor (household workforce), knowledge and skills (educational attainment level, labor skills trained, and experience), health conditions. Each of these factors contributes to engaging in and determining access to and implementing income-generating activities. In fact, lack of labor, suffering from illness, and lack of required skills and knowledge are among the most oft- cited causes of poverty. (c) Remittances (either domestic or abroad). Having members or relatives sending home remittances means that the household has already had access to income-generating opportunities outside the residing locations. It serves as a social network to induce future migration of the household labor or relatives. (d) Access to credit (both formal and informal). Access to credit is a complicated factor. With regard to formal credit, the poor (and near poor) are subject to preferential loans from GoVN through different credit lines of the Vietnam Bank for Social Policies (VBSP). VBSP loans are made through trust loans with the mass organizations including Women Union, Farmer Association, Veterans Association, and Youth Union. With low interest rates (around 6.5 percent per year, depending on the credit lines), these official channels provide a subsidized and inexpensive access to credit. The effects on household income and assets however depend on whether loans can be used effectively. With respect to informal credit, which could be in cash or in kind (in terms of loans of inputs), the cost of these informal credit sources is generally high (very high in some cases), and this could drive the household into loss of their property that encompasses lands and other assets. (e) Inputs or subsidies. Some available inputs or subsidies to poor ethnic minority households are part of policies or programs that could be supported by GoVN, development partners, or private sector. This support could be useful for engaging in productive activities; but long-term effects depend on how the recipients utilize these subsidies and whether they could unintentionally create ‘dependency’ on impermanent support. (f) Mass organization membership (e.g., Women Union, Farmer Union, or collectives, production group). Household representatives could be eligible to access or benefit the resources being made available to mass organizations. This is closely linked to the preferential credits from the VBSP, which are only made available for local residents with proven membership in such an organization. (g) Gender roles and intra-household power dynamics. These factors affect the decision- making power of women and household labor division. This particularly reflects women’s perceived roles in house chores and other household livelihood activities. (h) Political connectedness. Whether household members, relatives, or kinship hold key positions in local administration could affect access to information and resources of a household. (i) Social networks and kinship. Access to different types of support mechanisms serve as informal support mechanisms for a household in coping with shocks or seeking opportunities, especially existing opportunities in the labor market. Research Report 38 Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam FIGURE 3: Driving Forces of the Social and Economic Development of Selected Ethnic Minority Groups LAYER 2 – INDIRECT FACTORS Labour market Physical Basic public and economic services connectivity LAYER 1 – DIRECT FACTORS Political Market Social network connectedness linkages GoVN policies and kinship and programs Gender roles Landholding and power Vulnerabilities Wellbeing of and shocks Labour ethnic minorities Mass organization membership Donor Knowledge, supported skills Inputs, projects subsidies Local Health Credit governance conditions Remittances Misperceptions Traditional and stigma institutions Culture, spiritual beliefs, religion 26. Layer 2: Indirect factors of ethnic minority wellbeing. Being ‘indirect’, these factors generally contribute to and/or affect those direct factors that determine the wellbeing of ethnic minority households. Some of these factors could be considered as ‘external’ or ‘exogenous’ causes of the wellbeing. (a) GoVN policies and programs, projects supported by development partners, and availability of basic public services. These factors that include education and health care, could potentially and positively correlate to socio-economic development of ethnic minorities. (b) Physical connectivity (geographical distance and transportation conditions from the residing location of the ethnic minorities to the main towns or central administration) and economic connectivity (connectivity to economic clusters). This connectivity potentially affects transportation duration and respective costs, availability of production inputs, public service utilization, and access to labor market opportunities. Given its significance, improving physical connectivity in remote ethnic minority communities has been the major focus of many GoVN- led investments for socio-economic development of ethnic minorities. (c) Labor market (both formal and informal). Access to wage employment has been one of the most salient factors to drive millions of rural people and ethnic minorities out of poverty. Promoting labor export is a target of a numerous national policies, including the National Target Program on Sustainable Poverty Reduction (NTP SPR). In addition to formal labor market, there Who are the Top-and Bottom-Performing Ethnic Minorities? 39 are opportunities in informal labor markets domestically or cross border. Cross border market however often deals with illicit activity. (d) Market linkages Market linkages refer to how ethnic minorities obtain their production inputs and sell out their outputs as well as how dependent they are on local traders, retailers, or shop owners. Market linkages are partly affected by physical connectivity. In many cases, market linkages of the poor could be manipulated or even blocked by other market actors such as concerned traders. (e) Vulnerabilities and shocks. Populated mainly in mountainous areas, many ethnic groups are vulnerable to landslides and other climatic catastrophes. Droughts are other risks encountered in the central coast or the Central Highlands. Salinity is a natural threat to many of the Khmer in the Mekong Delta. In addition, there are market shocks, especially in terms of price fluctuations with agricultural and livestock products; these uncertainties pose a high risk for crop and animal production. (f) Local governance and traditional institutions. Local governance is linked to decentralization and investment ‘ownership’ status, meaning the decision-making authority by local government. Traditional institutions are reflected in the participatory roles of village elderly, family clan chiefs, or religious leaders throughout the decision-making process at village level. These traditional institutions might also be embedded in a number of existing customary laws or other traditional communal management. (g) Culture, spiritual beliefs, and religion. Among these complex factors, religion is often associated with political sensitivity. Misinterpretations of culture, spiritual beliefs, and religion usually deepen the existing social misperceptions and stigma against ethnic minority population. More importantly, an inadequate understanding could be counter-productive to policies or other external support for ethnic minority development or social justice in general. (h) Misperceptions and social stigma. These factors attached to ethnic minorities can hinder ethnic groups from accessing opportunities, especially income-generating work. These also affect the ways in which basic public services or policies are delivered to ethnic minorities. 3.4 Sampling for the Qualitative Research 27. Based on the above-discussed analytical framework, a set of research instruments was developed for the qualitative research to capture all these identified factors (see Error! Reference source not found. for a set of research tools). These factors and, more importantly, interactions among them are the driving forces that underlie why some ethnic groups managed to get to the top-performing positions while others ended up at the bottom. 3.4.1 Selection of Research Sites 28. The identification of the top-two and bottom-two performers as well as the two groups in between served as the basis for study’s qualitative sampling selection. The overall guiding principles for the selection of research settings are indicated as follows: Research Report 40 Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam • The most ‘relevant’ province is selected for each group. This relevance is based on the share of the ethnic minority group of the selected province in the total population of that specific ethnic minority group nationwide and whether the province is perceived as a ‘traditional’ area of the ethnic minority group under consideration. • The selection of study settings should allow each selected group to be studied in different locations. Among the candidates of relevant provinces (as above), the one that has a greater number of the selected ethnic minority groups is preferred. • The selection should result in a reasonable number of areas so that the field work and the qualitative study in general are manageable in scope. Accordingly, a number of between 6-8 provinces was projtected by the research team. 29. For the case of the Xơ Đăng, Kon Tum and Quảng Nam provinces account for nearly 85 percent of total Xơ Đăng population. Therefore, both Kon Tum and Quảng Nam were selected as the main study sites for the Xơ Đăng. Regarding the Khmer, Sóc Trăng and Trà Vinh provinces are the home of nearly 57 percent of the Khmer population; therefore, these two provinces were selected for the Khmer. This is to allow greater geographical variation of the study sites. 30. Given this, the provinces selected for fieldwork were finalized and reported in the bottom row of Table 1. Accordingly, each of the selected top- and bottom-performing groups appears in at least two respective provinces (with the exception of Sán Dìu). In particular, the study sites selected for the Mường are Sơn La, Thanh Hóa and Kon Tum; some districts of Thái Nguyên for the Sán Dìu;15 Sóc Trăng and Trà Vinh for the Khmer; Kon Tum and Quảng Nam regarding the Xơ Đăng, Sơn La and Thanh Hóa with respect to the Khơ Mú; and Sơn La, Thanh Hóa, and Thái Nguyên as pertained to the Mông. 3.4.2 Samples for Qualitative Research 31. Given the provinces selected for the fieldwork (i.e., Sơn La, Thái Nguyên, and Thanh Hóa), two or three districts were selected and two communes were proposed per districts accordingly. The selection of these districts and communes were based on the geographical concentration of the ethnic groups under considerations. In addition, it allowed the study team to consult more than one group per district or commune when possible. 32. The selection took into consideration different types of districts and communes in the study sites. For districts, the study sites encompassed the poorest districts (the so-called ‘districts of 30a’ named after the Decree No. 30a on supporting the sustainable poverty reduction in the poorest districts), the second poorest districts (i.e., those categorized as Tier 2-poor districts according to the Decision No. 275/QD-TTg dated Mach 2018), and non-poor districts. 15 Ideally, another province should be selected as a second study site for the Sán Dìu; in that case, Quang Ninh was suggested as a potential candidate. However, selecting another province in addition to the seven selected prov- inces was constrained by the timeline and allocated resource of this study. Besides, at the selection of the ethnic groups, according to the 53EMS, there was a small Sán Dìu community (of around 400-500 persons) in Kon Tum. Therefore, it was expected to visit this Sán Dìu community during the fieldwork in Kon Tum. However, this consul- tation was not practical since this small group of the Sán Dìu was scattered in Kon Tum. Who are the Top-and Bottom-Performing Ethnic Minorities? 41 33. The list of selected communes consists of various types of communes, including extremely difficult communes (usually called Program 135 communes named after the Program 135, being the major GoVN poverty reduction initiative for the ethnic minorities); area-Tier2 communes with “extremely difficult” villages,16 bordering communes, extremely difficult, and coastal communes; and non-poor communes, the ones that have more or less accomplished the set goals for the acquired ‘New Rural Development’ status. Within each commune selected, the fieldwork took place in the two ‘in-focus’ villages. 34. For the Xơ Đăng and the Khmer, with two provinces selected as the study sites for each group, only one district was selected per province to keep the workplan manageable. In addition, the team only consulted with the provincial-level stakeholders in one of the two provinces for each group (i.e., Kon Tum for the Xơ Đăng, and Sóc Trăng for the Khmer). For the Xơ Đăng, apart from Kon Tum – the main province where 63% percent of the Xơ Đăng reside, Quảng Nam offered an opportunity to study a subgroup of the Xơ Đăng, who are also known as the Ca Dong locally. 35. Table 3 summarizes the study sites selected for the fieldwork. TABLE 3: Selection of the Study Sites for the Qualitative Study Sán Dìu Mường Khmer Xơ Đăng Khơ Mú Mông Thái Nguyên Sơn La Sóc Trăng Kon Tum Sơn La Sơn La (Đồng Hỷ/Nam (Phù Yên/Mường (Mỹ Tú/Phú Mỹ; (Đắk Tô/Kon Đào; (Mai Sơn/Hát Lót) (Bắc Yên/Xím Hòa, Văn Lang; Thải) Trần Đề/Đại An 2) Ngọc Hồi/Bờ Y); Vàng) Thanh Hóa Phú Bình/Bàn Thanh Hóa Trà Vinh Quảng Nam (Mường Lát/ Thái Nguyên Đạt) (Ngọc Lạc/Thạch (Trà Cú/ Định An) (Bắc Trà My/Trà Tén Tằn & Mường (Đồng Hỷ/Văn Lập) Đốc) Chanh) Lang) Kon Tum Thanh Hóa (Ngọc Hồi/Bờ Y) (Mường Lát/Pù Nhi) Note: Names of provinces are bold, names of districts are normal, and names of communes are italic 36. The final selection of sites for the study consists of 7 provinces, 13 districts, 16 communes, and 32 villages. The study team held consultations on average with 15 provincial-level stakeholders in each province, 20-25 district-level informants in each district, 10-12 persons in each commune, and 24-28 persons are consulted in each village. In total, the study team consulted 105 stakeholders at the provincial level, 260 informants at the district level, 192 at the commune level, and 832 at the village and household levels (of which 48.5 percent are female), making a total of 1,389 persons in the sample. Details of the study sites are provided in Annex. 16 According to the Decision QĐ50/2016/QĐ-TTg, dated 3rd November 2016 on Classification Criteria on extremely difficult villages, communes in the ethnic minority and mountainous areas, period of 2016 – 2020, Articles 2 and 3, the classification criteria are based on the weighted sum of 5 items: natural conditions and residential location, infrastructure, social factors, production conditions, and living conditions. Research Report 42 Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam 3.4.3 Organization of Field Trips 37. The team of 12 researchers from the CEMA Academy for Ethnic Minorities was selected to take part in the fieldwork led by two World Bank consultants. From this core team, six researchers eventually joined the two consultants for fieldwork in each of the provinces selected. The CEMA researchers participated in rotation so as to ensure that each of the research team members was able to conduct fieldwork in two provinces. Given this, the fieldwork was carried out by two sub-teams working in parallel, each sub-team consisted of three members of the core research team led by one consultant. 38. In terms of fieldwork organization, CEMA was responsible for logistics in organizing the fieldwork. Before the fieldwork starts, the CEMA core team was trained with the research instruments to pre-test the tools and to provide inputs for the tool improvements. The first leg of fieldwork (for the Mường, Mông, and Khơ Mú) was carried out from May 5 to May 16 in Sơn La. The subsequent field visits to other provinces were organized with a lag of less than a week between each field visits. This arrangement was made to ensure that the CEMA researchers could both engage in the study while being able to catch up with their job duties at the Academy. The final leg of fieldwork took place between July 5 and July 17 in 2018 in Sóc Trăng and Trà Vinh. For each of the fieldtrips, the team made sure to hold at least two on-site team briefings to reflect on preliminary findings and research observations, plus another end-of-the-fieldtrip team briefing on the final day of each trip. These arrangements aimed to ensure consistency and quality of data collection. It also served as a platform for the research team to share on their observations and reflections from the field. Remnant of Tranquility in Muong Hoa Valley Who are the Top-and Bottom-Performing Ethnic Minorities? 43 What are the Main Drivers 4 of Socio-Economic Performance Among Ethnic Minority Groups? This section summarizes research evidence across the key factors that are identified to have impact on the social and economic development of different ethnic groups. For analysis purpose, these factors are grouped under the following factors: (a) physical and economic connectivity; (b) basic household livelihood assets; (c) market linkages; (d) labor market mobility; (e) access to education; (f) relevant aspects of healthcare services; (g) traditional institutions and local governance; (h) gender roles and intra-household power dynamics; (i) perceptions of ethnicity and dynamics of trust; and (j) access to external support initiatives. Generally, each factor is found to have exerted influences differently across the top and bottom performers. The findings are organized by the 10 factors with a summary at the end of each sub-section. 4.1 Physical and Economic Connectivity 39. In the context of this study, physical connectivity indicates geographical proximity to the administration center, such as the main town of the district or the main city of the province; road networks; and the transportation conditions. Economic connectivity refers to connectivity to economic clusters such as urban areas or industrial parks. In terms of physical connectivity, Table 4 reports statistics on the average distance from the households of the selected ethnic groups to some main public facilities such as schools, hospitals, and local markets using the data available from 53EMS. TABLE 4: Distances to Main Public Facilities (km) 53 EMS Sán Dìu Mường Khmer Xơ Khơ Mú Mông average Đăng Distance to nearest primary school 2.5 2 2.3 1.6 2.3 4 4.8 Distance to lower secondary school 3.6 2.8 3 2.7 4.6 7.1 7.1 Distance to upper secondary school 11.7 6.8 9.3 5.3 18.6 28.7 23.3 Distance to hospital 16.7 10.6 16.3 8.6 21 31.5 27.8 Distance to health center 3.8 2.8 3.1 2.9 4.1 7.3 7.3 Distance to market 9.1 4.6 5.9 3.4 21.2 25.7 17 Source: Authors’ calculation from 53EMS Research Report 44 Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam 40. Top performers are likely to be better connected. Compared to other studied ethnic groups, physical connectivity is found to be most convenient to the Mường and Sán Dìu and the Khmer, which was identified as close to the top-performing groups. The average distances from the Mường, Sán Dìu, and Khmer to important facilities are considerably shorter than those encountered by the Xơ Đăng, Mông, and Khơ Mú. Notably, the Mông appeared to be the worst connected compared to all other selected groups (see Table 4). Out of 48,364 villages with ethnic minorities, the 53EMS indicated that 53.5 percent having tarmac roads, 28.5 percent have cement concrete roads, 8.4 percent having rock road, meaning that only 19.6 percent have dust roads that could be difficult for travel to the village center during rainy season. Improved transportation conditions are usually associated with an expanding electricity grid network. In fact, 93 percent of these villages were shown connected to the national grid network in 2015 (about the time when the 53EMS data collection took place). 41. There is no doubt that physical connectivity contributes to top performance by reducing transportation costs; and improving access to markets and, more importantly, to other economic opportunities with mainstream society. Rural transportation has been the focus of public investments as well as almost all designated programs and projects; as a result, road connectivity has been substantially improved. This gives weight to economic connectivity to urban areas or industrial parks, where ethnic minority youth can find opportunities in the wage labor market, and ethnic minority farmers can participate in some market value chains. In this regard, some small- scale industrial zones were observed in many rural and poor districts to take advantages of a cheap and abundant labor force in the local areas. For instance, Phù Yên district of Sơn La, Ngọc Lạc district of Thanh Hóa, Trần Đề district of Sóc Trăng are examples where labor-intensive factories in garment and footwear manufacturing absorb a significant labor force of ethnic minority youth and create at least short-term income flows. This connectivity to industrial clusters helps enable many ethnic minority women to commute to work, which would otherwise be impossible for them. For the ethnic minority youth, it seems that participating in the labor market is more attractive and technically easier than having their families engaged in some value chains. The qualitative study reveals that ethnic minority youths see migration as an avenue to improve their status, learn new skills, and feel a sense of pride. Staying at home typically limits their access to land which is legally and traditionally under the possession of their parents, coupled with their role as unpaid farm work rather than functioning as farmers in their own right. The main barriers to entering the wage labor market lie at their limited mobility and lower secondary education attainments. 42. Distance is a factor that impacts local perceptions about women taking up paid work outside their homes; it influences how households weigh the options allowing women to seek work away from the area where they live. Generally, the further away the jobs, the less likely women are to pursue them. There are differences between the best performers and the least performers seen from cross-group comparison. Among the top-performing groups such as the Khmer, Mường and Sán Dìu, the distance factor does not really play an important role in women’s participation in the labor market. However, there are women among these groups who are not prepared to work far from their homes. In such cases the local perceptions of gender power and gendered familial roles have impact on the labor-market participation of married women. For instance, results of the 53EMS indicate in Sán Dìu, of the locals who had relatively high employment rates in industry, women were 15 percent lower than men. Besides distance, decisions for working away from home include nature of the jobs What are the Main Drivers of Socio-Economic Performance Among Ethnic Minority Groups? 45 available; income expectations; and prior experiences, especially interpersonal relations with workers and management who most often are Kinh people). Security and safety are also factors; dangerous mountainous and forest roads may be insecure and unsafe for women and children. Ethnic minority women are less likely to use or own motorcycles, so travelling distances of over 10 kilometer is a challenge for them (CEMA, Irish Aid, and UN Women 2017). These difficulties cause ethnic minority women living in remote areas with limited use of transportation to be more dependent on men when accessing basic social services. 43. It should be noted that even within a top-performing community, there could be locations that are still poorly connected and thus correlate to higher-than-average poverty. For instance, Luong Thien hamlet of Thach Lap commune of Ngọc Lạc district is divided geographically by a river, which makes one part of this Mường village completely isolated in the rainy season. Whereas other Mường hamlets of Do Son and Do Quan in the same commune appear to possess more production materials such as forest land; but, the connecting dirt road from the commune center (about 10 km) often becomes impassable in wet weather, especially to wheeled traffic. Plus, electricity has only been available to the villagers of these two hamlets since 2017. In this regard, connectivity also matters at the village level. As a matter of fact, for many of the studied communes, transporting from the remote villages to the commune center could be as difficult as getting from the commune center to the district main town. 44. Poor connectivity tends to be associated with bottom performers. Poor connectivity was evident in all the studied villages among the Mông and the Khơ Mú as well as all selected groups as shown in Table 4. For instance, the Mông hamlet of Cài Nòi in Pù Nhi commune (Mường Lát, Thanh Hóa) had no access to electricity at the time of data collection in mid-summer 2018. For the poorly connected locations, infrastructure factors continued to negatively impact the local residents’ capabilities to access consumer market and education/training opportunities. This indicates that poverty in Vietnam is synonymous with remoteness (World Bank 2018a). High- altitude habitat, lack of transport facilities, and frequent bad weather particularly render the Mông villages highly isolated, depriving villagers of trading opportunities and social contacts with other groups. Weak infrastructure like narrow roads (less than 2 meters wide) hinder car access for goods transportation, resulting in higher prices for the Khmer of remote hamlets in Sóc Trăng. Seasonality problems emerge when the quality of transportation is considered. It is reported that Mường Lát (Thanh Hóa) is prone to landslides which in turn inhibit lowland companies’ willingness (especially investors with bigger stakes) to invest in the area. 45. In addition to poor physical connectivity, the bottom-performing groups also experience poor economic connectivity. The Mông in Sơn La (Bắc Yên district), Thanh Hóa (Mường Lát district) or the Khơ Mú in Sơn La (Mai Sơn district) and Thanh Hóa (Mường Lát district) are examples of poor economic connectivity. Even though physical road conditions have been improved in recent years (with exception of few villages in Hát Lót commune of Mai Sơn district), a far distance from these areas to the provincial economic clusters or industrial parks make it less favorable for the Mường in Sơn La (Phù Yên district) or the Sán Dìu in Thái Nguyên (Đồng Hỷ district) where ethnic minority women could commute to factories located in the district or the nearby district to work. Research Report 46 Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam 46. Geographic distances should be seen in a specific regional context. The notion of distance or level of remoteness among various groups varies and does not necessarily reflect the realities on the ground, considering factors like weather or the current conditions of infrastructure. Taking the case of the Khmer in the Mekong Delta for instance, Table 4 indicates that the Khmer appear to have the shortest distance from homes to nearest public facilities (e.g., 1.6 km to primary school, 2.7 km to lower secondary school, and 5.3 km to upper secondary school) in comparison with that of other studied groups. Despite the fact that the distance from homes and schools is rather short compared with other groups, the dropout rate from primary to secondary levels among Khmer children is quite high (Section 4.5 provides further discussions). Thus, the distance factor does not necessarily explain the high dropout rate among Khmer children. This suggests that one should not consider ‘distance’ only in terms of physical geography but should include social and cultural elements as well. 47. Physical and economic connectivity in summary. The top performers are better connected in both physical and economic terms. Improvements in physical connectivity have been made for all groups, even the bottom-performing Mông or Khơ Mú. This gives more weight to economic connectivity where the gap in access between the top and the bottom performers is even more pronounced compared to that gap in physical connectivity. Having private investment, especially industrial zones, in the local areas contribute greatly to economic connectivity for ethnic minorities. Having physical and economic infrastructures is important for connectivity. But gender, culture, and perceptions also contribute in ‘connecting’ ethnic groups to these infrastructures. Better connectivity enhances economic empowerment for women, who would otherwise hardly find opportunities in the wage employment market. 