CLIMBING THE LADDER POVERTY REDUCTION AND SHARED PROSPERITY IN VIETNAM UPDATE REPORT 2018 CLIMBING THE LADDER POVERTY REDUCTION AND SHARED PROSPERITY IN VIETNAM VIETNAM POVERTY AND SHARED PROSPERITY UPDATE REPORT II CLIMBING THE LADDER: POVERTY REDUCTION AND SHARED PROSPERITY IN VIETNAM Currency Equivalents Exchange Rate Effective as of December 15, 2017 Currency Unit = VND (Vietnamese Dong) VND 22,760 = US$1.00 Fiscal Year = January to December Acronyms and abbreviations ECD Early Childhood Development FDI Foreign Direct Investment GSO General Statistics Office PPP Purchasing Power Parity VHLSS Vietnam Household Living Standards Survey VND Vietnamese Dong Regional Vice President : Victoria Kwakwa Country Director : Ousmane Dione Senior Practice Director : Carolina Sanchez-Paramo Practice Manager : Salman Zaidi Task Team Leader : Obert Pimhidzai CLIMBING THE LADDER: III POVERTY REDUCTION AND SHARED PROSPERITY IN VIETNAM ACKNOWLEDGEMENTS This poverty and shared prosperity update report was prepared by Obert Pimhidzai – Economist, with contributions from Linh Hoang Vu – Consultant, Sergiy Zorya (Senior Economist), Alwaleed Fareed Alatabani (Lead Financial Sector Specialist), Nga Thi Nguyen (Social Protection Specialist), and Sebastian Eckradit, (