4.2 Basic Household Livelihood Assets 48. Livelihood assets encompass a broad range of factors (under the Sustainable Livelihood Framework, for instance). Within the scope of this analysis, basic household livelihood assets in this sub-section fall under landholdings and household labor force. In addition, vulnerabilities and shocks affecting the livelihood assets and strategies are also addressed in this sub-section. 4.2.1 Productive Landholdings 49. In terms of arable land possession, the bottom-performing groups are better endowed than the top-performing ones. The average landholdings of the Khơ Mú and Xơ Đăng are nearly double those of the Mường or the Khmer. The Sán Dìu – identified as the best-performing group – possess the smallest amount of arable land. Other 53EMS figures indicate that only 7.5 percent of ethnic minority households revealed that lacked land for livelihood activities. Average landholdings of the selected ethnic groups are illustrated in Table 5. The total land areas indicated in Table 5 are not informative in terms of different types of landholdings or productivity of land, which are important for determining earning from landholding.17 Therefore, the following analysis relies upon findings of the qualitative study. 17 The 53EMS does not provide detailed information on different types of landholdings and hence it is not possible to further disaggregate household landholding in this table. What are the Main Drivers of Socio-Economic Performance Among Ethnic Minority Groups? 47 TABLE 5: Household Landholding of the Selected Ethnic Groups 53 EMS Sán Dìu Mường Khmer Xơ Đăng Khơ Mú Mông Average household landholding 7,268 2,417.4 5,066.8 4,724.2 9,600.6 9,248.8 4,229 (square meters) Source: Authors’ calculation from 53EMS 50. Earnings from productive lands contribute significantly to poverty reduction. Evidence from the qualitative research generally suggests that top performers such as the Mường and the Sán Dìu appear to be well endowed in terms of fertile landholdings (the total landholding areas, as indicated in Table 5, could be lower than those of the bottom performers). The Sán Dìu in Bàn Đạt commune (Phú Bình, Thái Nguyên) could diversify into annual and perennial crops while the Mường in Ngọc Lạc (Thanh Hóa) and Phù Yên (Sơn La) utilize their arable lands to diversify their crop choices, including new crops that have been successfully piloted in other places (i.e., orange tree was introduced as a new cash crop in Ngọc Lạc district of Thanh Hóa). However, landholding size does not always correlate with the best-performing groups. For instance, among the six groups selected for the qualitative study, the Khmer was found with the largest landholding size (by general average). But, land productivity of Khmer was found to be low due to subsequent high costs of hiring extra farmworkers. While the Khmer does not routinely practice labor exchange, it is found common among other ethnic groups in the North and Central Highlands. 51. Among the bottom performers, fertile lands are limited. Lack of fertile cultivation land is seen as the main cause of poverty, which passes from one generation to the next and referred to as ‘sustainable poverty’ (nghèo bền vững) or ‘inherited poverty’ (nghèo di truyền) by some Khmer informants. This in turn limits the opportunity or access to livelihood diversification for some; others with with sufficient land are able to invest in other sources of revenue like shrimp farming or raising special crops (i.e., taro in Trà Vinh). In the Mekong Delta, poor households, especially those owning no land, indicated in interviews with the study team that their only source of income comes from work as hired farm hands in neighboring areas (earning about 150,000 VND per work day). Recent farming mechanization has generally affected their livelihoods. The introduction of the mechanical harvester has wiped out a source of seasonal income for women, who are often hired at the harvesting stage, as reported among the Khmer in Trà Vinh.18 52. Lack of finance is a barrier for poor farmers making more efficient use of their land. For instance, the poor Mông households in Thái Nguyên see quite clearly the profitability of tea cultivation, but they cannot afford to rent machines for clearing land in order to expand their cultivation area or to buy new tea seedlings. Or some poor Sán Dìu households cannot afford the required investment to utilize fertilizers to boost their tea output. 53. There are still ways for the better-off to expand their arable land. For instance, the better- off Khơ Mú in Sơn La or Thanh Hóa accumulated more land by subcontracting to work on land of 18 ILO (2016) in a report on the Impact of Automation for the ASEAN-5 (Cambodia, Indonesia, the Philippines, Thai- land and Viet Nam) indicates that around 56 % of all employment has a high risk of automation in the next couple decades. Among the ASEAN-5, the share of jobs with a high probability of automation is lowest in Thailand (44 %) and highest in Viet Nam (70 %). Research Report 48 Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam their brothers or brothers in law – but not from outsiders – to grow corn, cassava, elephant grass and sugar cane. For the Mông (Bắc Yên, Sơn La) there are possibilities to expand rice-growing land through forest clearance, soil restoration, or creating terrace fields. In certain places such as Pù Nhi commune (Mường Lát, Thanh Hóa) or Xím Vàng commune (Bắc Yên, Sơn La), some Mông households reported earning substantial cash from engaging in some indigenous perennial crops such as tea and plum trees. Land productivity among the bottom performers also depends on whether they have learned the cultivation techniques of the more productive neighboring groups such as the Thái, Tày, Kinh, or Hoa. This may be related to the ways in which various groups interact with the production process. For example, the Kinh in Kon Tum (who migrated from Thái Bình province) use manure to enrich the soil in coffee planting, a practice promoted by the local authorities but was met with reluctance by the neighboring Mường. This is due to the Mường’s cultural perceptions of the body as permeable and therefore vulnerable to “dirty air” such as bad smells from human feces (Rheinlander et al. 2010). 54. Productive landholdings correlate with displacement and resettlement experiences in the past. Field observations among five ethnic groups (the Mường migrants and the Xơ Đăng in Kon Tum, the Ca Dong in Quảng Nam, the Khơ Mú in Thanh Hóa, and the Mông in Thái Nguyên) indicate that problems surrounding land resources and livelihoods arising from resettlement policies and related programmes vary from top-performing groups to the bottom-performing groups. Taking the case of the Mường in Kon Tum, for instance. The Mường — a top performer that migrated to Kon Tum from Đà Bắc district when the Hòa Bình Dam was built in the early 1990s — not only benefited from considerable state resettlement assistance but were able to utilize large areas of forest land in the early stages. Together with this head start, their cultural affinities with the Kinh in neighboring communities and their expertise in paddy and fruit tree cultivation enabled these Mường migrants to build a stable and rather prosperous community in the Central Highlands. 55. As performers near the bottom, the Xơ Đăng in Ngọc Hồi of Kon Tum and the Ca Dong in Bắc Trà My of Quảng Nam, had experiences with resettlement and loss of landholding. Their hamlets used to be located around the border post of Bờ Y. Since their livelihood relied on swidden, slash-and-burn cultivation, they often ventured into Laos territory to exploit the forests there. With the government sedentarization policy, the Xơ Đăng were resettled in fixed hamlets and were given land to take up paddy rice cultivation. Data from 53EMS indicate that the prevalence of slash-and- burn cultivation accounted for only 0.055 percent by 2015. However, many Ca Dong residents were unwilling to adopt to the new mode of livelihood and sold their allotted land to the migrant Kinh and resumed their traditional technique of swidden, slash-and-burn cultivation for their livelihood. When both the Vietnamese and Laos authorities “closed” (đóng cửa rừng) the forests, denying the Ca Dong of their traditional livelihood, these households returned to their hamlet and left with little alternative but to fall back on work as hired laborers for the Kinh or other ethnic groups on the land they once owned. 56. The Ca Dong in Nam Trà My District of Quảng Nam represents a case of involuntary resettlement caused by the recent boom in hydropower. The village 1 in Trà Đốc commune of Nam Trà My was resettled to its current location for the construction of Sông Tranh II hydropower. As part of the hydropower investment, basic infrastructures of the villages (access road, water supply, electricity, and the village common house) were constructed by the hydropower investor. What are the Main Drivers of Socio-Economic Performance Among Ethnic Minority Groups? 49 This involuntary resettlement resulted in the loss of paddy land, which was already limited among the Ca Dong before the hydropower construction. Acacia plantation was central to the income restoration program. Though acacia could provide a source of income, its short production cycle (to be harvested for paper industry) is five years, and many households found it difficult to find other short-term cash crops. 57. Likewise, the Khơ Mú in Mường Lát (Thanh Hóa) often crossed over to Laos to collect forest products and to trade goods at the market. Cross-border marriage among people of the same ethnic group is relatively common (though the 53EMS reports a national average of 0.1 percent women married foreigners) in the research settings. The two hamlets of Lách and Đoàn Kết of Tén Tẳn commune in Mường Lát have been set up since 1992 under the government policy of ‘hạ sơn’ (literally “coming down the mountain”). Similar to the Ca Dong, the Khơ Mú practice of slash-and-burn cultivation on mountainous slopes used to be considered as a form of “primitive production” mainly existing among “backward” peoples living in isolated and remote mountainous areas (Nguyễn Văn Chính 2008: 44). To many Kinh officials, this type of farming needs to be replaced with sedentarization along with new forms of agricultural production in order to improve quality of life and enhance the quality of the environment. However, decades after sedentarization and the introduction of wet rice cultivation, many Khơ Mú still exploit the forested slopes for a living. 58. Productive landholdings in summary. Total landholding owned by ethnic minorities could be a poor indication of livelihood assets. In fact, land productivity matters. The top performers such as the Mường and Sán Dìu are better endowed with fertile land compared to the bottom performers and their land usage pattern is also more diversified. The bottom performers are less endowed with fertile land and lack diversification in crop choices. Some groups were worse off due to displacement or resettlement having lost their most fertile (and sometimes limited) land and lacking the capability or being reluctant to adapt to new crops on their newly compensated lands. 4.2.2 Household Labor Force 59. As indicated earlier in Table 2, the bottom performers, especially the Mông, exhibit larger household sizes compared to the top performers. The 53EMS provides some useful information about ethnic minority participation in the labor force. The hardships that bottom- performing groups encountered have probably forced all adults to seek some form of economic activities. The ratio of bottom-performing groups ages 15-24 who were economically active was found to be greater to top-performing groups. 60. Table 6 reports the percentage of people age 15 or higher working at the time of the survey (and disaggregated figures by sex and age groups). It shows that the bottom- performing groups are more ‘active’ in the labor force. In fact, 94.8 percent of the Mông, age 15 and older, were working at the time of 53EMS while the corresponding figures of the top performers were 87.1 percent of the Sán Dìu and 89.1 percent of the Mường, and just 75.4 percent of the Khmer. Results of 53EMS further show that in 10 occupational groups, most ethnic minority workers participate in simple labor at 67.7 percent, followed by skilled labor in agriculture and forestry at 17.6 percent, manual labor at 4.9 percent and sales and service at 4.4 percent. Very few ethnic minority workers Research Report 50 Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam undertake occupations requiring medium- to high-level knowledge and skills like management labor or medium- and high-level expert. The Khmer was found to be one of ethnic minority groups that had a high rate of employment in service sectors (17.4 percent) as well the highest share of women employed in service sectors (nearly 22 percent). TABLE 6: Working by Sex and Age Groups of the Selected Ethnic Groups 53 EMS Sán Dìu Mường Khmer Xơ Đăng Khơ Mú Mông % people at age 15+ working 86. 1 87.1 89.1 75.4 90.8 93.2 94.8 By sex Male 52 53.2 51.8 55.1 50.4 50.3 50.4 Female 48 46.8 48.2 44.9 49.6 49.7 49.6 By age group 15-24 24.8 22.3 20 18.8 33.4 33.6 36.6 25-54 64 67.4 66.8 67.2 56.2 59.1 55.6 55-59 5.2 5.2 6.4 6.1 4 3.2 3.3 60+ 6 5 6.8 8 6.4 4 4.5 Source: Authors’ calculation from 53EMS 61. Observations from the field suggest that an adequate work force taking part in the production process is an important factor that helps some ethnic minorities rise to the top. Poverty is usually associated with lack of adult labor or when main laborers in the household were subject to medical problems or disability. These households are usually targeted by social protection policies, and poverty reduction is seen to be challenging to these beneficiaries. This type of poverty could be observed in all researched ethnic groups, regardless of how well they were performing. In fact, local officials tend to avoid targeting these households within poverty reduction policies. This was made on the assumption that without an adequate labor force, it is not possible for these households to pursue productive livelihood activities (that are supported by the respective policies). 62. Economic deprivation among households seems to be proportionate to the number of children not yet working age. This situation is often observed in some ethnic groups having a high birthrate due to the absence of birth control for religious reasons like the Xơ Đăng in Kon Tum, or because of preference for sons over daughters among the Mông in Sơn La, Thái Nguyên and Thanh Hóa. Both groups have a relatively high proportion of poor households. In those households, there are grown children able to supplement the household income by shouldering some agricultural tasks like herding cattle, or working outside the home as factory workers, especially among school dropouts at 2nd or 3rd levels of secondary. This helps explain why households considered to be under the ‘poor’ category are often those having children still at school age and thus below legal working age and illegible for paid jobs. 63. The multigenerational mode of co-habitation constitutes an important supplementary labor force in the production process, especially in the Mông areas where the practice of exchanging labor is common. Figures from 53EMS indicate that the Mông is characterized as the ethnic minority group having the largest households, each on average consisting of 5 or 6 members What are the Main Drivers of Socio-Economic Performance Among Ethnic Minority Groups? 51 living under the same roof. In the context of the market economy, the mobilization of supplementary labor force from kinship may help keep production costs down but may not always help increase productivity when these family members lack the skills required of the labor market. 64. Household labor force in summary. The bottom-performing groups exhibit the highest rates of labor market participation, which is likely driven by extreme hardship. School dropout to supplement the household economy is frequent among bottom performers. Exchange labor was also found common among some of the least-performing groups such as the Mông. Exchange labor in practice seems useful to keep the production cost low. Regardless, whether being on the top or at the bottom, the chronic poor households are usually the ones that lack labor force and should therefore be offered social protection. 4.2.3 Vulnerability and Shocks 65. Market price fluctuations represent a major type of shock that ethnic minorities have encountered in their major livelihood activities. At the time of fieldwork, the Mông in Bắc Yên district and the Mường in Phù Yên district (Sơn La) suffered from decreasing prices in maize as one of their two major crops (maize and rice). The Khơ Mú in Mai Sơn (Sơn La) experienced price fluctuations in sugar cane (in spite of having a factory located in the district that sourced from the Khơ Mú farmers). Just as the Sán Dìu in Phú Bình district (Thái Nguyên) were affected by decreasing prices in tea leaves, the Khmer in Trà Cú (Trà Vinh) reported decreases in the price of sweet potatoes (local variety). In terms of livestock, many of the households from all studied ethnic minority groups reported that pig raising, which was one of the most popular livestock options a few years ago, was no longer the choice (even for some local varieties that could still find niche markets) due to sharp reduction in prices in recent years. 66. Market shocks exerts negative influence on all the studied ethnic groups, but the top and bottom performers responded and coped with the shock differently. For the top performers such as the Sán Dìu in Thái Nguyên or the Mường in Ngọc Lạc (Thanh Hóa) and in Phù Yên (Sơn La), decreasing prices of maize, pigs, or tea induced the switch to other crops. Thạnh Lập commune of Ngọc Lạc district, for instance, invested in new perennial crops such as orange and pomelo. Notably, the ethnic minority youth of these top-performing groups sought wage employment opportunities. Some of them rented their productive land to those who stayed in the locations. The Khơ Mú or the Mông households indicated their difficulties in switching to other crops. The Mông in Xím Vàng commune of Bắc Yên (Sơn La) or Văn Lang commune of Đồng Hỷ (Thái Nguyên) continued to grow maize as they realized there were no other choices available given the land quality and their technical knowledge. 67. The bottom-performing ethnic groups are more prone to natural disasters. Natural disasters were not found to be as major a risk to the top-performing Sán Dìu or Mường. For the Khmer, salinity was reported to damage as much as 30 to 50 percent of the total paddy rice areas of the Khmer interviewed, but that serious salinity happened only in 2016. At the time of field trip to Sóc Trăng and Trà Vinh, salinity was no longer a risk given the rain level of the current year had been high. For the Mông and Khơ Mú, natural disasters appeared to be more problematic. In many areas of the Mông locations, landslides under heavy rain are a high risk, particularly in the hilly locations where many Mông reside. Serious drought was also reported by both the Mông and Khơ Mú, and to Research Report 52 Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam some extent the Xơ Đăng in Ngọc Hồi district (Kon Tum). Due to their geographical location, some of the ethnic minority groups are more prone to natural disasters such as typhoons causing flash floods and landslides. The bottom-performing groups while being more prone to natural disasters also have less capacity to cope with natural disasters compared to the top-performing groups. The top groups can rely on their generally solid social capital built on family and other relations that can be of great importance in the post-disaster relief and reconstruction. Suffering from natural disaster, many bottom-performing households fell into indebtedness to local money lenders. Unable to pay back their debts at high interest rates, as observed among the Khơ Mú in Mai Sơn district (Sơn La), some farmers had to sell their production resources, including land, thereby depriving themselves of their very livelihoods. 68. The high dependency on natural resources as a major livelihood strategy makes some bottom-performing groups more vulnerable. The case of Xơ Đăng in Bờ Y commune, Ngọc Hồi district (Kon Tum) illuminates a typical situation. For a few decades, the Xơ Đăng in that commune relied on illegal logging for timber as well as non-timber forestry products, which were mainly wild animals for sale. Rice and other staple crops (such as maize or cassava) were also cultivated for (mainly) household consumption. This was also given as a major reason for an influx of migrants to the commune. However, since the GoVN ‘closed’ the natural forest in 2016, this major livelihood of the Xơ Đăng was controlled and reduced to a modest degree at the time of the study. This proved difficult for many Xơ Đăng households. Over the past few decades, they have been heavily dependent on illegal logging activities, which were terminated. Many of them have sold their productive lands to the Kinh migrants (or some Mường migrants). As a result, many Xơ Đăng residents had no other choices but continued their illegal logging, which had become increasingly difficult. Some migrants moved to other districts over the past two years, causing the population in the Lek village of Bờ Y commune to decrease from more than 500 households to only 320 households at the time of the study. 69. The case of Khơ Mú community in Tẻn Tằn and Mường Chanh communes of Mường Lát district (Thanh Hóa) illustrates a similar story of vulnerability. In Đoàn Kết village of Tén Tằn commune, after the natural forest had been closed, the border guard forces and forest protection established tight control over illegal forest logging. To help the Khơ Mú change their livelihood strategy, a governmental support program on chinaberry or bead tree (i.e., xoan đào) cultivation was initiated in certain areas. While this program would offer an alternative source of income for Khơ Mú residents, increasing engagement in forestry plantation could only generate income after ten years; hence, the Khơ Mú have continuously struggled to secure other source of income (rather than from illegal logging). 70. Vulnerability and shocks in summary. Market shocks (especially in terms of price fluctuations) affect all ethnic groups. The top performers usually respond by diversifying to new crops while the bottom performers fail to diversify due to their lack of productive land and unwillingness to change. The bottom performers are found to be more prone to natural disasters, which could then lead to indebtedness to informal credit providers; that usually leads to transfer of land and assets to the money lenders. The top performers are usually less prone and more capable of coping with natural disaster. Heavy dependence of the bottom performers on natural resources makes them more vulnerable to both natural disasters and the shocks caused by tightening natural resource management (such as forest). What are the Main Drivers of Socio-Economic Performance Among Ethnic Minority Groups? 53 4.3 Market Linkages 71. Market linkages are critical for getting access to cash crops and household livelihood strategies. The top-performing groups appear to be better linked to the market and vice versa. The physical connectivity obviously attributes to the establishment of these market linkages. Table 4 indicates that the average distances to physical markets of the Mường, Sán Dìu, and Khmer are considerably shorter than those encountered by the bottom-performing groups such as Mông or Khơ Mú. Observations from our fieldwork reveal that having physical access to market locations does not correlate to households’ integration into cash crops. In fact, many markets were constructed but are poorly utilized. Having indigenous high-value crops such as specialty tea or temperate climate fruits seems to bring the market linkages to some remote villages. Seen from gender perspectives (observed in both top- and bottom-performing groups), males generally assume the task of dealing with outsiders when marketing household products; the male family members are more apt to speak the Kinh language coupled with their physical endurance and mobility for riding motorbikes over long distances. But, another reason for lacking access to the market is attributed to the pervasive prejudice toward the bottom-performing groups by the majority of Kinh people who mostly control the lowland markets. In this regard, males among the top-performing groups might be less likely to be subjected to such ethnic prejudices. Moreover, the degree of market power exercised by local traders, retailers, or shop owners, especially those who happen to be private money lenders, seems to be a major impediment to ethnic minorities’ capacity to shape their market linkages and livelihood strategies. 72. Cash crop agriculture is one of the main routes out of poverty in all studied provinces. The amount of cash cropping varies widely, from virtually none in upland areas [i.e., among the Mông in Bắc Yên (Sơn La) or Đồng Hỷ (Thái Nguyên)] to nearly all production in the lower-lying, but still highly sloping land of Mai Sơn among the Khơ Mú (e.g., sugar cane) in Sơn La, the Sán Dìu (e.g., tea) in Thái Nguyên, the Mường and Xơ Đăng in Kon Tum (e.g., coffee, rubber) or the Khmer (e.g., taro) in the delta area of Trà Cú (Trà Vinh) or Trần Đề (Sóc Trăng). The top-performing groups and the better-off households (regardless where they are in the top- or bottom-performing groups) appear to be capable of diversifying most of the cash crops available. Besides, a number of households have successfully diversified into high-value perennial crops such as Shan tea (the Mông) or, to some extent, coffee (the migrant Mường and some Xơ Đăng in Kon Tum). For these cash crops, selling at the farm gates is the dominant marketing arrangements. However, for many of the cash crops, selling through a couple of retailers was not seen as problematic as long as there are no (informal) bidding rules on farmers selling obligation (to certain suppliers). 73. Even for top performers, participating in the key agriculture value chains remains limited. In the fieldwork settings, coffee, rubber (in Kon Tum) and aquaculture (catfish, prawn) in Sóc Trăng and Trà Vinh are among the key agriculture value chains of Vietnam. However, ethnic minorities’ participation in these key value chains was rare. Take the case of the coffee chain in Đắk Tô district (being one of the two main coffee-producing districts of Kon Tum), only a few migrant Mường reported to be engaged in this value chain. Whereas, there is a rapid growth of small-scale rubber plantation (i.e., ‘cao su tiểu điền’) in Ngọc Hồi; all interviewed Xơ Đăng reported not partaking in that Research Report 54 Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam development activity in the district (even though the study team noticed some Xơ Đăng workers in the rubber fields in the district of Đắk Tô at the time of the study). Surrounding by arguably the most potential areas for aquaculture, the poor Khmer usually find themselves as farmhands in catfish or prawn commercial farms owned by the Kinh investors. 74. In this context, study observations indicate that the Mường and Sán Dìu are the top performers that have tried to ‘upgrade’ their livelihood strategies to engage in some high-potential value chains. For instance, several Sán Dìu in Đồng Hỷ and Bàn Đạt districts of Thái Nguyên have invested in free-range chicken farming. And some have managed to rise to medium- or large-scale suppliers of a few thousand chickens for various supermarkets and food chains. The Mường in Ngọc Lạc district of Thanh Hóa have tried with orange, citrus, and pomelo as perennial crops. Some migrant Mường in Kon Tum, who have been migrating to the Central Highlands for around two decades, have successfully engaged in coffee or rubber value chains. 75. Whether bottom performers engage in high-potential value chains depended partly on the availability of some valued indigenous crops. For instance, the Mông in Bắc Yên district of Sơn La have gained a stable source of income from Tà Xùa tea, and Son Tra (a sort of forestry fruit that can be used in pharmaceutical industry). Some Mông in Mường Lát of Thanh Hóa have also earned side income from temperate climate fruit trees such as plums and pears. Mông in Bắc Yên district of Sơn La is known as one of the best Shan tea areas in the country, or Sơn Tra (Mông apple). These products are not considered as key agriculture value chains nationally, but locally they help to create an important source of income for some Mông households. 76. Informal credits, both in production inputs and instant cash provided by local traders and retailers, could lead to unexpected setbacks to fair competition. In many cases, ethnic minority farmers tend to owe traders and retailers for input supplies; and hence at harvest they are bound to repay the debt. This informal credit link is usually at the expense of poor farmers. The traders, retailers, or “investors” (called by Khơ Mú residents in Sơn La) usually impose the output prices to their advantage; this is possible due to the pre-existing loans that tie the indebted farmers to them. There seems to be a vicious circle of indebtedness of poor households in all provinces visited for the study. Emergency loans often come with high interest: a million VND loan may require payments of VND 150-300,000 per month, “being tightly bound” (bị thắt ngặt) in the Mekong Delta. 77. However, good road (and river) connectivity is likely to make some Khmer farmers (in Trà Vinh) less dependent on local retailers or shop owners (mainly the Kinh and the Hoa). Some Khmer farmers said that they keep track of market prices through television or the internet. They indicated that there was almost no sizable difference between prices paid by local retailers and those paid by traders elsewhere – a result of price equalization thanks to private funding through relatives. This is different from the situation in the North where ethnic minority farmers, especially those in the bottom-performing groups, are usually forced to sell their products at pre-fixed prices to the traders or shop owners, following the borrow-now, pay-back later formula. 78. The market power of shop owners and traders is generally weaker in areas with better connectivity, physically and economically. In areas with good connectivity, the top performers What are the Main Drivers of Socio-Economic Performance Among Ethnic Minority Groups? 55 usually have competing choices in terms of where to buy inputs and sell outputs. Such options keep the margins and interest rates charged by shop owners and traders relatively competitive. Access to information is an important reason that contributes to this bargaining power of the top-performing groups. Table 7 provides the different aspects of access to information using 53EMS. It is evident that the top-performing groups have experienced better access to information in all indicators used having television with access to the broadcasting programs Voice of Vietnam or Vietnam Television. With respect to cross-group comparison among the least performers, the Mông appear to have the higher rate of (cell) phone possession (55.9 percent) than the other two groups, Xơ Đăng (38.9 percent) and Khơ Mú (33 percent). The Mông can be considered more mobile in terms of means of access to information. Whereas the Xơ Đăng prefer to rely on television as the means of access to information. Are there any associations between these ethnic preferences and the ways in which cultural sensitivity has been addressed in existing communications for rural development? The current 53EMS dataset does not provide information on this dimension. 79. In a related note, the study team observed that collective buying or selling was rarely practiced by either the top or bottom performers. Despite efforts from both GoVN-led programs and donor-supported projects to establish production groups (e.g., shared interest groups), farmers seem to keep doing their livelihood activities on an individual basis. This tendency, in turn, makes them more vulnerable to market power exercised by traders or agri-input suppliers. Further in-depth study is required to look into the ways in which this phenomenon has been linked with trust within and among people of the community and, more importantly, to what extent the trust-building mechanism has been colored by existing perceptions of ethnicity and the major interactions between the upland groups. TABLE 7: Means of Access to Information 53 EMS Sán Dìu Mường Khmer Xơ Đăng Khơ Mú Mông Households having television (%) 84.9 95.7 92 87.2 64.5 53.6 50.1 Households with access to 65.1 79.8 66.9 78.1 65.4 20.1 43.3 information from VoV (%) Households with access to 88.8 97.5 95.5 87.7 81.5 56.4 60 information from VTV (%) Individuals having phones (fixed or 75.4 90.5 85.8 72.4 38.9 33 55.9 cell) (%) Households with computer (%) 7.7 6.6 5.1 4.8 1.4 1 1.7 Households with broadband internet 6.5 7.9 3.7 4.1 0.5 0.1 0.6 (%) Source: Authors’ calculation from 53EMS 80. The issue of manipulation of market and prices is reported to be more significant in Mường Lát district (Thanh Hóa), and among the Khơ Mú in Sơn La. A group of lowland Kinh monopolizes a whole collecting system, setting low prices for producers across the district as well the two neighboring districts of Quan Hóa and Quan Sơn. It is reported that the selling price (set by these Kinh traders) was much lower than the production costs so local farmers, including both the Thái and the Khơ Mú of Mường Lát district, have been forced to quit large-scale maize production since 2016. While the lowland Kinh traders dominate the links to market, there are some cases where ethnic Research Report 56 Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam traders or shop owners have tried to compete. However, the lowland Kinh traders were reported to be more ‘well connected’ and thus could easily push the ethnic competitors in corners (such as the case of some Mông grocery shop owners in Xím Vàng commune, Bắc Yên/Sơn La). Adding to these unbalanced practices, high transportation costs and seasonal availability make the bottom- performing groups even more dependent on traders (mainly dominated by the lowland Kinh). 81. Market linkages in summary. Agriculture cash crops remain a main route out of poverty for all ethnic groups that remain largely ‘outside’ the key agriculture value chains of the country. The top performers such as the Mường or Sán Dìu have tried to engage with some potential cash crops that are not (yet) dominated by the Kinh in the local markets. For the bottom performers, engaging in value chains is observed only for some unique indigenous crops (such as Shan tea, Mông apple etc.). Selling at farm-gate prices is the practice adapted by all ethnic groups, but traders (and input suppliers) in the areas resided in by the top performers charge relatively low margins due to market competition. In contrast, the bottom performers are over-reliant on traders and shop owners, who are usually the lowland Kinh, for buying inputs or selling outputs at high margins. At times, the bottom-performing households end up with high indebtedness to these agents; this often leads to transfer of land and assets in lieu of cash repayment for the high-interest loans. 4.4 Labor Market Mobility 82. Labor market opportunities have been increasingly crucial for poverty reduction in the ethnic minority areas over the past five years. While agriculture remains the major source of employment for ethnic minorities, there has been increasing opportunities in the labor market, both formal and informal, especially for ethnic minority youth. Figure 4 shows the relative importance of different income sources for ethnic minority households using 53EMS. Nonfarm income sources are found to be more important than income generated from agriculture production. On average, wage and other non-wage nonfarm income account for about 50 percent of the total household income for all ethnic minorities, while the corresponding figure of agriculture is 42.1 percent. More importantly, Figure 4 indicates that the top performers earn the majority of their household income outside agriculture while the bottom performers remain heavily reliant on agriculture as their main source of income. FIGURE 4: Structures of Household Income Mông 14.3 1.5 75.8 8.3 Khmu 16.8 1.2 74.7 7.3 Xơ Đăng 27.9 0.9 59.0 12.2 Khmer 49.9 9.7 30.7 9.7 Mường 47.5 7.1 38.2 7.3 Sán Dìu 48.0 12.9 34.6 4.5 53 EMs 39.6 9.9 42.1 8.4 0 20 40 60 80 100 Wage employment Non-farm non-wage employment Agriculture income Other sources Source: Authors’ calculation from 53EMS What are the Main Drivers of Socio-Economic Performance Among Ethnic Minority Groups? 57 83. With regard to wage labor market, there are increasing opportunities for factory work either in the districts or nearby. In almost all the studied districts or in other provinces, there exists a range of manufacturing factories or industrial parks. These are manufacturing factories mainly in electronics, garment, footwear, and aquaculture processing (the later applicable only for the Khmer in the Mekong Delta). These formal wage labor opportunities are mainly available to the ethnic minority youth who attain at least secondary education. From policy perspectives, these labor market opportunities also cover a scheme on Vietnamese labor export, which has been set up as the objective of GoVN policies, especially within the National Target Program on Sustainably Poverty Reduction (2016-2020) as well as its previous phase. With respect to the informal labor market, private small and medium-size enterprises, household businesses, and farms offer opportunities for ethnic minorities, either young or middle age, in manual labor, including civil engineering. The informal labor market also includes migrants who illicitly cross the border (with China, Laos, or Cambodia) to engage in factory work or farm employment. All studied groups, ranging from the top performers to the bottom performers, recognize the emerging importance of labor market mobility even though the nature of labor market jobs may differ greatly across ethnic groups. 84. For the top performers, there are usually available options for wage employment either in nearby locations or other provinces. The Mường, Sán Dìu, and Khmer youth particularly who have completed their lower secondary schools or higher educational level can easily find a job in the factories located in their home districts or nearby. They can commute to the factories on their own motorbikes or shuttle bus services either operated by the respective factories as in Trần Đề (Sóc Trăng) or Trà Cú (Trà Vinh) or some private service providers as in Phú Bình or Đồng Hỷ (Thái Nguyên). Jobs within village reach are usually lower paid and less diverse as compared to the opportunities offered in outside provinces. It is understandable that many ethnic minority youth head to other provinces to seek better wage employment. The local job positions seem more suitable to many ethnic minority women, especially if married, as these women could daily commute to work and come home at the end of the day to fulfill their role as housewives. 85. For the Mường in Phù Yên (Sơn La) and Khmer in Sóc Trăng, incomes from remittances from household members working far from home play a vital role in changing the physiognomy of the village: new houses with concrete roofs and new home appliances. Migration has significantly increased over the past decade. Most households have a member who has migrated out of the village, at least on a seasonal basis. Among the Sán Dìu of the Bàn Đạt commune (Thái Nguyên), remitted incomes come from gold mining business. In the Mekong Delta, factory work has become so popular that state jobs are no longer considered as prestigious as they once were. Migration, though increasingly popular, is highly gendered: migration (from out of some studied sites) was almost exclusively by men, especially among the Mông in Sơn La, Thái Nguyên, and Thanh Hóa, and to a lesser extent for the Sán Dìu (Thái Nguyên) and the Khmer in the Mekong Delta. The realization that agricultural production does not bring high economic returns has led many Khmer households — those having average cultivation area less than 5-6 công19 — to sell their land and look for work in factories in the province or elsewhere like Ho Chi Minh City or Bình Dương province, but mostly in lower-end manufacturing (i.e., garment, textile and footwear production). 19 One công đất is equivalent to 1,000 square meters in local measurement practices. Research Report 58 Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam 86. Wage employment opportunities are increasingly available for the bottom performers, but mainly in lower-end jobs. Physical connectivity is less likely an issue when it comes to labor market opportunities. From team observations in all the field sites, even in the remotest communes or villages, there are job-opening advertisement boards and brochures widely available in public spaces (e.g., the Mông in Bản Tềnh village of Văn Lang commune, Đồng Hỷ district of Thái Nguyên or the Khơ Mú in Đòan Kết village, Ten Tan commune of Mường Lát district of Thanh Hóa). Labor-intensive companies, with an attempt to seek for low-cost manual workers, go through the district units of labor and social affairs or even the Commune Peoples Committees to approach and recruit potential ethnic employees. Compared to wage employment opportunities, which are usually made available for the top-performing groups, the bottom-performing groups often get their wages in the lower-end activities. However, these jobs are generally for unskilled labor and thus could be considered as low- end wage employment options. For these cases, language barriers and low educational attainment (only primary school attendance among bottom-performing groups) are the factors that prohibit ethnic minorities from securing the decent job opportunities. As indicated in Table 8, the percentage of working people ages 15 or older with technical skills training is lower among bottom performers than top performers. Language barriers are also less favorable to bottom performers and women. For many interviewed, the option of outmigration for factory work is less favorable for married women unless the couple could migrate together. TABLE 8: Ethnic Minority Women with Technical Training and Language Skills 53 EMS Sán Dìu Mường Khmer Xơ Đăng Khơ Mú Mông % working people age ≥15 with 6.2 6.6 7.5 3.5 1.9 1.6 2.1 technical training % working female age ≥15 with 5.9 3.6 7 3.2 1.3 1 0.9 technical training % who read and write the national 79.2 94.8 95 73 72.4 61.6 46.6 language % female who read and write the 72.8 92.1 93.6 67.3 64.5 48.1 30.9 national language Source: Authors’ calculation from 53EMS 87. The informal labor market is an increasingly popular choice among the bottom- performing groups. While accessing to formal labor market or labor export appears to be out of reach to most of the middle aged or those enduring language barriers or low education attainment, cross-border migration (regular or irregular) has been seen as an alternative and increasingly important source of household income for some ethnic minority groups such as Mông, Ca Dong, or Khơ Mú. The Mông in the Northern provinces (i.e., Sơn La and Thái Nguyên) secure income- generating opportunities in neighboring provinces of China, thanks to their linguistic proximity and the manual nature of available farm employment. Likewise, some Khmer engaged as cross-border commuters in Cambodia, as well the Ca Dong cross-border workers in Laos. Apart from the studied provinces, there seems to be an upsurge of illicit cross-border migration in communes and districts of, for instance, Lào Cai (Mường Khương, Si Ma Cai, Bắc Hà) and Bắc Kạn (Pắc Nậm) provinces. The What are the Main Drivers of Socio-Economic Performance Among Ethnic Minority Groups? 59 illicit cross-border movement carries lots of risks.20 Coupled with no literacy requirement or technical skills, the lure of higher wages led a growing number of ethnic minorities to cross illicitly into China. Of course, the lack of education reduces access to better employment or productive opportunities among the poor, resulting in their low earnings and taking risks upon themselves (see also World Bank 2018a). 88. Wage employment opportunities, although a pathway out of poverty for ethnic minorities, also present some challenges. With respect to formal wage employment, criteria limiting working age between 18 and 35 and requiring a secondary education often rule out the participation of (middle aged) women and men in factory jobs. In addition, remuneration and benefit systems in place are not transparent and seem unjust toward ethnic minority employees in some factory work. There is uncertainty about long-term employment since enterprises in these areas only provide one-year (or even six-month) base contracts. The fact that many local youth are not prepared to accept a strict working schedule at Kinh-owned-and-operated enterprises may have something to do perception of being treated unfairly (difficulties in getting leave, too-short leave; hazardous working environment such as chemical contamination and other risks in Samsung factory in Thái Nguyên). In addition, the pay is not really attractive (around VND 3 to 4 million) for jobs offered by the factories in nearby locations. Higher-paid jobs are available in other provinces, but migrating to other provinces also incurs additional costs. In discussions with the study team, some youth revealed that working in the cities or industrial parks might earn them good wages but they are faced with high accommodation costs, and temptations like gambling and drugs. Working in their (native) province allows the youth to come home in the evening, saving most costs. In areas visited by the study team, both workers and state officials raised concerns about the lack of benefits for eligible wage workers, including social insurance, workplace protection and safety rules, and other non-wage benefits. In addition, there is no evidence that the labor market is open for ethnic minority people with disabilities. The 53EMS indicates that there is around 1.01 percent of the ethnic minority population with disability. Access to labor market as well as other opportunities for ethnic minority people with disabilities is reported to be very limited. 89. Engaging in the wage labor market also raises family-related issues. Migrants such as the Sán Dìu in Thái Nguyên who work in factories away from home often rent out their land for others to cultivate; this is particularly the case for people too old for factory work and older than 40. Likewise, migrant Khmers working in Ho Chi Minh City often rent out their fields in their absence to the landless. In addition, there is an emerging issue of children left behind by their migrant parents (UNICEF 2016). Grandparents or other relatives often look after children of pre-school age while their parents are at work (in a far-away province) and generally communicate with the young in their native tongues. These children only begin to learn to speak the Kinh language when they enter kindergarten at age five, giving them just about one year to get familiar with the national language before starting the first grade. Interviewed teachers point out this often causes some children to lag behind in class, especially among groups with high rates of non-proficiency in the Kinh language like the Mông, Khơ Mú, and Xơ Đăng (see Table 8 for detailed figures). In contrast, the Sán Dìu children under 13 in Thái 20 There was anecdotal evidence that some Mông and Sán Dìu who illicitly crossed the border to seek jobs were ar- rested by the Chinese police force and subjected to mistreatment in detention. Research Report 60 Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam Nguyên are generally fluent in the national language due to their daily contacts with the Kinh and can hardly speak the language of their familial ethnic group. In addition, lack of concern for children’ schooling during their parents’ long absences often result in poor schoolwork and increased dropout as observed among the Mường and the Khơ Mú in Sơn La, and the Khmer in the Mekong Delta (UNICEF 2016). Information from focus group discussions with Sán Dìu youth indicates that, in view of the job opportunities in the industrial sector, a number of youths with their parents’ encouragement have tried to finish upper secondary school to qualify for these jobs. 90. There exists an ethnic differentiation in job structure and share of labor export market. As a result of the policies to promote labor export in the poor districts covered under the National Target Program on Sustainable Poverty Reduction (NTP SPR, 2016-2020 and its earlier phases), there has been a modest number of ethnic minority youth to pursue labor export. The 53EMS shows that around 0.7 percent of the ethnic minority labor force is now working abroad. The labor markets such as Japan and South Korea, which demand workers with certain technical skills and language fluency, attract mainly Kinh applicants and some ethnic minority groups considered to be socially and economically ‘advanced’ like the Thái and the Mường (as observed in Sơn La). These returnee workers are generally reintegrated easily with good jobs in the local labor market thanks to their working experiences abroad. Whereas, the Middle East countries, hiring mainly for domestic work, attract female workers, including women from ethnic groups in the remote areas like the Ca Dong (Quảng Nam), the Mông, and the Khơ Mú (Thanh Hóa). These migrants have little education and their weak social position make them highly susceptible to exploitation and abuse. Their employment prospect hardly improves after their return. Some young men in ethnic minority groups have found themselves doing manual work, mainly civil engineering work, in the Middle East or some other countries in the region (e.g., Malaysia, Indonesia or Taiwan). Discussions with some returnees revealed that this was best described as a sort of ‘lottery’ game due to lack of information on the destination countries. Some returnees found out the hiring companies were bankrupt, some got a lower salary than promised because their employer was not getting the necessary work contracts. The young returning workers therefore ended up with indebtedness from having to borrow money from relatives or VBSP to pay for the air tickets and the pre-work training fees. One may ask the question of should the respective policy on labor export among ethnic minorities be continued or not as part of the poverty sustainable reduction process? 91. Labor market mobility in summary. Labor market participation has become the major source of income for the top-performing ethnic groups. While more than half of total income of the top performers is earned from the non-farm labor market, bottom performers earn less than one-fifth. Out of all labor market opportunities being sought, wage employment is dominant. Factories that offer higher wages usually employ the top performers while bottom performers find their own opportunities in lower-end or even illegal cross-border jobs. Economic connectivity reflected in having factories in local areas provide ethnic minority women opportunities to commute to work, which would be otherwise not possible. Becoming a major pathway out of poverty, there are also problems encountered or caused by ethnic minority youths engaging in the labor market. Notably, the workplace conditions are usually poor and unequal treatment does exist toward ethnic minority workers, especially from the bottom- performing groups. Opportunities for women to migrate to other provinces for wage employment is more limited compared to their male counterparts; this is most pronounced for women of the bottom- What are the Main Drivers of Socio-Economic Performance Among Ethnic Minority Groups? 61 performing groups. As youth and parents commute or migrate to find wage employment opportunities, the communities are left with the middle aged or elderly to both take care of pre-school and school-age children and engage in agriculture activities. 4.5 Some Aspects of Access to Education 92. Low-educational attainment of the bottom-performing groups compared to top performers is partly reflected in HDI rankings (discussed in Section 3.3). Using 53EMS, Table 9 elaborates further on educational attainment levels of the selected ethnic groups. It shows that there are no materialized differences between the selected groups in pre-school and primary school. This might be attributed to the universalization of free primary school in the ethnic minority regions. Table 4 shows that primary schools are within a radius of 2.5 kilometers from ethnic minority households. The prevalence of attaining no education remains high among the bottom performers. Notably, 37.2 percent of the Mông have received no formal education. For secondary schools or higher levels, the top performers exhibit considerably higher levels of educational attainments. These numbers also suggest that many pupils from the bottom-performing groups drop out after primary or lower secondary education. TABLE 9: Educational Attainment Levels of the Selected Ethnic Minorities (%) 53EMS Sán Dìu Mường Khmer Xơ Đăng Khơ Mú Mông average No education 15.6 3.6 3.3 20.7 18.3 26.5 37.2 Pre-school 25.6 22.4 20.7 33.4 30.3 25.2 26.1 Primary 27.8 31.9 29.4 30.1 30.6 29.9 22.0 Lower secondary 19.2 27.6 30.3 10.2 16.5 14.5 10.7 Upper secondary 7.1 9.5 10.3 3.1 3.0 2.7 2.4 Vocational 2.5 3.2 3.5 1.0 0.8 0.8 0.9 University & College 2.1 1.8 2.4 1.5 0.5 0.3 0.5 Source: Authors’ calculation from 53EMS 93. School dropout prevalence is reportedly due to geographical distance between home and school and lack of social contacts leading to language barriers, including limited contact with outsiders in the national language. Dropout in the years following lower secondary school or high school is reported in all studied provinces. Early dropout in secondary school is found to be more prevailing among the bottom-performing groups such as the Khơ Mú in Sơn La and the Mông in remote Bản Tềnh village of Thái Nguyên. The high school dropout rate is also common among the top performers like the Sán Dìu, Mường, and other bottom performers such as the Mông in lowland Khe Cạn hamlet of Thái Nguyên, the Xơ Đăng in Kon Tum, the Ca Dong in Quảng Nam, and the Khmer in Mekong Delta. Lower rates of educational enrollment are partially explained by household poverty status itself; they correlate to the lack of connectivity among the bottom performers. 94. There is a widespread perception (often held by the majority Kinh and the other top- performing ethnic groups such as the Thái, Mường, Tay) that the least performers like the Research Report 62 Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam Mông, Khơ Mú, Xơ Đăng are generally taciturn and timid. Thus this ‘so-called’ ethnic characteristic impacts negatively on the interactions between those ethnic minority pupils and teachers (who are mostly the Kinh). When the teachers do not make extra effort to improve the situation, the result is school aversion and absenteeism among those ethnic minority pupils. This may be due to the fact that those least-performing ethnic minorities and the Kinh approach childrearing differently. The Mông, Khơ Mú, Xơ Đăng, and Ca Dong rarely beat or scold their children whereas the Kinh follow the adage “spare the rod, spoil your child”. Moreover, in the Mông home “father and mother don’t talk much” (quoted from the interviewee). More importantly, the lack of preschool facilities in remote areas means that children under age 5 often stay at home, delaying their entry in primary education and having the disadvantage of unfamiliarity with the Kinh language (as mentioned in Section 4.4). In this regard, the disadvantage may compound over generations, with children of linguistically excluded groups facing an additional barrier to access learning and jobs, public services, and democratic spaces (World Bank 2013). Both Kinh teachers and local officials generally blame the poor performance of the Khơ Mú, Mông, Xơ Đăng, and Ca Dong pupils on their inherent poverty and the lack of support from parents, which they consider the failure of ethnic minority parents’ childrearing. 95. Conversely, a lack of cultural sensitivity coupled with popular discourse about backwardness, self-deficiency, and superstition with regard to the least-performing groups like the Khơ Mú, Xơ Đăng, Ca Dong, and especially the Mông contributes to misunderstandings and an inability to communicate on educational issues. The language barrier between Kinh teachers and pupils has already been mentioned as an issue of perceived prejudice. The ability to read and write in Vietnamese is a factor that cuts directly to the perceptions or prejudices of the majority Kinh toward those ethnic minority groups.21 The ability to read and write among women of minority groups was particularly low in the Mông (30.8 percent) and Khơ Mú (48.1 percent) (refer to Table 8 above). This factor may sharpen existing prejudices toward groups that follow religions or beliefs that might be unfamiliar and “conflicting” to local authorities and public servants or having potentially political risks like the Protestant faith or the Dương Văn Mình belief, or to a lesser extent the Catholics in Kon Tum. 96. In-depth interviews with local residents (from both the best- and least-performing ethnic groups) in all study sites generally revealed that those students who do not relate to privileged relationships with power-holders have little chance to benefit from job opportunities in the state sector. In particular cases, for instance, some Mường and Sán Dìu students, who managed to attend universities or colleges, could not find jobs in the state sector, that being their first choice in remote areas. In such cases, the problem had to do with the lack of ‘political connections’ rather than just ‘physical connectivity’. The staffing in local authorities is centrally regulated by the Ministry of Home Affairs (MoHA), and positions are usually open only after existing officials are promoted or retired. Observations in the field indicated that a few ethnic minority graduates have found their skilled jobs in the private sector. The secondary choice is to look for wage 21 Stereotypes that are often ingrained in service providers can have a damaging effect on the way they treat their clients, but these issues are not insurmountable. A growing practice focuses on ‘cultural competencies’ among service providers. Holding providers accountable not merely for their presence and their technical skill but for cultural competence as a performance indicator is likely to go a long way toward ensuring that cultural minorities feel comfortable in accessing services (World Bank 2013). What are the Main Drivers of Socio-Economic Performance Among Ethnic Minority Groups? 63 employment in the labor market. Many ethnic minority graduates ended up with manual work in factories not unlike those who completed secondary schools. Most factory employers do not want to recruit university graduates for manual work, and so having a university/college decree could be a disadvantage that could even lead to being ‘disqualified’. This dilemma has caused a sense of disillusion among all studied ethnic minority groups about the benefits of pursuing higher education but perhaps contributes more to the lower academic achievement at high school level, which in turn determines less access to better paying jobs for the least-performing groups. As a result, the lower tertiary education attainment of the least-performing groups excludes them from the most rewarding jobs (World Bank 2018b). Whereas those from the best-performing groups are more likely to have a secondary school education or higher (shown in Table 9). 97. More importantly, those who have only completed primary school are astoundingly 3.1 times more likely to be in a high-risk occupation in Vietnam (ILO 2016). From gender perspectives, women are more likely than men to be employed in high-risk, automation-related occupation. The resultant unemployment will have a more disruptive impact on household welfare given that most ethnic minority households now rely on wages for the bulk of their income. Figure 5 suggests a correlation between the educational attainments of the household heads and the importance of wage employment using the aggregated data from the 53EMS. It suggests that the higher educational attainment that an ethnic minority household head obtained, then the higher proportion of wage income in the total household income. Notably, when the head attains tertiary education or advanced vocational training, more than two-thirds of his/her household income is generated from wage employment. It is also noted that regardless of the educational levels of household heads, wage employment was found as a major source of household income (which is consistent with figures reported in Figure 4). FIGURE 5. Income Structures and Educational Attainment of the Household Head Postgraduate 83.1 10.4 University 71.9 8.2 College 73.3 11.8 Vocational - advanced 79.0 8.4 Vocational - intermediate 59.8 20.3 Vocational - primary 65.4 16.7 Vocational - short-term 53.9 16.9 Upper secondary 40.5 32.9 Lower secondary 38.1 42.7 Primary 35.6 46.3 Below primary 36.8 45.4 0 10 20 30 40 50 60 70 80 90 100 Wage Agriculture Nonfarm Other Source: Authors’ calculation from 53EMS 98. Provision of vocational training is one potential area that could contribute to skill development for ethnic minority youths who seek labor market opportunities. In all the Research Report 64 Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam districts visited, there are district centers that provide vocational training under Project 1956 or the NTP NRD 2016-2020. This vocational training for the poor and ethnic minorities, according to these policies, are offered free of charge. In principle, the contents of the vocational training should be demand driven, based on proposals from the communes. In fact, however, commune authorities proposed the content, based on local livelihood potential, to the district authority. The district authority, usually comprising District Division for War Invalids, Labour, and Social Affairs (DDoLISA), would then compile the proposals from communes and submit to the District People Committee’s Chairman or Vice Chairman for approval. Therefore, the free vocational training was in fact not really demand driven. In addition, as indicated in Table 8, the training has reached very modest numbers of ethnic minorities. Even among the top performers, the percentages of working people age 15 or older having any technical training was around 7 percent while bottom performers were ignored for the most part. 99. Looking further to the content, most of the vocational training offered by the district authorities was focused on agriculture or livestock and organized in the form of short-term courses. This short-term training, usually less than three months, was delivered in the evenings or weekends at the CPC meeting room or community learning centers and conducted in the national language, using some traditional classroom (or blackboard) approaches. Representatives of the district vocational training (and continuing education) centers in Sơn La, Thanh Hóa, and Trà Vinh reported some difficulties in this free provision of vocational training. Notably, it was not easy to register sufficient numbers of young participants; the young aspirants were not interested in vocational training on agriculture or livestock, but in fact were looking for labor market opportunities outside agriculture. However, for these non-farm skills (e.g., manual work in aquaculture processing, garment, footwear, or electronics assembling) the factory employers usually provide basic, on-the-job training for their recruited labor. 100. Access to education in summary. The bottom performers have experienced lower educational attainment compared to top performers. Notably, as high as one-third of some bottom-performing Mông or Khơ Mú have no educational attainment. In addition, many ethnic minority children of the bottom- performing groups dropped out in primary or lower secondary schools. Distance to secondary schools, language barriers, and lack of cultural sensitivity with regard to the bottom-performing groups held among the Kinh teachers were found to be contributing dropout factors. Limited opportunities for tertiary graduates have led to disillusion about the benefits of pursuing higher education in the ethnic minority areas. Some young ethnic minority graduates, unable to find employment in the state sector, ended up hiding their decrees from employers in order to be recruited as workers in labor-intensive factories. Vocational training, expected to enhance employability of ethnic minority youths primarily in agriculture, failed to offer the skills needed for the nonfarm labor market. 4.6 Some Aspects of Healthcare Services 101. The Government of Vietnam has invested in the grassroots health care services and the 53EMS data show that commune-level health centers are found to provide most basic medical services. Of these commune health centers, 68 percent are in solid structure buildings, What are the Main Drivers of Socio-Economic Performance Among Ethnic Minority Groups? 65 serviced by 26,557 commune health staff (of which 12 percent are doctors and 49 percent are ethnic minorities). Figures show that healthcare investment also goes to the village level. Of the 48,364 villages in the areas with ethnic minorities, 85 percent was reported to have village health workers. Table 10 shows that around 44.8 percent of all the ethnic minorities were provided with free health insurance cards (meaning around 6.67 million health insurance cards were provided for free to the ethnic minorities). 102. In recent years, there has been an increasing willingness among ethnic groups to visit modern health services provided by health centers or hospitals. In the past, when sick, the local residents usually consulted spiritual shamans or traditional healer. But notably, the provision of free health insurance for the poor and ethnic minorities contributed significantly to the willingness to seek help from healthcare services. It is likely that the top-performing ethnic groups use far more health care service than the bottom-performing groups.22 With the improvement of healthcare, for example, the Mường, the Sán Dìu, and the Khmer no longer had much need for shamans. TABLE 10: Access to Healthcare Services 53 EMS Sán Dìu Mường Khmer Xơ Đăng Khơ Mú Mông Having free health insurance card (%) 44.8 38.2 32.0 47.8 52.9 53.2 40.5 Infant mortality (%) 18.01 14.98 17.26 15.94 22.86 26.89 27.77 Fertility rate (# of children per 1,000 2.09 2.55 2.09 2.07 3.56 2.80 3.57 women) Married women using contraceptive 76.89 76.86 83.12 75.00 65.54 77.17 68.02 measures (%) Pregnant women age 15-49 having 70.9 78.9 82.4 84.1 66.3 47.7 36.5 check-up (%) Women age 15-49 having homebirth 36.3 15.4 21.1 8.2 68 73.3 77.4 delivery (%) Women age 12-49 delivered 3 or more 0.2 0.19 0.13 0.16 0.3 0.31 0.38 children (%) Source: authors’ calculation from the 53EMS 103. Maternal health service utilization by ethnic minority women remains especially low. Among the bottom-performing groups of the Mông and Khơ Mú, the incidence of women who undertook pregnancy check-ups is considerably lower (36.5 and 47.7 percent, respectively) than those in the top-performing groups (78.9 percent among the Sán Dìu, and 82.44 percent among the Mường as shown in Table 10). Notably, the rates of homebirths were found high among the bottom performers, with more than two-thirds of pregnant Khơ Mú and Mông women delivering at home while the corresponding figures of the Sán Dìu and Mường women were 15.4 percent and 21.1 percent, respectively. High frequency of homebirths within the Mông, Khơ Mú, and Xơ Đăng communities was confirmed in the study sites and was attributed to the speed of delivery. A primary reason for 22 This might not reflect in the prevalence of having health insurance cards. In fact, the bottom performers are usu- ally residing in the communes classified as ‘extremely difficult communes’ covered by the Program 135, where free insurance cards were provided to all citizens. Research Report 66 Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam homebirths, the average distance from the Khơ Mú and Mông households to hospital was around 30 kilometers (refer to Table 4). Findings from a recent study (MoH and UNFPA 2017) reflect ethnic minority women’s views that pregnancy and childbirth are healthy and normal processes and it is therefore not necessary to visit health professionals unless complications arise during pregnancy or labour. A prominent view amongst women is that maternal healthcare services are under-resourced and the benefits do not outweigh the potential risks and costs of attending health facilities. Quantitative data from the 53EMS indicate that commune health centers are poorly equipped to provide, with only 20.5 percent of these health facilities were reported to have met national standards. Additionally, the lack of cultural sensitivity and language barriers among healthcare workers contribute to misunderstandings and a poor ability to communicate with regard to health issues (MoH and World Bank 2018). This might relate to a lack of effective policy effort to incorporate ethnic minority populations in modernisation programs through centralized “one-size-fits-all” approaches ( White et al.2012). 104. The Mông have experienced high incidence of child marriages. The empirical findings of this study are consistent with 53EMS that the Mông has the highest percentage of child marriages (59.7 percent). It has been reported that the prevalence of underage marriages among the Mông even happened to adolescent daughters of some local officials in the studied villages; this was confirmed from the field visits. Apart from the Mông, as indicated in Table 11, no other studied least- performing group was listed among the ethnic minorities with very high rate of underage marriage. It is worth noting that all best-performing groups were found pertaining to relatively high prevalence of child marriage (i.e., between 13.6 percent to 17 percent). There is a dire need to investigate the socio-cultural background of the underage marriage phenomenon and the risk factors involved in specific geographic areas and within specific ethnic groups. Within the scope of this study, there are indications that the underage marriage situation among the Mông and other ethnic groups may stem from the fact that teenage sex may be taboo in these societies, which make teenagers turn to early marriage as an alternative. TABLE 11: Rates of Child Marriage and Consanguineous Marriage 53 EMS Sán Dìu Mường Khmer Xơ Đăng Khơ Mú Mông average Rate of consanguineous marriage (%) 6.5 1.1 1.7 15.9 4 25 11.4 Rate of child marriages (%) 26.2 16.9 13.6 16.5 26.6 44.3 59.7 Source: authors’ calculation from the 53EMS 105. Early marriage, early childbirth, and multiple children are sure signs of inherent poverty. In such cases, these immature mothers are more likely to have left school early and are thus less likely to know about reproductive health and contraceptive measures. The bottom- performing Mông and Xơ Đăng have a high fertility rate (3.57 and 3.56 births, respectively, per 1,000 women). The top-performing Mường and Khmer have very low fertility rates (Table 10). Underage marriage and bride kidnapping prevail among the Mông in Bắc Yên, Sơn La but not among the Mông in Thái Nguyên, which is due to the impact of religious faith (the Protestant faith or Dương Văn Mình belief) on attitudes against violence toward women and girls. With regard to underage marriage with mutual consent of both boys and girls, parents tend to accept these arrangements What are the Main Drivers of Socio-Economic Performance Among Ethnic Minority Groups? 67 grudgingly, most likely due to parental behavior as pointed out earlier. Mông parents tend not to object to these because they believe in the custom that once a Mông girl crosses the main door of the house of the abductor or ‘husband-to-be’, she is considered to be part of that family. Once married, many girls are requested by her in-laws to drop out from school, soon give birth, and take care of the children; whereas the husbands could be allowed to continue to complete their study at secondary or up to high school level. These young women’s lower educational attainment may deepen further their dependency on their husband economically, and may expose the women to intimate partner violence. A recent multisectoral nutrition assessment and gap analysis (World Bank 2018c) reveals that early marriage and pregnancy is associated with stunting and with intimate partner violence (IPV). Since adolescent pregnancies and the condition of adolescents and young women before and during pregnancy is of such importance to the prevention of stunting, this is a topic of high importance, the same report states. 106. Quantitative results demonstrate high rates of consanguineous marriage among the Mông and Khơ Mú. Consultations with provincial officials often indicated that consanguineous marriage was most rampant with the Khơ Mú, among the bottom-performing ethnic groups (Table 11). This discussion suggests that consanguineous marriages are controlled by a number of policies and propaganda efforts. Nevertheless, qualitative findings reveal that the practice of consanguineous marriage is uncommon among local residents in the studied sites. This is consistent with the information from direct consultation with the Mông and Xơ Đăng. In fact, curbing consanguineous marriage was widely understood as a policy target by the local authorities and villagers themselves (as a result of the recent propaganda). Under that pressure, the informants might understate the actual incidence of consanguineous marriage in the communities. 107. Aspects of healthcare services in summary. There are significant improvements in access to healthcare services in the ethnic minority areas. More than a half of bottom-performing ethnic minorities were provided with free health insurance cards; commune-level health centers are available in all communes while 85 percent of villages have health workers. Notably, there is an increasing willingness to use the healthcare services, especially among the top-performing ethnic minorities. However, utilization of maternal healthcare services remains low among women of the bottom-performing groups with low rates of maternal check-up during pregnancy and very high rate of homebirths. Child marriage is very high among the bottom-performing Mông and Khơ Mú. There is the association between early marriage, stunting and intimate partner violence. 4.7 Traditional Institutions and Local Governance 108. Official and traditional systems of village local governance co-exist and play a role in implementing government and development policies at grassroots level. All studied ethnic groups had traditional leaders such as the village elderly (già làng), head of village (trưởng bản), and head of family clans (trưởng họ). And, more recently, the person with prestige (người có uy tín)23 23 As per Decision No. 18/2011/QD-TTg of 2011 by the Prime Minister, each ethnic minority village is eligible to elect one or two “prestigious persons” annually from a list proposed by each village and which must be approved by both Commune Peoples Committee and the District Peoples Committee. The elected persons will receive ‘creden- tials’ and some money from the District Peoples Committee. Research Report 68 Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam was elected by residents in the village (not a hereditary post), and there were the religious leaders (especially the protestant leaders in many of the Mông villages and monks in the Khmer villages). 109. Along with the village mass organizations, the village head is a key player in conveying government policies to villagers. As observed in all study sites, the village head seems to take an active role in providing people with alternatives to develop their livelihood, especially in terms of livestock, loans, and employment. But many heads had only completed secondary school and felt the strain of a heavy workload (Dutta 2018). In the surveyed villages, it is estimated that less than 10 percent of the population was under 30 years of age and had a college degree and so were dependent on the services of the village head.24 In general, villages tend to be more ethnically and economically homogenous than communes (Dutta 2018). Most of the villages have only one dominant ethnic group, with few from other ethnic background (e.g., the Kinh shop owners, those settled in the village through cross-ethnic marriage). Therefore, it is usually the case that the village head represents either the only or the more dominant ethnic group in the village. Many interviewed women indicated that the family clan chief appears to be the main source of information about development of agriculture production, access to loans, and jobs for them. Some young Mông women (Bắc Yên, Sơn La) said that they feel more comfortable seeking advice from the village head regarding access to bank loan and jobs. It is observed that in the top-performing groups, the roles of village heads are generally more pronounced than in the remote villages of the bottom-performing groups where community leadership is probably more influenced by traditional persons such as village elderly or clan chiefs. 110. There is a relatively high percentage of ethnic minorities in the commune administration system. Ethnic minorities comprise 52 percent of total civil servants at the commune level, and 54.7 percent of the Communist Party’s commune committee (Table 12). However, compared to the dominant proportion of the ethnic minorities in these communes, these figures still under represent ethnic minorities in the local administration system. Evidence from field visits reveals other important aspects that are not captured in 53EMS or other quantitative databases. Most key positions such as the Chairman of the Commune Peoples Committee and the Commune’s Secretary of the Party, for instance, are held by Kinh cadres whose qualifications are perceived far better than the local ethnic minorities. With current government policies aimed at reinforcing grassroots structures by moving young Kinh cadres to remote areas, there is little chance for ethnic minority locals to break into these positions, even for the Sán Dìu in Thái Nguyên or Mường in Kon Tum. By virtue of their rather “advanced” social and economic standing and affinities with the Kinh culture, members of groups like the Sán Dìu in Thái Nguyên and the Mường in Kon Tum would be expected to dominate the local administration. However, looking at profiles of the key positions, the Sán Dìu and Mường were not found in the key positions in these locations. 24 For instance, the paper work required for bank loan applications is often perceived as complex and the village cadres often do not have the necessary skills to properly file a form. What are the Main Drivers of Socio-Economic Performance Among Ethnic Minority Groups? 69 TABLE 12: Representations of the Selected Ethnic Minority Groups in the Local Administration Communist Party Total staff (person) 11,774 % ethnic minority staff 54.7 % female ethnic minority staff 7.9 Commune Peoples Council Total staff (person) 9,681 % ethnic minority staff 60.2 % female ethnic minority staff 7.9 Commune Peoples Committee Total staff (person) 59,389 % ethnic minority staff 52.0 % female ethnic minority staff 12.8 Others Total staff (person) 23,746 % ethnic minority staff 57.5 % female ethnic minority staff 17.1 Total commune staff Total staff (person) 10,4590 % ethnic minority staff 54.3 % female ethnic minority staff 12.8 Source: Authors’ calculation from 53EMS 111. Besides key positions in the administration system, ethnic minorities hold deputy or non-key positions or leadership in mass organizations. The ethnic minority groups are usually the most populous in the areas where they serve in deputy or non-key positions. The ethnic groups with modest proportions in the commune rarely serve in local administration. Regarding non-key positions, ethnic minority participation seems to be strongest from larger and indigenous ethnic groups (Wells-Dang 2012). This is particularly so in communes with a majority population from a single ethnic group as observed among the Mông in Xím Vàng commune (Bắc Yên, Sơn La) and the Mường in Mường Thái commune (Phù Yên, Sơn La). Small ethnic minority kinsmen are rarely represented, even in these non-key positions. This could put these ethnic groups into disadvantaged positions in terms of access to resources available from policies and other external support in the commune or villages (Hager 2006). In fact, ethnic minority cadres are viewed as effective in mobilizing their communities and connecting with residents through understanding of language and customs. However, limited educational background and national language skills (which are required to be recruited in the administration system) are seen as the two main constraints to recruiting officials from ethnic minority groups. 112. Village elderlies also play an important role in village management, in some cases, together with hội đồng già làng (village elderly committee). The elderly committee consists of all men who are 60 years old and above, but seems to be common only among Central Highlanders. In reality, the village elderly committee has no actual power but merely symbolic. The village elderly however is often a senior member of the leading clans, having vast experience in agricultural production, and well versed in local customs. He advises the village head on important matters in day-to-day management and mediated conflicts among clans and households. The reconciliation process may include providing advice from a moral perspective and having consultations with each Research Report 70 Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam household involved and participation of all concerned people. In general, the village head, the village elderly, the clan heads, the military commander, and the local religion leaders (and priests where available or Theravada Buddhist abbot and acha among the Khmer in the Mekong Delta) are the key figures in managing the village life. Among the Protestant Mông in Sơn La, Thái Nguyên and Thanh Hóa, the Protestant priest is the most important person in the village. He encourages people to help each other and supports medical expenses for them. Protestants’ network can be a potential pathway for women helping each other in economic activities. In fact, practicing religion was found highest for the Khmer, Xơ Đăng, and Mông (Table 13). Data from 53EMS reports 8,080 religion leaders in the ethnic minority areas.25 Nevertheless, it should be noted that for the studied villages of the Ca Dong in Quảng Nam, the role of the village’s traditional institutions seems to be blurred or have even disappeared from community life. The reason is partly due to the fact that villages lost control over their land and forests, which is an important factor contributing to the sense of community, due to public ownership over land and forest (Nguyen Cong Thao et al. 2012). TABLE 13: Religion Practices among Top and Bottom Performers 53EMS Sán Dìu Mường Khmer Xơ Đăng Khơ Mú Mông % Practicing religion 19.5 0.2 2.6 78.3 45.2 0 19.7 Source: Authors’ calculation from 53EMS 113. For most of the studied ethnic groups, clan membership (dòng họ) is the source of identity, serving as a management function and a binding force among local residents. Kin members are obliged to help one another in such tasks as settling disputes with outsiders, solving debt problems (to a lesser extent for the Khmer in this regard) and, above all, upholding the clan’s reputation. Generally, having relatives working as local cadres or participating in local mass organizations would give the household an edge in getting state credit through proper application procedures and eventual production diversification possibilities. Household members and other relatives, especially those who work for (local) government, give households an advantage when it comes to help in applying and getting capital, hearing about tips on job openings, and being informed of new development schemes. Some households’ good economic position did not come from inherited wealth or land holdings but from good political connections. Nevertheless, kinship and relations do not seem to play an important role in the process of assessing household poverty and gaining access to resources at grassroots level in the Mekong Delta as observed in other settings. Since migration is more common among the Khmer, especially in Trà Vinh, these villagers are less tightly knit communities and their values might be more individualistic. 114. For the Mông, the Mường (in Kon Tum) and some other ethnic minority groups, there are unwritten customary provisions have been handed down from generation to generation, specifying the proper conduct of social behavior and family relations. These customary laws apply to all members of the village at large, regardless of geographical location or clan background. They form the basis for village management, wedding and funeral rituals and other lifecycle events. The links between these customary provisions to economic wellbeing is not 25 Of all 53 ethnic minorities, around one-fifth practice religion (53EMS). Of these figures, 44.4 percent are Buddhist, 19.1 percent are Christian, and 31 percent are Protestant. What are the Main Drivers of Socio-Economic Performance Among Ethnic Minority Groups? 71 straightforward. In some cases, it was found that customary provisions are important to operate community-based mechanisms such as community-based forest management or tourisms, which could then translate into income sources for the households. But it is observed that, with exception of the Mường migrated community in the Central Highlands, the top-performing groups are less subject to customary provisions compared to the bottom-performing ones. This tendency might be partly attributed to the fact that the role of traditional institutions, specifically the eroding power of the chieftainship, seems to likely decrease among these top performers as a result of their perceivably successful integration of these groups into the mainstream society of the lowland Kinh. Further empirical studies are required to better judge this matter. 115. Apart from their spiritual functions, cultural activities and traditional festivities serve to strengthen ethnic and community cohesion. These are also convenient occasions for locals to share information and experiences in agricultural production — Lễ cơm mới (festival of worshipping the new harvest) among the Ca Dong in Quảng Nam, Ok Om Bok (festival of worshipping the moon), Chol Chnam Thmay (Traditional New Year Festival), and Pchum Banh (Thanksgiving to the family ancestors) among the Khmer in the Mekong Delta. Supporting the preservation and practice of these cultural activities is part of the policies for the ethnic minorities. The 53EMS data indicates that 62.5 percent of the villages having community houses and 44 percent have ‘cultural’ houses (e.g., usually in the forms of traditional houses of the ethnic groups where traditional festivals usually take place). It seems that the bottom performers such as the Mông and Khơ Mú have done relatively well in preserving their traditional cultural activities in the studied sites. The situation for the other top- performing groups is however mixed. While traditional festivities of the Mường in Phù Yên district, Sơn La, were viewed as any normal holidays following the national calendar, the migrant Mường in Kon Tum tends to pay more attention to preserving its own ethnic identity. The Sán Dìu in Thái Nguyên have been adaptable to the Kinh society, but there have been recent efforts to preserve and promote the Soọng Cô traditional folk performance. How do these cultural activities and traditional festivities contribute to socio-economic performances of different ethnic groups? This question goes beyond the scope of this study but warrants further study in the future. 116. Women’s Union and Farmers’ Union are the two mass organizations with branches operating at the grassroots (village) level. In order to develop effective empowerment programs for women in collaboration with Women’s Union and Farmer’ Union, it is necessary to assess the extent and nature of local people’s participation in the activities of these organizations. The organizational structure and modus operandi of these organizations are highly gendered. Women’s Union activities focus on mobilization and propaganda related to women’s issues. Farmers’ Union concerns itself with cultivation issues and can be seen as gender neutral. In reality, its activity is highly gendered since it often deals with individual households and, specifically, household heads who are mostly men, being fathers, husbands or (eldest) sons. Since men generally have better schooling than women and are more attuned to technical matters, they are more likely than women to participate in Farmers’ Union activities and get informed about new trends in cultivation and animal husbandry. Women often have a lower educational level – some have never gone to school – and usually shy away from community activities, leaving decisions on agricultural production to male family members. In addition to their main mandates related to propaganda, managing the VBSP preferential loans is the main channel for these organizations to contribute to economic wellbeing of ethnic minorities. In development Research Report 72 Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam partner-supported projects, Women’s Union (and Farmers’ Union in some cases) have been involved as an active implementing agency of subcomponents that are designated for women as the primary beneficiaries. In most cases, this has proven to be an effective implementation arrangement for women’s economic empowerment. Outside the areas targeted by development partners, contribution of these mass organizations to socio-economic development of different ethnic groups is a case-by- case basis, and largely depends on the capacity of the key personnel at the local Women’s Union or Farmers’ Union. 117. Traditional institutions and local governance in summary. Village heads, village elderly, clan chiefs, and religious leaders are key community leaders who exert influences on local institutions and governance. Village heads are more influential to the top performers while more traditional community leaders such as clan chiefs or village elderly are more influential in the bottom-performing communities. Figures indicate high representation of ethnic minorities in the local government system (with more than a half of commune officials being ethnic minorities). But this remains under representative if comparing to the dominance of ethnic minorities in the location. In addition, ethnic minority cadres usually pursue non-key positions. For these positions, the representation of the large and indigenous groups is usually strong, and this could put the groups with small populations at disadvantages. Customary laws apply to all members of the village at large, regardless of geographical location or clan background; but the top- performing groups are less subject to customary provisions compared to the bottom-performing ones. With regard to traditional culture, it seems that the bottom performers have done relatively well in preserving their traditional cultural activities while the situation for the other top-performing groups is mixed. 4.8 Gender Roles and Intra-Household Power Dynamics 118. In the seven study provinces, some ethnic minority groups are patrilineal (the Mường, Sán Dìu, Mông, and Khơ Mú), and some are bilineal (the Xơ Đăng and Khmer). Among some highly patriarchal groups like the Mông, Mường and Sán Dìu, women, particularly married women, are expected to play a subordinate role, with men making most decisions in the household and community. Among bilineal groups like the Khmer and Xơ Đăng, gender relations are more or less equal, with both sons and daughters having equal inheritance rights.26 There is a tendency to equate patrilineal or matrilineal systems with the roles and positions of women/wives and men/husbands in family relations. In reality, patrilineal or matrilineal systems do not necessarily entail patriarchal or matriarchal power structure. A woman’s position is rather important in societies considered to be patriarchal as indicated among the Sán Dìu in Thái Nguyên nowadays. On the contrary, in bilineal societies the position of men/husbands can be considerable due to their role in agricultural production and other off-farm activities, which brings ready cash for day-to-day expenses as reported among the Khmer and the Xơ Đăng. Therefore, generalizations and misinterpretations of social systems as simply patrilineal and matrilineal can easily gloss over the complexities of local 26 The cases of the Khmer and Xo Dang echoes findings from India that making women joint landholders in land re- distribution or resettlement projects or reserving land use quotas can increase their access to opportunities while also empowering them (World Bank 2013). As Deininger, Goyal, and Nagarajan (2010) pointed out, coparcenary rights for women in property can have a positive effect on the education of girls within the household. What are the Main Drivers of Socio-Economic Performance Among Ethnic Minority Groups? 73 notions and practices about gendered roles and social economic functions in family relations within these selected minority groups. 119. As observed across the studied groups, gender roles in livelihood activities are more pronounced, with men involved in harvesting, construction, and animal trading, and women leading in handicraft and clothing production, planting, and animal raising. The question of who manages the household’s money is not of primary importance. In some households it is the task of the husband; in others it is the wife who is in charge; there are cases when the husband and the wife keep their money separately. Among the Mường in Sơn La, for example, entrusting the money-keeping task to the husband is indicative of the high regard the wife has for her man — a good husband who does not indulge in alcohol or gambling. In such cases the wife will be free from household financial matters and is not accountable for paying off debt. 120. At one time, men were the primary participants in agricultural extension training and community meetings, but officials and NGOs now report greater participation of women. There are many women participating in hamlet meetings among the Khmer and the Xơ Đăng. Some state officials explain this rather active participation reflects the cultural aspects of bilineal systems among the Khmer and the Xơ Đăng. But this does not necessarily signal an improvement in gender equality. The fact that it is almost exclusively women who attend these meetings reflects a clear gender differentiation in household participation in community activities. But it should be noted that women of the top-performing groups are actively involved, which could correspond as livelihood habits also change such as among the Mường (Sơn La, Thanh Hóa, and Kon Tum) and the Sán Dìu (Thái Nguyên). Among these ethnic groups, Women’s Union representatives mentioned positive impacts of credit and savings programs in fostering participation as well as fostering a model for better-off women in a village getting together to help one or more poor women rise out of poverty. 121. Women still face social and economic inequalities and barriers of access to resources and livelihood opportunities, yet with a varying degree among the best- and the least- performing ethnic communities. The study found among the top performers like the Sán Dìu and Mường and the Khmer that, where women have had some schooling and are fluent in the Kinh language, they tend more often to have be more vocal on agricultural matters. Among the least performers, the Mông, Xơ Đăng, and Khơ Mú women, especially those who did not go to school, rarely or never take part in community activities, leaving production decisions to their menfolk. At another level, while men have the final say in decisions concerning household cultivation and livestock, these decisions are affected by the way they interact with local cadres. Information from in-depth interviews indicates these men are disappointed at the way local cadres respond to the hardships their households have to bear and are generally negative about activities organized by local cadres. It is likely that this sort of ‘avoidance’ attitude is reported more significant among men of the least-performing groups. This is partly coupled with the pre-existing discourses about backwardness, self-deficiency, and superstition that are more likely associated with the least- performing groups, rather than the top-performing groups (or perhaps with a lesser degree). All this does not help the efforts to improve women’s livelihood through credit-granting schemes since men, as household heads, may veto them. Research Report 74 Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam 122. New labor dynamics for women are prompting shifts in customary gender-ascribed divisions of labor in young households. There are existing factory jobs in the vicinity (e.g., Samsung Electronics in Thái Nguyên; Ngọc Hà footwear company in Phù Yên, Sơn La; Pan Pacific garment company in Ngọc Lạc, Thanh Hóa; and Mỹ Phong garment factory in Trà Vinh), which are particularly attractive to local women who want to be near home to do house chores and look after their children after work. There is evidence that in some Mông and Khơ Mú communes in Sơn La and Thanh Hóa, and Sán Dìu in Thái Nguyên where traditions favoring men over women still prevail but are beginning to change such as when local men engage directly in domestic chores and share household workloads with female members. 123. Unpaid care work appears to be a critical impediment to women’s economic opportunities as observed in both best- and least-performing groups. Societal prejudices that have been internalized by women themselves — particularly their traditional housewife role — impact the participation of women in non-agricultural economic sector activities, for example, by seeking work in the cities or abroad. Recent study on gender and off-farm work opportunities (Nguyen 2018, Le 2015) have shown that some men use traditional gendered discourses about feminine and masculine work to reject household chores and react negatively to their wives’ frequent and prolonged absences from the house. Their negative reactions to their wives’ work-lives take various forms, ranging from mockery to outright confrontation, including wife battering (see Bonnin and Turner 2014). As a result, the bulk of agricultural labor and lack of transportation means that women, also burdened with childcare, are not able to establish and maintain social networks on par at the same level as men, especially outside the village. Furthermore, young couples working far from home often leave their children under the care of their parents. This situation creates a kind of unpaid care work often observed among women above age 45, thus depriving them of non- agricultural work opportunities in the labor market outside their localities. Therefore, programs providing subsidized child care not only can help women improve their labor market outcomes, they also can have other positive externalities as well (World Bank 2013). 124. According to 53EMS, the rate of women-headed households is particularly low among the Mông (5.84 percent) and the Khơ Mú (8.96 percent), a situation reflecting the inherent patriarchal system among these two least-performing ethnic groups, in which women play subordinate, dependent roles in the home and community. The qualitative research found that the categorization of male-headed households or female-headed households does not seem to reflect the unequal situation of women who live in male-headed households. The sex of the household head seems to be a misleading proxy for gender relations. While male-headed households are composed of households in which both spouses are present, female-headed households are made up mostly of households in which a husband is not present, either by death or divorce, or migration. A woman could head a household if she does not marry (as among the Khmer and Ca Dong). In such cases, it is important to distinguish between de jure female-headed households (headed by divorced or widowed women) and de facto female-headed households (in which the husband is absent but may contribute to household finances). The latter situation seems to be applicable to the Khmer as seasonal migration is typical in the economic life of the Khmer. What are the Main Drivers of Socio-Economic Performance Among Ethnic Minority Groups? 75 125. In male-headed households, particularly when the husband’s parents make all decisions on expenditure, a young wife who does not have any access to household resources is in a vulnerable situation. This situation is quite pronounced among the Mông in Sơn La where the traditional practice of cohabiting with the wife’s in-laws prevails. It reveals the intra- family gender and generational dynamics of control between husband and wife, between parents themselves, and between parents and young couples. Therefore, qualitative research findings shed light on the complexities inherent in livelihood strategies of minority ethnic communities. Also, notably, an aunt among the Mông bears responsibility toward the children of her brother(s) even after her marriage. She has a say in arranging the marriage of her nieces. She is engaged in arrangements concerning the funerals of her own brother(s) in consultation with the clan’s head. 126. Gender roles and intra-household power dynamics in summary. There is a tendency to equate patrilineal or matrilineal systems with the roles and positions of women and men in family relations. It seems that patrilineal or matrilineal systems do not necessary entail patriarchal or matriarchal power structure. In reality, gender roles in livelihood activities are more pronounced, with men involved in harvesting, construction, and animal trading; and women leading in handicraft and clothing production, planting, and animal raising. Most notably, women face a number of social and economic inequalities and barriers of access to resources and livelihood opportunities, with those in the bottom-performing groups at disadvantages. Labor market being an increasing important source of income also creates new labor dynamics for women that prompt shifts in customary gender-ascribed divisions of labor in young households (especially in the top-performing groups) where women find more opportunities for wage-earning employment. Nevertheless, unpaid care work appears to be a critical impediment to women’s economic opportunity, as observed in both the top- and the bottom-performing groups. Finally, it is noted that the sex of the household headship seems to be a misleading proxy for gender relations. It is important in policy targeting to distinguish between households headed by divorced or widowed women and female-headed households in which the husband is absent but may contribute to household finances. 4.9 Perceptions of Ethnicity and the Dynamics of Trust 127. In all the studied provinces, a stigma of inter-ethnic discrimination was reported not only among top performers toward bottom performers, but also the more economically complex among the Thái toward the Mường in Sơn La, the Mường toward the Mông in Thanh Hóa, the Thái toward the Khơ Mú in Sơn La, and the Hoa toward the Khmer in Sóc Trăng and Trà Vinh. Overall there exists the prejudiced attitude of the majority Kinh toward minority populations. With increasing improvements in road networks in upland areas, lowland settlers have moved to the remote mountain frontier and the ethnic minority hamlets far away in Bắc Yên (Sơn La) and Mường Lát (Thanh Hóa), which used to be perceived by the Kinh as highly unattractive places to live. Nevertheless, improved infrastructure has turned these places into “promised lands” yielding economic potential, and thus subsequently helps bring more lowland settlers to these upland areas (Bonnin 2011). Research Report 76 Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam 128. Qualitative results indicate a concern that the economic activities undertaken by Kinh groups in upland areas sometimes act to further entrench ethnic minorities in poverty (World Bank 2009). As observed among the least-performing ethnic groups’ dwellings like the Khơ Mú and the Mông, Kinh settlers tend to build their houses along the side of connecting roads and/or at the entrance of the hamlet. The Kinh tend to turn the front room of their house into a mini grocery store, providing foodstuffs, electronics, and other basic commodities to ethnic minority residents. A commonly held perception by ethnic minority residents is that the Kinh are much more skillful and resourceful than most ethnic minorities (except the Chinese). It was also the reason given in interviews for the failure of some Mông and Khơ Mú shopowners who could not stay competitive in the market as compared with the Kinh lowland settlers, who are the late-comers yet much more well-connected politically and socially. Eventually the lowland Kinh appear to dominate and monopolize the whole village trade and business, advantaged by their extensive social networks and greater mobility. These Kinh settlers, who most often own the local shops, tend to also play the role of emergency money lenders to the ethnic minority villagers who are charged rather high interest rates. This seems to further widen social distancing between the Kinh and these ethnic minority residents in the uplands. The situation seems less pronounced among the top performers such as the Sán Dìu, Mường, and the Khmer given the bigger share of these ethnic groups in local trading activities at hamlet level. 129. Evidence indicates that many upland groups tend to judge themselves by lowland standards and internalize their inferiority (Nguyen 2016, Well-Dang 2012). This sense of inferiority and timidity is perceived as a matter of degree when it involves cross-ethnic group comparisons. For example, this perception is documented more significant/higher among the Mường in Sơn La as compared with their Thái neighbors who are more dominant socially, culturally, and politically in the Northwest region. Nevertheless, this degree of timidity is considered to be less significant among the Mường in Thanh Hóa, and in Kon Tum given they are the largest group in the province as compared to the Thái, especially the Mông and the Khơ Mú in Thanh Hóa, or they are better off economically and socially than the Xơ Đăng in Kon Tum. This sense of inferiority and timidity, which has been reported to be more common among least performers like the Khơ Mú, Mông, or Ca Dong, has an impact on the participation of these groups, especially young adults in non-agricultural economic sector activities, who will in turn seek work in the cities. But for those groups with adequate representation in local leadership, this sense of inferiority and timidity seems to be changing; change is also seen in greater mobility as migrant workers in urban contexts such as the Khmer in the Mekong Delta (cf. Dutta 2018). Observations from our fieldwork indicate that social separation between the Khmer and Kinh is rarely witnessed in Trà Vinh and Sóc Trăng. 130. When it comes to poverty reduction, the groups that have risen to the top tend to attribute ‘risk avoidance mentality’ to bottom performing of the other ethnic groups. In this line of thinking, the least-performing ethnic groups adopt a ‘play safe’ attitude in agricultural production as well as non-farm livelihood activities. For instance, the locals do not want to take out bank loans but will accept ‘investment’ from local agents or traders even though bank interest rates are lower than the pay off to their investors/creditors. The required collateral — their “red book” (property certificate) — causes the aversion to bank loans. Locals do not want to hand over their property nor pay monthly interest to the bank. However, the interest to the traders or What are the Main Drivers of Socio-Economic Performance Among Ethnic Minority Groups? 77 investors is hidden in the crop yields that the locals have to hand over to their trader or investors at harvest’s end. 131. From the ‘insider’27 point of view, most poor households among the least-performing groups are reluctant, more conservative, and do not trust outsiders (i.e., in this case with the lowland Kinh given prior negative experiences).28 The research team came across many narratives in which the locals do not trust outsiders because of past experiences with the Kinh traders because of unfulfilled promises about providing seeds and fertilizers and guarantees of buying the crops. These broken promises affect the locals’ attitudes toward agricultural diversification opportunities and their participation in decision-making in these practices. Participation was also influenced by economic risk. The poor households are concerned about their own food security and facing a high degree of price and non-price risks, especially in the presence of missing or imperfect credit and insurance markets (Markussen 2017). In fact, they have no cash but must wait until the crops are harvested. Actually, few businesses invest in agriculture, the profit margins are small while the risks are great. Moreover, all studied ethnic minority groups often adopt a ‘wait and see’ attitude; if someone shows they can do it, the rest will follow suit.29 This eventually leads to oversupply and depresses prices. Also, this can result in intra-ethnic tensions and shared perceptions regarding the poor’s aversion to change. According to the assessments of several non-poor families and local officials, a major reason for the poverty of poor households is often attributed to their timid nature and the aversion to change, shying away from other livelihood options and/or new production methods. Within ethnic communities, this has sometimes led the better-off ethnic minority households to stigmatize the worse-off ethnic households for not trying hard enough to lift themselves out of poverty. 132. Perceptions of ethnicity and cultural elements formed the identity of each ethnic minority against the backdrop of their promoted integration into mainstream society and the globalization process. Table 14 indicates that some ethnic minority groups such the Mường, Mông, Khmer, and Xơ Đăng have preserved their own ethnic cultural traditions, including languages, songs, and dances, while other ethnic minority groups like Sán Dìu have not achieved the same result. Among the studied ethnic groups, two top performers, Mường and Khmer, and one bottom performer, the Mông, were found to be able to preserve their traditions and identity. Whereas the best- performing group, Sán Dìu, had almost no households with members knowing their ethnic traditional dance. One may ask, is there a trade off between ethnic identity and involvement in mainstream society? The case of Sán Dìu seems to illustrate this tendency, which runs opposite to other top performers, the Mường and Khmer. A possible reason maintaining ethnic identity could be partly attributed to population size; the Mường as the third largest group and the Khmer as fifth largest group of the country’s 53 ethnic minorities outnumber the small Sán Dìu population. 27 In anthropological research, the ‘insider’ point of view is accordingly associated with that of the emic perspective on a culture, meaning to view the world as a member of that culture views it. 28 In some cases, this mistrust was translated into protests or strikes. The 53EMS reports 1,851 cases of protests and strikes in the ethnic minority areas that involve around 23,990 people in 2015. In addition, 20,889 cases of land disputes and 28,056 cases of disputes (of other natures) were officially recorded by the commune authorities. 29 Members of minorities are not passive subjects; they may not be keen to embrace the workings of the social structure of the dominant group, which has already had a great impact on every aspect of their daily lives (Nguyen 2016). Research Report 78 Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam TABLE 14: Preservation of Traditional Cultures 53EMS Sán Dìu Mường Khmer Xơ Đăng Khơ Mú Mông % use ethnic language when talking 87.0 55.6 86.2 90.6 98.4 96.5 99.5 in household % know how to play traditional music 6.4 0.6 6.4 2.7 16.9 7.4 15.4 instruments % know traditional songs of the 16.5 9.8 18.3 29.8 20.2 14.5 29.1 groups % know traditional dances of the 16.5 0.9 12.0 33.4 30.0 24.2 15.9 groups Source: Authors’ calculation from 53EMS 133. Perceptions of ethnicity and dynamics of trust in summary. There is a stigma of inter-ethnic discrimination held by the majority Kinh toward minority populations as well as among the top performers toward bottom performers. Among the ethnic minorities, there is a commonly held perception that the Kinh are much more skillful and resourceful than most of the ethnic minorities. This is a reason that explains the dominance of the Kinh in many key cash crops and trading (e.g., input supply or selling) activities. In addition, many ethnic groups, especially the bottom-performing ones tend to judge themselves by lowland standards and internalize their inferiority. The sense of inferiority and timidity is reported to be more common among the least performers like the Khơ Mú and Mông. This sense of inferiority and timidity seem to be changing, especially among those with adequate representation in local leadership as well greater labor market mobility. When it comes to poverty reduction, top performers tend to attribute the ‘risk avoidance mentality’ to bottom performers, who adopt a ‘play safe’ attitude in agricultural production as well non-farm livelihood activities. Nevertheless, from the insider point of view, most of the poor households among the least performing groups are reluctant, more conservative, and do not trust outsiders (mostly the lowland Kinh). 4.10 Access to External Support Initiatives 134. Ethnic minorities in general have been targeted by GoVN policies and programs and the projects supported by development partners. Notably, Program 135 has been implemented over the past two decades as a major GoVN-led investment on poverty reduction among ethnic minorities. Apart from Program 135, there are more than 187 resolutions, decrees, and decisions related to poverty reduction and 30 poverty reduction programs (with more than 120 components) aimed at ethnic minorities or mountainous areas (EMWG 2014). There have been several projects supported by development partners over the past two decades or so. After Vietnam was declared its middle-income country status in 2010, development aid has been steadily decreasing to modest levels to such times many development partners completed their project execution strategy between 2010 and 2015. For the development partners who are still active, persistent poverty of ethnic minorities is their main concern. These external support initiatives, either by GoVN or development partners, have resulted in important outcomes in many aspects of living standards for the ethnic minorities. Most notably evident are improvements in physical connectivity, access to public services (education and healthcare), housing conditions, and public utilities (electricity, What are the Main Drivers of Socio-Economic Performance Among Ethnic Minority Groups? 79 drinking water) in many ethnic minority areas, including the poorest and remote communes; the study team visited many of these areas. In this regard, the investments from GoVN and development partners have been a great success. 135. The outcomes in ‘soft investments’ such as production support, capacity development, giving voice, and participation of ethnic minorities in community development are less evident. The top-performing groups tend to benefit from these non-infrastructure investments more that the bottom-performing groups. On one hand, this might be related to better endowment of the top performers such as the Mường or Sán Dìu in terms of basic livelihood access and market linkages compared to those of the bottom performers. On the other hand, this might reflect the willingness to change as observed to be strong within the Mường and Sán Dìu. For instance, many Mường households in Mường Thái (Phù Yên, Sơn La) or in Thạnh Lập commune (Ngọc Lạc, Thanh Hóa) and the Sán Dìu households in Bàn Đạt commune (Phú Bình, Thái Nguyên) indicated their efforts to try different crops and livestock, including the new crops and improved production practices. In contrast, the Mông and Khơ Mú groups were more averse to change. Those interviewed households who have managed to fight their way out of poverty contend that other poor households would not be able to emulate their success because these lack determination and hard work, as indicated earlier. 136. One important area of external support is to improve access to credit for the poor and ethnic minorities. For GoVN-led policies, the preferential lending through VBSP is the major channel of access to credit for ethnic minorities. The VPSP indebtedness was found to be high across all ethnic groups. Figures from the Agriculture Census 2016 indicate that 75.2 percent of households in the areas covered by Program 135 have taken loans from some sources; this number is slightly higher than the national average of 73 percent. Of these households, 74.4 percent of ethnic minorities have borrowed from VBSP while the corresponding figure of the Kinh in these communes are 44.4 percent. The average loan size taken by ethnic minorities was reported at VND 21.4 million. The dominance of VBSP as the principal creditor for ethnic minorities in Program 135 communes is linked to the GoVN policy to provide preferential loans for the poor and the near poor. Evidence of ‘reluctant mentality’ to take bank loans as indicated in some of the existing studies was not really evident in the current work, except for some poorest households. In fact, almost all the poor households in this study have borrowed from VBSP preferential credit lines. The merits of low interest rates without collateral requirements were widely perceived. 137. Top performers Mường and Sán Dìu tend to use their loans more effectively in investing in crops and livestock. Most of the Mông and Khơ Mú households were found to use VBSP loans for cattle-raising. In some cases, buying cattle was even stated implicitly as a condition to approve a loan application. But many of the interviewed households indicated problems with raising cattle that were purchased with bank loans, including reported high mortality rate caused by diseases or “accidents” while grazing. A few of the poorest households revealed that they, anticipating difficulty in paying back the loans, hesitated to borrow from VBSP. With regards to external supports from donors, access to credit is usually addressed by a micro-finance component with focus on establishment and functioning of saving and credit groups (SCGs) with an active role of the local Women’s Union. As saving is a condition to make this model work, it was reported that the bottom-performing ethnic households with more limited ability to save are less likely to benefit Research Report 80 Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam from this micro-finance arrangement. Therefore, in the areas with the bottom-performing groups, the SCG sustainability after completion of the external support initiative was found to be less likely as compared to the areas with the top-performing groups. 138. For poverty-targeted programs, households must be on the official poor list to be considered eligible. Statistics on poverty and poverty reduction bear limited resemblance to the realities on the ground and are often heavily manipulated to suit local political expediency. For example, when a commune needs to show it is meeting certain poverty reduction objectives set by the local Socio-Economic Development Planning (SEDP) or NTP SPR, it produces figures purportedly showing a sufficiently high number of households considered to have escaped poverty. In almost all communes visited, the number of households escaping poverty was stated as an important target indicator of the local SEDP, and these numbers were in fact ‘guided’ by the target-percentage of poverty reduction. Once the target-percentage was set, the commune authorities would work out the ‘quota’ of poor households that could be maintained in the planning year. Likewise, when it comes to seeking funds from the same program, the authorities tend to produce figures showing enough households considered as ‘poor’ in order to be eligible for accessing GoVN policies and subsidies for the poor areas. Field contacts indicated that villagers are frustrated by the way they are classified as ‘poor’ one day and ‘no longer poor’ the next cycle. Moreover, the distinction between poor household and near-poor household is quite arbitrary (as the official poverty assessment criteria, while being centrally defined for all areas, are interpreted by the local officials) and heavily influenced by the power dynamics displayed throughout the poverty assessment process at the hamlet or village level. In this regard, the qualitative findings indicate that representation in the local administration system or village authorities becomes a factor that could influence, in one way or the other, the results of poverty assessment. 139. Related to this poverty assessment and targeting, there is a power dynamic at play between different groups within the community. Policies are not simply imposed top down, implemented by local cadres, and obediently obeyed, but often meet with resistance from locals when their interests are threatened. These are manifested increasingly in community activities where officials are pressured to change their minds. Village heads reportedly fear poor households most. If local officials remove a household from the list of poor households, for instance, members of that household would swear at the officials, accusing them of all kinds of things. This type of response reportedly crosses all the studied groups. Among the eligible benefits from the existing policies and programs, the poor are provided free health insurance cards and school tuition fee exemptions and other related subsidies such as meals or accommodation subsidies for (semi)boarding schools. In fact, many better-off households requested the local staff to leave them on the list of the poor or near poor for the benefits. The evaluation of household poverty seems to proceed smoothly; there have been no adverse reactions on this count among the Khmer in the Mekong Delta. There seems to be a sense of fairness shared between the authorities and Khmer residents. This is unlike the case of the Khơ Mú in Thanh Hóa when a number of residents tried to hide their property to gain eligibility. In the absence of reliable poverty assessment data, figures reported on poverty reduction could be a potentially misleading indication of the development results of policies and projects for poverty reduction of ethnic minorities. What are the Main Drivers of Socio-Economic Performance Among Ethnic Minority Groups? 81 140. Decentralization and participation have been two core principles of the strategies and policies for poverty reduction. Among many mechanisms to operationalize these two principles, the most notable ones are participatory planning and commune investment ownership, meaning the decision-making power for the usage of poverty reduction resources are decentralized to the authority at the commune level. However, there has been such a large gap between basic principles in policy designs and actions in practice. With regard to participatory, there are still a lack of a consistent and enforced guidelines for making participatory planning at the local level and having the prioritization processes of the poverty reduction programs as part of that participatory planning. At the provincial level, more than 30 provinces have institutionalized some participatory socio-economic development planning procedures as a result of donor- supported projects such as the World Bank in the northern mountain provinces and IFAD in its participating provinces.30 Observations from the field where external support from donor-funded projects remains, participatory meetings are organized to get the proposals from villagers in the prioritization process of the socio-economic development resources available. However, the quality of participation especially of ethnic minority women was found to be low. In other areas without external support, participatory planning was not usually enforced even in the provinces where the process was already institutionalized. The lack of participation in the socio-economic development planning (and hence the prioritization process for the policies and projects in the areas) impedes the voices of many ethnic minorities in the local decision-making process. In this regard, it was also observed that out of the limited participation found on the ground, the top performers seem to be most active while participation of the bottom performers such as the Khơ Mú was extremely limited. This might be partly attributable to the lack of diversification in production support policies for ethnic minorities. Within the NTP SPR (and its component Program 135), for instance, cattle-raising and some small processing machines garner the most popular types of support. It is sharply in contrast with the diversified livelihood portfolio support made by the donor-supported projects in the same locations. 141. Another mechanism of decentralization is commune investment ownership (CIO) under the CPC authority. With CIO decentralized to the commune level, it was expected that the resources would be used in a locally responsive manner (rather than decided by the district or even the provincial authorities). Official statistics from CEMA indicated 100 percent of Program 135 communes have exercised CIO for the production support sub-components of Program 135 and around 75 percent for infrastructure sub-component development; though the figure varies among provinces with very high prevalence of infrastructure CIO (Hóa Binh with 100 percent) to those with very low incidence of the infrastructure CIO (Dien Bien with 20.3 percent). Observations from the field indicates that CIO has been stronger in the areas with the top-performing groups. This is partly attributed to CPC capacity in the areas with good connectivity. 30 There are no statistics in this regard. The number of 30 provinces was estimated by the authors by working with different provinces and study reports. Out of the provinces visited in this study, Sơn La (with support mainly from the World Bank-supported Northern Mountains Poverty Reduction Project, Phase II), Thái Nguyên (as a result of cooperation with Plan International for nearly two decades), and Trà Vinh (as a result of two rounds of program support from IFAD) have institutionalized a participatory SEDP process; Nam Trà My district of Quảng Nam and Ngọc Hồi district of Kon Tum have experienced participatory planning from various projects, including the World Bank-supported Central Highlands Poverty Reduction Project. Research Report 82 Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam 142. Gender mainstreaming is another principle that is reflected in most of the current GoVN-led policies and programs. Most notably, the NTP SPR 2016-2020, being the largest poverty reduction initiative, was designed with the principle of giving priority to women in poor households and ethnic minority women for participation in and benefit from the program activities. A mid-term assessment conducted by MoLISA, Oxfam/CARE/SNV (2018) indicates that the implementation of this gender principle remains limited due to lack of enforced guidelines and local capacity to implement gender mainstreaming. The challenges for gender equality in the ethnic minority areas found in this study and discussed earlier are similar to the assessment by MoLISA and others. 143. According to interviewed local officials, development schemes sponsored by the state or NGOs seem to create a mentality of dependency among some people. For some people, it is an “easy come, easy go” approach to tasks: if they succeed, that’s fine; if not, that’s fine, too. In cases of failure, they expect further support from the state or NGOs. This might be a result of having many GoVN policies and programs that provide production inputs and other support for free. Such provision of inputs is not a result of a participatory process where the needs of the beneficiaries are reflected in the types of input support actually needed. In other words, the poor ethnic groups are targeted by policies, but decentralization and participations were not sufficiently strong to transform them into agents of change. Such dependency appeared to be more pronounced in the bottom- performing ethnic groups. Development partner-supported projects were found to be more effective in participatory planning with a requirement to make counter-contributions; therefore, ownership by beneficiaries and communities is stronger, and dependency is less an issue. 144. Fragmented and thinly resourced policies have not yet covered the particular cultural and structural needs that marginalized ethnic minorities have and may require different interventions and services.31 Existing policies and legal framework reveal little diversity among target groups.32 The problem lies in the fact that most policies are designed without the participation of ethnic minorities (i.e., regarding the Program 135 in the process of designing and developing the socio-economic plans at commune level). These minorities are seen merely as ‘target groups’ but not as active agents of change with the potential to contribute (Dutta 2018) despite the driving principle of NTP SPR being that production development activities and livelihood diversification should be determined by the needs of the beneficiaries. However, these objectives are rarely followed, except in the case of Trà Vinh. Elsewhere most production and diversification support schemes in other provinces still revolves around cattle-raising, production mechanization, and small-scale manufacturing in communes placed under Program 135. This seems to run counter to the livelihood diversification objectives spelled out in projects funded by IFAD, the World Bank, and other development partners in the same target areas. The main reason is linked to limited 31 Seen from policy perspectives, the limits set by P135 do not reflect the realities on the ground. For instance, the solution to provide the landless farmers (or the poor with little land) with loan to buy back the land. Loan support of 30-40 million VND is barely enough to repurchase (chuộc) 1 ‘công’ (equivalent to 1300m2 in Southern measure- ment), and such small measure does not help solve poverty reduction problems. 32 Among the Khmer in the Mekong Delta, the practice of out migration among the Khmer in Trà Vinh dates back many years. The driving factors for out migration include limited land suitable for growing specific crops, i.e., taro… Historically regular contacts with other groups made the Khmer in Trà Vinh more adaptable to other social/cultural settings, whereas the Khmer in Sóc Trăng are more tradition-bound, and less inclined to move out of their native environments. What are the Main Drivers of Socio-Economic Performance Among Ethnic Minority Groups? 83 participation of ethnic minorities in the decision-making process on production and diversification options in the NTP SPR. 145. Access to external support initiatives in summary. External support initiatives, either by GoVN or development partners, have resulted in important outcomes in many aspects of living standards for ethnic minorities. These are reflected in improvements in physical connectivity, access to public services, housing conditions, and public utilities. The outcomes in ‘soft investments’ such as production support, capacity development, gaining voice, and participations of the ethnic minorities in community development are however less evident. The top-performing groups tend to benefit from these non-infrastructure investments more that the bottom-performing groups. One important area of external support is to improve access to credit for the poor and ethnic minorities. The ‘reluctant mentality’ to take bank loans was not really evident in the study, except for some poorest households. However, the top performers tend to use their loans more effectively in investing in crops and livestock. 146. For poverty targeting, households need to be on the official poor list to be considered eligible. Statistics on poverty and poverty reduction bear limited resemblance to the realities on the ground and are often heavily manipulated to suit local political expediency. There is also a power dynamic at play between different groups within the community when it comes to poverty assessment. In the absence of reliable poverty assessment data, the figures reported on poverty reduction could be potentially misleading indication of the development results of policies and projects for poverty reduction of ethnic minorities. Decentralization and participation have been two core principles of strategies and policies for poverty reduction. Among many mechanisms to operationalize these two principles, the most notable ones are participatory planning and commune investment ownership. However, there has been such a large gap between basic principles in policy designs and actions in practice. In addition, development schemes led by GoVN or development partners reportedly create a mentality of dependency among some people at times. It is noted that such dependency appears to be more pronounced in the bottom-performing ethnic groups. Family to the Weekly Market Research Report 84 Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam Policy Recommendations 5 How have gaps between the top- and bottom-performing ethnic minorities been addressed by the Government of Vietnam in collaboration with other development partners that are interested in poverty reduction for ethnic minorities? The answer to this question and the reported findings help to formulate the policy recommendations of this study. 5.1 How the Existing Policies Have Addressed the Gaps? 147. This study did not review existing policies on ethnic minority development but instead used the existing reviews to map the analytical framework developed for this study (CEMA, 2015; CEMA, Irish Aid, and CARE, 2018; CEMA and UNICEF, 2015). The CEMA has reviewed existing policies to formulate the basis for proposing new programming (mainly after 2020) for ethnic minorities (CEMA, Irish Aid, and CARE, 2018).33 One major policy review round was made during the design of the NTP SPR and NTP NRD in the period 2016-2020; reviews were made especially during the National Assembly’s supreme supervision of policies and programs on poverty reduction in 201434, which led to Resolution 76/2015 that laid the background for designing of poverty reduction policies in the 2016-2020 period.35 The ‘mapping’ shown in Figure 6 helps to plot how existing policies and programs have addressed the factors that determine the varying levels of development of different ethnic groups. The mapping is the basis for some useful observations as further discussed. 33 As part of this initiative, a review workshop was hosted by CEMA in October 5, 2018, to discuss the existing policies on ethnic minorities. 34 This Supreme Supervision of the National Assembly was decided by the National Assembly in 2014 to review all the policies and legal documents on poverty reduction in Vientam. The Supervision was managed by the NA’s Committee for Social Affairs (CSA). This involves a number of consultation rounds by the NA members and ex- perts to the field. Part of this Supervision was supported by the development partner community (mainly UNDP and Irish Aid though the PRPP Project). The Supervision Report was discussed by the National Assembly and that served as the background for the Resolution 76/2015 of the National Assembly. 35 Out of this round, there was one review made by CEMA (2015), where 130 policies related to socio-economic de- velopment of the ethnic minorities were listed. These were confined to the Decrees of GoVN or Decisions of the Prime Minister. If circulars or decisions of CEMA and other ministries are mentioned, the number would be very high. Notably, only 9 out of the 130 policies are under direct management of CEMA; the others are under the man- dates of line ministries. Policy Recommendations 85 FIGURE 6: Mapping of How Existing Policies Have Addressed the Gaps 86 • Preferential credit lines • Women’s Union, Farmers’ Union, and Youth Union for the poor through VBSP; to promote its membership enrollment. • Few preferential credit This links to VBSP credits lines for agri promotion • Modest support for labor export; Research Report • Limited nonfarm vocational training LAYER 2 – INDIRECT FACTORS • Capacity building for local sta • Subsidies for sta in ethnic minority areas and sta rotating • Investments in rural infrastructures Labour market Local governance • Local investment promotion LAYER 1 – DIRECT FACTORS • Mobilization of the people with high reputation (who could be village Membership elderly, religion leaders…) in propaganda Physical and Credit of associations Remittances economic Traditional connectivity institutions Landholding Gender roles and power Wellbeing • Policy to support landholding Labour of ethnic minorities • Promotion of participatory planning; for ethnic minorities Political • Principle of ‘prioritize women’s Basic public participation and bene ts’ services connectedness Knowledge, skills Market Health linkages Social network conditions • Promotion of cooperatives, and kinship Inputs, ‘tổ hợp tác’, ‘One Commune, One Product’ • Investment in education subsidies and healthcare facilities; • Fee exemptions and reduction Vulnerability Culture, spiritual and shocks Misperceptions • Disaster rescue and recovery support; • Production support mainly belief, religion • Limited support for planning, and stigma in the forms of provision enhancing resilience Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam of inputs and materials • Other direct support such as housing, water and sanitation Source: Authors’ compilation from reviewing the policies 148. Illustrated in by the bold arrows in Figure 6, the main focus of existing policies are shown as physical connectivity, access to public services (education and healthcare), and access to credit.36 With regard to physical connectivity, investing in rural infrastructure is arguably the most resource-consuming area of the existing polices. In addition to policies from central GoVN, the local authorities (both provincial and district) have tried to attract private investment in industrial zones and other public infrastructures. Although promoting this investment might not be mainly motivated by addressing poverty reduction for ethnic minorities, it could exert positive spillover to create job opportunities for ethnic minority youths. It could also contribute to improve economic connectivity to the ethnic minority areas. With regard to public services, GoVN has invested in education and healthcare infrastructures and offered fee exemptions or waivers to ethnic minorities to encourage the utilization of basic public services. Regarding access to credit, VBSP manages different credit lines for preferential loans for the poor and the near poor with ‘trust loans’ through the four mass organizations in the country (see subsection 4.10 for more details). In addition, there are several micro-finance initiatives within development partner-supported projects. These micro-finances mainly target women, some focus on ethnic minority women. 149. Illustrated by the light arrows in Figure 6 are the areas of moderate focus in existing policies. These moderate areas have been captured by a number of policies but are less as intensive as those on rural infrastructure and public services. Production support (mainly in forms of provision of inputs, free or subsidized) that promote income diversification, commodity production, market linkages is supported by many GoVN-led programs, policies and donor-supported projects. Placing a greater emphasis on this area does not seem to translate yet to practice. Within the NTP SPR 2016- 2020, for instance, infrastructure accounts for 76 percent of the total funding while production support is about 18 percent. Local governance is another moderate focus area. This has been reflected in subsidies for the GoVN staff working in the ethnic minority areas, capacity building for local staff, and staff rotation. 150. Other areas with light and dotted arrows shown in Figure 6 have been partly addressed by existing policies but with weak or limited focus. These weaker areas include some support to labor market (e.g., modest support for labor export, limited nonfarm vocational training); landholding (e.g., policies to provide production land for landless ethnic minorities), traditional institutions (e.g. mobilizing the people with reputation, including village elderly and religious leaders in propaganda for the related policies); market linkages (e.g., support to cooperatives, cooperation group — tổ hợp tác — or the recent ‘One Commune One Product’ movement); and vulnerability and shocks (e.g., having price-stabilizing arrangements rice, disaster rescue and recovery support, direct investment for climate change adaptation, and other direct support to the poor for housing, water, and sanitation). In addition, gender roles and power are also partly addressed by adaptation of participatory socio-economic development planning, incorporating the principle of ‘giving priority 36 The most useful method to make this assessment is to do a public expenditure review (PER) for policies and pro- grams on poverty reduction for ethnic minorities. However, there has been no PER of this type available to inform how important (at least in terms of resource allocations) different policies are. This is clearly an issue for future research agenda. For this study, the assessment is made based on existing review reports as well as observations from field visits, especially from the consultation with GoVN officials. However, this warrants caution in interpreting the observations drawn here. Policy Recommendations 87 to women’ in targeting strategies of some policies and programs for poverty reduction. It is noted that although there are some policies in these areas, the policies are usually under-budgeted and lack actual enforcement in practice. 151. Other areas that are highlighted in red letters in Figure 6 can more or less be considered ‘blank’ in relation to the existing policy agenda. Regarding Indirect Factors in the mapping of Figure 6 culture, spiritual belief, religion; and misperception and stigma are generally not addressed by existing policies; though these issues were discussed in some policy documents, concrete actions are lacking that could contribute to addressing these areas. With respect to Direct Factors, ‘social network and kinship’, ‘political connectedness’, and ‘remittances’ are not subject to the existing policies for poverty reduction. Most importantly, setting aside designated policies for ethnic minority groups with low populations, it is arguable that existing policies have not addressed the gaps across different ethnic minorities groups. By focusing, to a different extent, on factors that determine the wellbeing of ethnic minorities, the ultimate policy target aims at poverty reduction for ethnic minorities and hence closes the gap between the ethnic minority groups and the majority group. The findings from this study indicate that the top-performing minority groups are generally subject to better access and endowments, and hence tend to benefit more from the existing policies compared the bottom- performing groups. This could suggest that the gaps between the top and the bottom performers will widen in the future. 152. There are two important policy implications to take away from this sub-section. First, there are many factors that affect wellbeing of ethnic minorities but the existing policies toward ethnic minority development have focused on a small number of factors with intensive investment. There are many important factors that have either subject to minor focuses or even neglected. It will be important to ‘revisit’ existing policies to ensure that the different factors that determine the wellbeing of ethnic minorities are addressed. And, second, in order to close the gap between the best- and least-performing groups, the bottom performers should benefit more than the top performers; or, at a minimum, existing policies should benefit the bottom performers. Otherwise, the gap between the top- and bottom-performing groups will be widened. 5.2 Policy Recommendations 153. Policy recommendations are structured into two groups: Group 1 - refocusing ethnic minority development; and Group 2 - enhancing existing policies and mechanisms for ethnic minority development. 5.2.1 Group 1 – Refocusing Ethnic Minority Development 154. Infrastructure investments should be refocused equally to maintain the existing infrastructures in most areas. Unless being in the remotest areas or most isolated hamlets or villages, physical connectivity has been substantially improved and upgraded over the past two decade or so. This extensive focus on infrastructure investment was justified in the early stages of policies and programs in the ethnic minority areas. It was the stages when geographical locations of ethnic groups might be most important. After more than two decades of intensive investment, Research Report 88 Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam physical connectivity (as shown in this study) has been improved substantially. The focus should be now re-directed to (a) only the areas with the poorest physical connectivity where the bottom- performing groups reside (instead of spreading across all the Program 135 communes) and (b) operation and maintenance (O&M) investment, both in physical terms and in O&M capacity, in the remaining areas where physical connectivity is already high. At this stage, O&M remains at the modest level of 6.3 percent of this funding for infrastructure in the GoVN-led investments for infrastructures. In addition, it raises another question of how small-scaled infrastructures in poor areas could be linked to the national road system to improve connectivity to economic clusters, which, as reported in this study, is increasingly important for poverty reduction of ethnic minorities. 155. Soft investment, especially in production support, capacity development, and participation should be the main focus of polices for poverty reduction in the ethnic minority areas. There has been a notable increasing recognition for putting emphasis on this soft investment. However, this has not been translated into changes in the funding structure of the existing policies and programs. For example, in the current design of the Program 135 as part of the NTP SPR 2016- 2020, funding for production support accounts for 19 percent of the Program funding while capacity development accounts for 3 percent. During the design of the current Program 135 phase, there was discussion in terms of allocating at least 35 percent of the funding for production support; but this was not reflected in the current phase. In fact, the funding structure for Program 135 has been stable since the second phase of the Program in 2006-2010. At this stage, given that important improvements in physical connectivity have been made, funding poverty reduction initiatives for ethnic minorities should be re-focused on production support, capacity development, and participation. This is also a way to make better usages of the investment in improving physical connectivity. Of all these areas, giving voice and participation of ethnic minorities, especially the bottom-performing groups, should be at the core for interventions (further details follow on recommendation for participation). 156. Access to labor market should be the main focus of future policies. An important finding of this study is the increasingly important role of labor market participation in income generation and poverty reduction. This finding is evident for almost all ethnic groups, including the bottom performers. At this stage, access to labor market has not been at the center of poverty reduction efforts. The major initiative has been Project 1956 and one sub-project of NTP NRD that aims at improve vocational skills for the youth in the poor areas. Within NTP SPR, there is a sub-project (accounting for 1.3 percent of the total program funding) for supporting labor exports. To make labor market work for the poor ethnic minorities (especially women and those in the bottom-performing groups), this requires a holistic approach rather than some initiatives to address a few bottlenecks in improving access to labor market opportunities. Equal opportunity legislation needs to be consolidated with affirmative actions with regards to non-discrimination for ethnic minorities at the workplace, especially in the foreign-invested sector. One rationale behind the growing opportunities for ethnic minorities in the labor market is the low labor cost in some labor-intensive sectors; it is evident in this study that ethnic minority youths in these sectors have encountered a lot of difficulties and to some extent unequal treatment. Strengthening the enforcement of the Labor Code and the related legislation in this context is needed but equally challenging as many stakeholders, especially at the local level, might be resistant to changes in order to keep their ‘comparative advantages’ in attracting private investment. Policy Recommendations 89 157. Women’s economic empowerment should be strengthened in response to many disadvantages faced by ethnic minority women. Ethnic minority women were found with ‘double inequality’: one is the conventional gender inequality and the other is inequality of being an ethnic minority. It was found that ethnic minority women, especially those in the bottom- performing groups, are at disadvantaged positions for almost all types of access that contribute to their economic empowerment. Most importantly, the division of labor within the household (as an economic unit) constrains women from engaging in more productive income-generating activities in the villages and elsewhere. Even for the high-performing groups of the Mường, Sán Dìu, and Khmer, women working in factories are usually disadvantaged compared to men in the labor market. For women in the bottom-performing groups, access to formal wage employment is near impossible; they are ‘forced’ to engage in low-end or even illegal and risky job opportunities. There are life-changing opportunities available in the labour market, but the gaps for ethnic minority women, especially the bottom-performing groups, are wide and prevalent. The mid-term review of NTP SPR recently suggested that gender mainstreaming in poverty reduction policies requires improvements in policy framework, approaches used, and capacity to implement an upgraded gender mainstreaming agenda (MoLISA, CARE/Oxfam/SNV, 2018). 158. Focus on future policies should be placed on supporting the bottom-performing groups. This study highlights that top performers and bottom performers have responded or are affected differently to the factors that contribute to socio-economic development of ethnic minorities. It implies that external interventions from policies or projects would produce different outcomes across the ethnic groups along the wellbeing rankings. It might suggest that external interventions that work for the majority of ethnic minorities might not work for the bottom performers, thus widening the gap across different ethnic groups. For those at the top end of the wellbeing distribution, equal opportunities in the labor market and improved access to credit to promote market engagements might be most effective in overcoming poverty. For those in the bottom end of the distribution, continued investment in physical connectivity is still needed; but more focus should on improving their basic livelihood assets so that they can diversify their income sources and increasingly engage in the labor market. It is these bottom performers to whom inter- ethnic discrimination and stigma are most pronounced. It is also in these bottom performers that women are most disadvantaged. The economic positions and disadvantages of bottom performers warrant the focus for future poverty reduction efforts particularly along the bottom end of the wellbeing ranking. 159. Addressing misperceptions and social stigmas should be addressed explicitly by the policies. Misperceptions and social stigma are not explicitly recognized in most existing policy documents on ethnic minorities, though these are discussed in studies and various consultations between the GoVN and development partners. As found from study field visits, these misperceptions and stigmas held by the lowland Kinh cadres, civil servants, teachers, and medical staff define how policies, resources, and services are delivered to the ethnic minorities targeted. Given its significance, the issues of misperceptions and social stigmas against ethnic minorities, especially the bottom performers, should be officially recognized as an impediment to socio-economic development of ethnic minorities and hence could then be considered as a target of concrete policy actions. Research Report 90 Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam 160. Rationalizing and systemizing the policies and programs on socio-economic development for the ethnic minorities should be coordinated among CEMA, MoLISA, and MARD as the three key ministries involved in management of poverty reduction policies. As highlighted earlier, in a policy review by CEMA (2015), 9 out of 130 policies listed out were under the direct mandate of CEMA; the remaining 121 policies were under custody of the other line ministries. Rationalizing this existing plethora of policies and programs on ethnic minority development is challenging and have been in the process in the recent years (especially at the time of designing the NTPs for the 2016-2020 period). In the period 2016-2020, the number of NTPs has been reduced from 16 to only 2 programs, including NTP SPR and NTP NRD; and the long-lasting Program 135 is now under the NTP SPR, 2016-2020. However, there are still many other policies and programs targeted for ethnic minorities that are under the mandate of other ministries, of which CEMA has little influence. At the center of this rationalization, an effective coordination platform comprising CEMA, MoLISA, and MARD is required. Rationalizing the existing policies on ethnic minority development requires a restructuring process of poverty reduction mandates across these key players; and hence it is not possible for this rationalization process to be driven by any one of these three ministries. 5.2.2 Group 2 – Enhancing Existing Policies and Mechanisms 161. Market access, including efforts to address informal credit, should be improved. Evidence suggests that market access requires much more than improving physical connectivity. Supporting market institutions and individuals who support doing business in ethnic minority areas is essential. Recent movements toward private sector development in ethnic minority areas include the promotion of business start-ups, innovations, and public-private partnership. Business interests in community-based tourism and in agriculture, especially introducing unique, indigenous specialty products that are unique to remote areas, can provide new opportunities for ethnic minorities, including the most poorly connected, to engage in the potential market systems. New approaches that incorporate blockchain and information and communication technologies could ease many of the constraints encountered by ethnic minorities to connect with local, national, or even international value chains. The inability of ethnic minorities to engage in high-value chains (e.g., coffee or tea) is commonly understood. A breakthrough is needed to generate value-added capacities for ethnic minority farmers who fall victim to farm-gate collection and cross-border exporting. In a related note, black informal credits from private money lenders lock in many of the poorest and bottom-performing ethnic minorities in a cycle of high-cost borrowing then selling at low or pre-determined prices. In many cases, these types of unfortunate transactions end with the poor borrower having to transfer land and other assets in repayment. The SCG or community revolving funds could be pathways out of being over-reliant on informal credit sources. Promoting vertical integration along supply chains of cash crops so that processors and exporters can develop engagements with producers (e.g., in terms of advances for fertilizers or other credit support) could partly address this constraint. 162. Local governance and traditional institutions should be strengthened. Interactions are complex among local governance and institutions to socio-economic development of different ethnic minority groups and communities. But it could be argued that local governance and institutions contribute to poverty reduction when they represent or respond to the needs of the target audience, especially the needs of the most disadvantaged, bottom-performing ethnic minorities. Capacity Policy Recommendations 91 building for local-level GoVN officials should be a focus especially when ensuring cultural and ethnic sensitivity in addressing poverty for ethnic minorities. In addition, presentation of ethnic minorities in local government should be strengthened. In this regard, until ethnic minorities have key positions in local government, the policy interpretation and implementation will continue to reflect the views and experiences of the Kinh cadres who now fill these key positions. Therefore, while promoting presentation of ethnic minorities in local administration, having ethnic minorities in key positions, though challenging, should be a priority. 163. Initiatives to involve community leaders in implementing policies and projects for poverty reduction on the ground should be continued in future policies for ethnic minority development and should be expanded beyond the purpose of propaganda. Village chiefs, village elderly, and clan leaders assume important roles in community cohesion and decision-making process at the community or kinship levels. In many areas, the roles of religious heads (e.g., Protestant leaders in the Mông community) are as important as village elderly or clan leaders. To make this useful for change, it is important to put this effort in the wider context of decentralization and participation. As observed in the field, these community leaders could best contribute to community development when there are participatory processes that respect their roles and voices. These community leaders should be the focus of capacity development in the future. In addition, the roles of religious leaders in local economic development should be recognized in order to avoid the perceived hesitation by some authorities in dealing with religious leaders, especially Catholic and Protestant priests or religion group leaders. 164. While continuing to focus on capacity development for community and state officials in the ethnic minority areas, the capacity development approach and content should be completely revisited and renewed. Capacity development is recognized in almost all policies and projects for ethnic minority development. However, capacity development is usually seen as short on- off training courses or study tours in GoVN-led policies and programs. Regarding capacity development for state officials, capacity development is rarely seen as organizational development. The content of most training courses are simply to explain the policy documents and imbedded guidelines. Training on new perspectives and approaches on community development and poverty reduction was only observed in some development partner-supported projects. Regarding capacity development for communities, the standardized training framework that consists of general principles and policies of GoVN toward ethnic minority developments, Program 135, and other issues has been in place for years to serve as the main content for training of community leaders. There has been a misleading impression that capacity development activities are sufficient; there was a case of a complaint regarding too frequent training for state officials. In fact, such capacity development activities could be more or less redundant at places. What is needed is capacity development that is viewed as a process, with focus on approaches and thinking (rather than to articulate the policies and principles) in relevant capacity-building modality (rather than relying on on-off training or study tours). 165. Vulnerability and shocks should be targeted in all policies. The findings in this study indicate that the bottom performers are more vulnerable to external shocks while less capable in coping with these shocks. While market-price shocks could be the subject of improving market linkages (which is the subject of another recommendation), making the targeted ethnic minority households Research Report 92 Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam more resilient is challenging yet important in the next phase of poverty reduction programing for ethnic minorities. Recent experiences with increasing incidence of natural disaster in many ethnic minority areas suggest that investments for poverty reduction could be easily destroyed and wasted due to natural disasters. In this regard, an approach for building up resilience should be developed to fit with the context of ethnic minorities in the country. That resilient framework should be simplified to facilitate policy intervention planning and targeting. Based on that framework, resilience should be built into policies on ethnic minorities. Given the complexity of this focus, it might be useful to start some pilot interventions with support from development partners. Experiences and lessons learned should be shared for future policy dialogue in this regard. 166. The role of provincial government should be promoted in designing ethnically responsive measures. To avoid one-size-fits-all interventions for ethnic minorities, the call for being more locally and ethnically responsive has been discussed in various policy dialogues between GoVN and development partners. However, progress has been slow in this regard. With exception of the designated policies for ethnic groups with small populations, examples of how to make locally and ethnically responsible interventions remain rare. In this regard, provincial governments could pursue an important role. One principle in almost all policies on ethnic minorities introduced by GoVN at the central level is to decentralize the power to the provincial authorities. Following this principle, provincial authorities such as the Provincial Peoples Council (PrPC) and Provincial Peoples Committee (PPC) would develop specific mechanisms and guidelines to make the policies responsive to localP context and socio-economic characteristics of beneficiaries. Decentralization of power to provincial authorities could be potentially important prerequisite for making policy interventions locally responsive. However, observations from the field indicated that neither PrPC nor PPC follow procedures for making policies more responsive to the specific characteristics of ethnic groups in the provinces. In reality, the provincial line departments, which are custodian agencies of relevant policies such as PCEM, DoLISA, or DARD, submit to the PPC-relevant mechanisms for PPC to consider; if PPC agrees, it submits to the PrPC for appraisal and approval. However, observations of practice have witnessed that the provincial line departments usually mirror the central-level policies and, thus, the response to local context never materializes. To enhance decentralization, the capacity of the provincial line departments, especially PCEM, should be strengthened in order to draft policy proposals for the PPC and PrPC for making central-level policies more responsive to the ethnic minorities in the local areas. 167. Participatory planning and decentralization should be strengthened in ‘real’ terms. In relation to making policies more locally and ethnically responsive, participatory planning and decentralization are expected to contribute to this call for responsiveness by having the voices of ethnic minorities raised and reflected in the local SEDP and placing the decision-making power at local level, especially the community. However, both of these mechanisms have been undermined by various factors. With regard to decentralization, the ‘commune investment ownership’ is a very important element of local governance. Over the past 15 years or so, this investment ownership has been institutionalized. But this has been subjected, in one way or the other, to continued district- level control, especially in infrastructure investment (to where the majority of funding for poverty reduction goes). Commune block grants and commune development budgets are evidence of good practices for such decentralization. Community force accounts have also been recognized as a special investment arrangement for rural infrastructures within the two current NTPs in the period 2016- Policy Recommendations 93 2020, but implementation remains limited, especially within the resources of NTP SPR. Evidence is sufficiently available to continue this decentralization, but strong enforcement of decentralization can only be ensured by clear instruction and strong determination from central and provincial levels. 168. Participatory socio-economic development planning should be institutionalized at the central level to guide the local prioritization process. Participatory planning is a central mechanism for ethnic minorities to engage in the prioritization process for resource allocation at the local level. Participatory planning was introduced by some development partners-supported projects in the 1990s as a mechanism to prioritize the project activities. It was then used in Program 135, second phase 2006-2010, as well as some other poverty reduction programs in the country. Based on the lessons learned from participatory planning for projects, the ideas of making local SEDP participatory (i.e., to mainstream the participatory planning to the GoVN system) were tested and then scaled up over the past decade or so. At present, about half of the total provinces in the country has institutionalized the participatory SEDP process (at least at the commune level). However, adaptation of participatory planning in the policies and programs for ethnic minorities remains challenging. In the provinces where the SEDP processes were institutionalized, the participatory SEDP is usually followed in the area targeted by the development partners-supported projects. In the projects’ non-targeted areas, the level of participation varies from limited participation to none. The lack of a uniform regulation or guideline on local SEDP from the central level represents a major impediment to the adaptation of participatory SEDP in practice. In this context, evidence of strong participation of ethnic minorities in planning the policies and programs on poverty reduction was rarely observed in the sites covered by this study. Therefore, it is recommended that the participatory SEDP should be institutionalized consistently from the central to the provincial level; and accordingly, all resources available from poverty reduction policies and programs should be integrated into a single SEDP process where ethnic minorities (and other vulnerable groups) can share their influence in the decision-making process. 169. Availability and quality of data on ethnic minority development, especially the 53EMS, should be strenghthened. The 53EMS 2015 used in the current study is the first in the series of the survey on 53 ethnic minorities in Vietnam. It is expected that this survey will be repeated every five years.37 This represents a major development in data availability on ethnic minority development in Vietnam. It is recommended that the survey content should be revised and improved, especially the parts related to livelihood activities, culture and tradition, and household income. In addition, a module on household expenditure should also be considered in the next round. Having rotating modules on thematic areas should be considered as another options. Notably, data dessimination policy should be revised to get the data available in the public domain for research and policy analysis. 37 Refer to the Decision 02/QD-TTg dated May 05, 2015 of the Prime Minister on approval of the series of the Survey on 53 ethnic minority groups. Research Report 94 Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam Conclusions 6 To conclude the study, the following summary in Table 15 shows how different factors exert influences that differentiate the top- and bottom-performing ethnic minorities group. In addition, some caveats are discussed with suggestions for future research agenda. 6.1 A Summary Driving Forces of the Top or Bottom Performers 170. The analysis in Section 3.3 discussed how different factors described in Figure 3 that affect the development of the selected top- and bottom-performing ethnic groups differently. A summary of how these factors influence the top- and bottom-performing groups is below. Sparkling terraced rice fields Conclusions 95 TABLE 15: Summary of How the Factors Affect the Selected Top- and Bottom-Performing Ethnic Minority Groups 96 Mường Sán Dìu Khmer Xơ Đăng Khơ Mú Mông Some basic Ling (Vietic); Pop Ling (Chinese); Ling (Mon-Khmer); Pop Ling (Mon-Khmer); Pop Ling (Mon-Khmer); Pop Ling (Mông-Mien); indicators (163,650); HDI Pop (1,378,938); (1,279,567); HDI (0.524); (195,673); HDI (0.4494); (84,344); HDI (0.414); MPI Pop (1,244,137); HDI (0.5868); MPI HDI (0.5679); MPI MPI (34.5) MPI (82.83) (91.72) (0.3852); MPI (88.17) (27.09) (40.77) Areas of field 03 provinces, 03 01 province, 02 02 provinces, 03 districts, 02 provinces, 03 districts, 02 provinces, 02 districts, 03 provinces, 03 Research Report visits districts and 03 districts and 02 and 03 communes: Sóc and 03 communes: Kon and 03 communes: Sơn La districts, and 03 communes: Sơn La communes: Thái Trăng (Mỹ Tú/Phú Mỹ; Tum (Đắk Tô/Kon Đào; (Mai Sơn/Hát Lót), Thanh Hóa communes: Sơn La (Phù Yên/Mường Nguyên (Đồng Trần Đề/Đại An 2), Trà Ngọc Hồi/Bờ Y); Quảng (Mường Lát/Tén Tằn & Mường (Bắc Yên/Xím Vàng), Thải), Thanh Hóa Hỷ/Nam Hòa; Phú Vinh (Trà Cú/Định An) Nam (Bắc Trà My/Trà Đốc) Chanh) Thái Nguyên (Đồng (Ngọc Lạc/Thạch Bình/Bàn Đạt) Hỷ/Văn Lang), Thanh Lập), Kon Tum Hóa (Mường Lát/Pù (Ngọc Hồi/Bờ Y) Nhi) Physical and Better access (e.g., Better access (e.g., Better access (e.g., Less access (e.g. Ca Limited access (e.g., settle on Limited access due to economic residential pattern residential pattern residential pattern Dong living in isolation hillside with medium slope high-altitude habitat; connectivity in lowland area and in lowland area) in lowland area); yet, dispersed in small gradient) lack of transport this also remains for differences emerge villages) facilities; some areas the migrant Mường regarding physical were only connected e.g. to the Central geography with the road system Highlands) in most recent years Basic Better endowed, Better endowed; Better endowed; more Less access, yet, Xơ Đăng Limited access, less Limited access, less household more experience more experience experience with farm have more experience experience with farm experience with livelihood with farm with farm diversification, yet, with cash crop cultivation diversification, yet, farm diversification; assets diversification; but diversification. differences due to than Ca Dong; high differences emerge high dependency on differences emerge Better capacity to variable geographic dependency on natural regionally due to history of natural resources; regionally as a cope with shocks, conditions; resources; settlement; high dependency Differences emerge result of migration less risk averse to Land ownership varies Reported practice of on natural resources; as a result of Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam experiences. adapt changes significantly from selling land by the poor Reported practice of selling resettlement. Better capacity to excessive land possession to the better-off Xơ Đăng land by the poor to the Limited capacity to cope with shocks, to completely landless in the same hamlet; better-off Khơ Mú in the same cope with shocks, high less risk averse to Moderate capacity to Less capacity to cope hamlet risk averse to adapt adapt changes cope with shocks, less risk with shocks, less risk Limited capacity to cope with changes averse to adapt changes averse to adapt changes shocks, high risk averse to adapt changes Mường Sán Dìu Khmer Xơ Đăng Khơ Mú Mông Market Better access, an issue Better access, able Better access, yet less Less access, less Limited access, being Limited access, being linkages of manipulation of to keep track with available informal dependent on local obliged to sell their obliged to sell their market and report market prices via credit from family & retailers; less available products at pre-fixed products at pre-fixed by Kinh retailers but television, the friends; informal credit from prices to agents - prices to agents - varies regionally; internet; Emergency loans family & friends; following the ‘borrow following the ‘borrow able to keep track No issue of market (from local money Emergency loans (from now, pay back later’ now, pay back later’ with market prices manipulation lenders) often charge local money lenders) formula; less available formula; less available via television, the reported; the high interest; able often charge high informal credit from informal credit from internet; availability of availability of credit to keep track with interest family & friends; family & friends; informal credit from from family & friends market prices via Emergency loans often Emergency loans often family & friends at television, the internet charge high interest; charge high interest; low rates, yet varies limited or no market limited or no market regionally power in dealing with power in dealing with shop owners or traders shop owners or traders Labor market Better access, but Better access, known Better access, known Less access, but Limited access, but Limited access, mobility preference and as actively engaging in as actively engaging preference varies (e.g., differences emerge but mainly men or significance vary cross-border trading in migration & a handful of Ca Dong regionally (e.g., unmarried, female regionally due to activities; good skills cross-border trading start engaging in both Khơ Mú TH start youth who dropped the availability of in gold panning, activities (with Khmer domestic migration engaging in both out in public secondary manufactures in the mining; fellows in Cambodia); & international labor domestic migration education or high vicinity & household Remittances and incomes from migration (mainly Saudi- & international labor school engage in farm workloads; wage income account remittances play a Arabia); local youth are migration (mainly interprovincial remittances and for a large share of vital role in changing not prepared to accept Saudi-Arabia); local migration for factory wage income account household income the physiognomy of strict working schedule youth are not prepared work; local youth for a large share of the village at enterprises; experience to accept strict working are not prepared to household income of unjust treatment & schedule at enterprises; accept strict working non-transparent salary experience of unjust schedule at enterprises; payment in some factory treatment in some experience of unjust work factory work treatment in some Conclusions factory work 97 Mường Sán Dìu Khmer Xơ Đăng Khơ Mú Mông 98 Some aspects of Lower rate of Lower rate of A vernacular High rate of school Significant rate of Significant rate of access to education secondary school secondary school education system dropouts at 2nd or 3rd school dropouts at school dropouts at dropout; higher dropout; higher teaching Khmer levels of secondary; 2nd or 3rd levels of 2nd or 3rd levels of attendance in college; attendance in college; language in Buddhist Limited attendance secondary; secondary; a sense of disillusion a sense of disillusion wat; in college; a sense of Very limited Very limited about the benefits about the benefits Lower rate of disillusion about the attendance in college; attendance in college; Research Report of pursuing higher of pursuing higher secondary school benefits of pursuing a sense of disillusion a sense of disillusion education education dropout; higher higher education about the benefits about the benefits attendance in college; of pursuing higher of pursuing higher a sense of disillusion education; language education; language about the benefits barrier is more barrier is more of pursuing higher pronounced pronounced education Some aspects of Better use; low fertility Better use; low fertility Better use; low fertility Limited use; high Limited use; the Limited used; access to health care rate; low homebirths; rate; low homebirths; rate; low homebirths; fertility rate to the frequency of significant fertility rate services child marriage is less child marriage is less child marriage is less absence of birth homebirths; child due to the absence of common common common control for religious marriage is high; birth control because reasons; the frequency consanguineous of preference for of homebirths; child marriage is very high sons over daughters; marriage is quite high compared to other the frequency of groups homebirths; child marriage is high; relatively high consanguineous marriage Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam Mường Sán Dìu Khmer Xơ Đăng Khơ Mú Mông Traditional Based on strong clan/ Based on strong Community cohesion The significant role Based on strong Based on strong institutions & local lineal ties, yet ties clan/lineal ties; the is built on religious of village elderly; clan/lineal ties; clan/lineal ties; governance become looser in new village head is a key faith, with the pivotal the village head is both the village both the village settlements; player in conveying role of the wat, abbot an emerging key head & the family head & the family Social cohesion relies government policies & achar both in the player in conveying clan head are key clan head are key on factors such as to villagers, along spiritual world & social government policies players in conveying players in conveying living in the same with village mass and cultural activities; to villagers; common government policies government policies locality and sharing organizations; no influence of kinship practice of labor to villagers, along to villagers, along the same place of common practice in the process of exchange; less active with village mass with village mass origin; common of labor exchange, assessing household participation in mass organizations; very organizations; practice of labor no representation in poverty and gaining organizations; common practice important role of exchange; local administration access to resources Political participation of labor exchange; Protestant priest or Generally better (commune level); at grassroots level; seems to be strongest limited participation ‘group leader’; very political participation, more active nonexistence of labor among Xơ Đăng in mass organizations; common practice yet varies regionally; participation in mass exchange practice; (commune, district, no political of labor exchange; more active organizations Significant provincial level), participation limited participation participation in mass representation while the Ca Dong are in mass organizations; organizations throughout the state rarely represented in political participation apparatus; more commune level is only pronounced in active participation in Mông communes mass organizations Conclusions 99 Mường Sán Dìu Khmer Xơ Đăng Khơ Mú Mông 100 Gender roles & intra- Traditionally Traditionally Bi-lateral succession, Bi-lateral succession, Patriarchal, favoring Highly patriarchal, household power patriarchal, men patriarchal, favoring equal division of women’s presumably men over women; men is the decision dynamics assume control of men over women, household property, important and early marriage & maker; but more everything; yet, beginning to regardless of gender; traditional roles; men consanguineous reported cases of Women participate in change i.e., men Both young men and are understood to marriage rarely housechore shared productive activities, share the household women increasingly engage in matters practiced between (young) Research Report take care of children burden with female experienced that are ‘important’ or husband and wife; and other household household members; outmigration; ‘vital’, financial control by work; both young men and Middle aged i.e., nonfarm income- women in the role of Both young men & women increasingly women & men get generating activities mother in-law; young, women increasingly experienced limited access to that sustain the married women experienced outmigration; more factory jobs (due household. have no access to outmigration; reported cases of joint to low educational household resource; Middle aged household decisions attainment); family issues of women & men get made by husband & Unpaid childcare jealousy and social limited access to wife. burden on middle- stigma regarding factory jobs (due aged and elderly women migrating to to low educational women. other places on their attainment); own; bride kidnapping Custom of cohabiting prevail among the with elder parents Mông in SL, TH but after marriage; more not among the Mông reported cases of joint in Thái Nguyên household decisions made by husband & wife, but the wife voluntarily let the Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam husband have the final say. Mường Sán Dìu Khmer Xơ Đăng Khơ Mú Mông Perceptions of No language barrier; No language barrier; Believers of Theravada Catholic followers; Language barrier A proportion of ethnicity & dynamics considerably more Chinese linguistic Buddhism; a small Language barrier is more significant Mông is Protestant; of trust adaptable, and better proximity; more proportion of middle- is less significant; among women, language barrier is ‘integrated’ (albeit adaptable and better aged women pop considerably especially the middle extremely significant selectively) into ‘integrated’, a sense of facing language less adaptable to and elderly; yet, an among women; yet, mainstream culture, interiority compared barrier, still; more the mainstream increasing number an increasing number a sense of interiority to the Kinh adaptable and better culture and society; of secondary schools of secondary schools (compared to the ‘integrated’, a sense of internalized to some completed women completed women more predominant interiority compared extent popular with more Kinh with more Kinh ethnic group like the to the Kinh & Hoa discourses on their language proficiency; language proficiency; Thái in Sơn La, Kinh in backwardness, considerably least considerably least Kon Tum) self-deficiency and adaptable to the adaptable to the superstition mainstream culture mainstream culture and society; strongly and society; strongly internalized with internalized popular popular discourses on discourses on their their backwardness, backwardness, self-deficiency and self-deficiency and superstition superstition Access to external Significant access; Better access; Significant access; Less access due to Limited access due Limited access due support initiatives the issue of more the issue of more the issue of more lack of finance & a to lack of finance to lack of finance dominant family dominant family dominant family sort of risk avoidance & a sort of risk & a sort of risk clans mostly taking clans mostly taking clans mostly taking mentality, or avoidance mentality, avoidance mentality, advantages; a sort advantages (to a less advantages (to a less reluctance of taking said dependency , said dependency of risk avoidance degree) degree) bank loans; limited on support, on support; limited mentality (to a less participation in limited political participation in degree & varies planning connectedness; planning regionally) no participation in planning Conclusions Source: Compilation from the main findings 101 6.2 Main Drivers of Socio-Economic Development among Ethnic Minority Groups 171. The findings of this study indicate an intricate web of interrelated conditions and factors that significantly contribute to the route out of poverty among selected ethnic minorities in Vietnam. A summary of both successful and failed episodes is presented in the following. 6.2.1 Pathways to the Top 172. Successful outcomes in the task of reducing or escaping poverty among the best- performing groups indicate a combination of key factors: • Residing in the areas that are easily connected with basic infrastructures and economic clusters that offer job opportunities. Given the recent improvements in physical infrastructures, connectivity to economic cluster have become increasingly important. • Engaging in cash crops that are not already ‘controlled’ by the lowland Kinh in the locations such as the key agriculture value chains of the country. These might be the cash crops that serve local niche markets or the ones that are indigenous to the area. • Actively looking for labor market opportunities, especially as wage employment. This is adopted as a household-level strategy so that within-household division of labor was changed accordingly to give opportunities for the ‘employable’ adults, including women, to engage in the labor market opportunities. • Finding ways to interact actively with opportunities and stakeholders that determine allocation of resources and, on that basis, make good usage of the resources available from different policies and projects, including access to credit. The ability to finding these ways depends on political connectedness, participation in the local administration, or local mass organizations. • The crucial role of the key community members (hamlet cadres, representatives of traditional institutions, and local residents) provided that these are given proper conditions to develop self-reliance and relational autonomy in partaking in community development initiatives towards poverty sustainable reduction. • Factors constituting and enabling connectedness depends on the ability of ethnic minorities to adapt selectively to certain aspects of Kinh culture and therefore become less subject to misperceptions or social stigma inserted by the Kinh-dominated society. This contributes to closing the gap and creating ‘equal footing’ between these ethnic groups and the Kinh majority in accessing opportunities as well as the returns from these opportunities. 6.2.2 Trapped at the Bottom 173. The trap of ‘being left behind’ in the fight to reduce poverty or to escape poverty sustainably among the least-performing groups is caused by a combination of the pivotal factors that are indicated as follows: Research Report 102 Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam • Being poorly connected to physical infrastructures though mainly attributed to distance or quality of access rather than availability of access to infrastructures, which has been substantially improved over the past two decades. Being poorly connected to economic opportunities, especially labor market opportunities, is a major constraining factor for escaping poverty. • Lack of availability of cash crop opportunities, which are caused either by poor livelihood potentials in the locations or external factors such as being resettled or the loss/interruption of access to natural resources (such as forest) that contribute to the household livelihoods. More importantly, lack of capabilities (e.g., labor, productive landholdings) and risk-taking to engage in (existing or emerging) cash crop opportunities. • Constraints to access to labor market opportunities, especially wage employment in factories. These constraints are linked to low educational attainment levels, language barriers, cultural barriers caused by lack of exposure to new environment, and gender-biased household division of labor that traps women in house-chores and agriculture work in the local areas. • Inability to develop an active, proper role for various community members (hamlet cadres, representatives of traditional institutions, and local residents) as well as an inability to create opportunities to develop self-reliance and relational autonomy in partaking in community development initiatives toward poverty sustainable reduction. This is intensified by the absence of effective mechanisms to facilitate voices and participation; as a consequence, the bottom- performing groups tend to share little in local decision-making process. • Inability and/or reluctance to establish connectedness and to adapt to Kinh culture and society, thus perpetuating prejudices especially among least-performing groups. These prejudices are widespread among the Kinh cadres, teachers, healthcare staff, and employers, and therefore undermine access to opportunities as well as public services. 6.3 Limitations of the Study and Directions for Further Research 174. Some limitations of the study should be noted when interpreting the findings in this report. Regarding sampling for the qualitative research, a limitation to this study was that research sites were selected based on direct consultation with authorities dealing with ethnic minority affairs at the provincial and district levels based on criteria as mentioned in Section 3.4. Such an approach inherently lacks objectivity. The research team tried to correct this by complementing research data with available secondary sources of information (mainly from 53EMS and existing studies) and use its expertise and fieldwork experience to arrive at a well-balanced picture. Moreover, the fact that participants for in-depth interviews and group discussions were selected via village cadres (even though there were general criteria available for this selection). These might be relatives or acquaintances of these cadres and therefore might not reflect the real poverty situation on the ground. To obtain a more complete picture, the study team managed to gather additional information from contacting local residents of various backgrounds in their daily activities and in informal settings. However, this was constrained by the pre-arranged agenda. 175. The nature of qualitative research data collection could result in a social-desirability bias, meaning that respondents could answer questions in a way they believe makes them look good or distressed. However, this was mitigated because the researchers conducting the Conclusions 103 interviews and group discussions were experienced in remaining non-judgmental and objective during interactions with study participants. Also, the researchers were independent of the intervention developments, thus minimizing the immediate social desirability characteristics during the data collection interviews or discussions. There were few language barriers on sites since most interlocutors were able to communicate in Vietnamese. The team did meet some women who could not communicate in the national language among the Mông in Sơn La and Thanh Hóa, and the Khơ Mú in Thanh Hóa. In these cases, the team had to rely on local interpreters, usually local cadres or members of the Women’s Union. A co-research approach, meaning to find and collaborate with some locals to be co-researchers, was considered at the start of this study but was considered to be not useful for the capacity-building component of the research. 176. For future research, there are a few issues that this study did not address (or briefly discuss) either because of constraints in the in-depth information available or to keep the report balanced across different factors that determine the socio-economic development of different groups. These factors should be covered in future research agendas. First, as highlighted in the study, wage employment is now the major driving force for poverty reduction. There are increasing numbers of ethnic minority youth who migrated to engage in these labor market opportunities. However, it was not possible to examine the living and working conditions at the destinations, especially workplace features. These aspects are important to reflect the ‘quality’ of these opportunities and hence to understand the dynamics of labor market participation by ethnic minorities. 177. Second, the issue of religious practices was reported as an increasingly important aspect of wellbeing, at least from the spiritual side. There are some interactions between religious practices and other socio-economic activities of ethnic minorities. In particular, religious practices affect choices in terms of gender roles, power dynamics, job opportunities, and even interactions with the local authorities. Therefore, it might be reasonable to argue that such religion practices might exert influence on the path to the top or bottom. However, this issue has not been captured thoroughly in this study and would be an interesting area for future research. 178. Finally, it is noted that comparisons within the selected ethnic groups are not informative in this study report. Although the selection of the study sites was made to ensure that one selected ethnic group could be studied in different locations, these differences were not pronounced from the fieldwork and as a result the report is quite salient in informing such within- group differences. An elderly Hmong woman Research Report 104 Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam References Bonnin, Christine and Sarah Turner. 2014. “‘A good wife stays home’: gendered negotiations over state agricultural programmes, upland Vietnam.” Gender, Place & Culture 21.10 (2014): 1302-1320. Bonnin, Christine. 2011. Markets in the mountains: upland trade-scapes, trader livelihoods, and state development agendas in Northern Vietnam. McGill University. CEMA and UNICEF. 2015. Multidimensional Child Poverty: Situations from the Program 135 communes, a research report commissioned by UNICEF to IRC consultants. CEMA, 2015. Reviewing the policies on poverty reduction for ethnic minorities, a review made by CEMA to send to MoLISA to compile the report to the GoVN on reviewing the poverty reduction as part of the Surpreme Supervision of the National Assembly on poverty reduction strategies and policies. CEMA, Irish Aid, and CARE. 2018. Review of policies on ethnic minorities 2016-2018 and orientation for 2021- 2025, a study commissioned to IPSARD (work in progress). CEMA, Irish Aid, and UN Women. 2017. Figures on Ethnic Minority Women and Men in Viet Nam 2015. Based on the Results of the Survey on the Socio-economic Situation of 53 Ethnic Minority Groups in Viet Nam 2015 (UN Women: Hanoi). CEMA, UNDP, and Irish Aid. 2017. The Ethnic Minority Survey Report, a research commissioned to MDRI consultants. Deininger, K., A. Goyal, and H. Nagarajan. 2010. Inheritance Law Reform and Women’s Access to Capital: Evidence from India’s Hindu Succession Act. Policy Research Working Paper 5338, World Bank, Washington, DC. Dutta, Puja Vasudeva. 2018. Access and use of social assistance benefits among ethnic minorities in Vietnam. Draft report. In. Hanoi: World Bank. EMWG (Ethnic Minorities Working Group). 2014. “Critical Issues in Achieving Sustainable Development of Ethnic Minorities in Vietnam”, in, International Conference on Sustainable Development and Ethnic Minority Poverty Reduction in Mountainous Regions, June 11-13, 2014. World Bank and Thái Nguyên University Publishing House). Hager, Jorg. 2006. Effects of the Land Allocation Process and Market-Oriented Economy on Common. Grazing Landof Ethnic Minorities in Northwest Vietnam. Paper presented at the Eleventh Biennial Global Conference of the International Association for the Study of Common Property (IASCP): “Survival of the Commons: Mounting Challenges & New Realities”, June 19 – June 23. Bali, Indonesia. http://dlc.dlib. indiana.edu/dlc/bitstream/handle/10535/1987/Hager_joerg_effects.pdf?sequence=1&isAllowed=y ILO (International Labour Organization). 2016. ASEAN in transformation. The future of jobs at risk of automation. Bangkok: International Labour Organization. https://www.ilo.org/actemp/publications/ WCMS_579554/lang--en/index.htm Le Dan Dung. 2018. Moving out of Home: Negotiating Gender for Personal Transformation of Mông References 105 Women Working in Tourism-related Activities in Sa Pa, Vietnam. Journal of Mekong Studies 14 (1): 1-16. Markussen, Thomas. 2017. “Social and Political Capital”, In F. Tarp (ed) Growth, Structural Transformation, and Rural Change in Viet Nam: A Rising Dragon on the Move. Oxford, UK: Oxford University Press. MOH and UNFPA. 2017. Exploring barriers to accessing maternal health and family planning services in ethnic minority communities in Viet Nam. Hanoi: UNFPA MOH and World Bank. 2018. Social Assessment Report. Hanoi: Investing and Innovating for Grassroots health care service delivery reform project.MOLISA, OXFAM, CARE, SNV. 2018. Gender assessment at the Mid Term of the National Target Program on Sustainable Poverty Reduction 2016-2020, a thematic study commissioned by OXFAM, CARE and SNV (within WEAVE Project) to contribute to the MTR. Nguyen Cong Thao, Tran Hong Hanh, Mai Thanh Son and Nguyen Van Suu. 2012. Vietnam’s Central Highlands: Analysis of Local Institution and Leadership. Study Report. Hanoi: The World Bank. Nguyen Thu Huong. 2016. “The Red Seedlings of the Central Highlands’: Social Relatedness and Political Integration of Select Ethnic Minority Groups in Post-War Vietnam”, in P. Taylor (ed.) Connected and Disconnected in Vietnam: Remaking Relations in a Post-socialist Nation. Canberra: ANU Press. http://press. anu.edu.au/titles/vietnam-series/connected-and-disconnected-in-viet-nam/ Nguyen Thu Huong. 2018. “Conceptualization of gender relations in agriculture: Intersections between household dynamics, social capital, and ethnicity among selected ethnic minority groups in Northwestern Vietnam.” Seminar co-organized by the Australian Centre for International Agricultural Research and the Department of Anthropology, Vietnam National University, Hanoi, May 17. Nguyen Van Chinh. 2008. “From Swidden Cultivation to Fixed Farming and Settlement: Effects of Sedentarization Policies Among the Kmhmu in Vietnam”, Journal of Vietnamese Studies, 3 (3): 44-80. Rheinländer, T., Samuelsen, H., Dalsgaard, A., & Konradsen, F. 2010. Hygiene and sanitation among ethnic minorities in Northern Vietnam: does government promotion match community priorities? Social science & medicine, 71(5), 994-1001. Silver, H. 2013. Framing Social Inclusion Policies. Background paper, World Bank, Washington, DC. UNICEF. 2016. The apparel and footwear sector and children in Viet Nam. Hanoi: UNICEF. https://www. unicef.org/vietnam/reports/apparel-and-footwear-sector-and-children-viet-nam Wells-Dang, Andrew. 2012. “Ethnic Minority Development in Vietnam: What Leads to Success?” In. Hanoi: Background Paper for the 2012 Programmatic Poverty Assessment. World Bank. White, Joanna, Pauline Oosterhoff and Huong Thi Nguyen. 2012. Deconstructing ‘barriers’ to access: Minority ethnic women and medicalised maternal health services in Vietnam. Global Public Health, 1-13 World Bank. 2018a. Climbing the ladder: poverty reduction and shared prosperity in Vietnam (English). Washington, D.C.: World Bank Group. http://documents.worldbank.org/curated/ en/206981522843253122/Climbing-the-ladder-poverty-reduction-and-shared-prosperity-in- Vietnam ———. 2018b. Vietnam Future Jobs Report.World Bank: Washington DC. ———. 2018c. Vietnam Multisectoral Nutrition Assessment and Gap Analysis. Draft report. Hanoi: The World Bank. ———.2013. Inclusion Matters. World Bank, Washington, DC. ———. 2011. Vietnam Country Gender Assessment. Report 65501. Hanoi: The World Bank. ———. 2009. Country Social Analysis: Ethnicity and Development in Vietnam (Vol. 2): Main report (English) (World Bank: Washington, DC). Research Report 106 Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam Annex Details of the Study Sites Annex 107 HONG DUC PUBLISHING HOUSE Vietnam Laywers Association Hong Duc Publishing House • Address: 65 Trang Thi Street, Hoan Kiem District, Hanoi • Email: nhaxuatbanhongduc@yahoo.com • Tel: (84 24) 3 9260024 – Fax: (84 24) 3 9260031 Publishing responsible: Bui Viet Bac, Director Editor: Nguyen Thi Phuong Mai Graphic design: Hong Duc Publishing House Cover Photo: Family to the Weekly Market Print 150 copies in English, size 20.5cm x 28.5cm at Xuan An Co. Ltd. Address: 199 Quan Tho 1, O Cho Dua, Dong Da, Hanoi Publishing permit: 903/QĐ-NXBHĐ Publishing registration plan: 1304-2019/CXBIPH/63 -21/HĐ ISBN: 978-604-89-8586-8 Compeleted and archived in 2019 With support from: 8 Dao Tan Street, Ba Dinh District, Hanoi, Vietnam Telephone: +84 24 37740100 Facsimile: +84 24 37740111 Website: www.dfat.gov.au 8th Floor, 63 Ly Thai To Street, Hoan Kiem District, Hanoi, Vietnam Telephone: +84 24 39346600 Facsimile: +84 24 39346597 Website: www.worldbank.org/en/country/vietnam Research Report 110 Drivers of Socio-Economic Development Among Ethnic Minority Groups in Vietnam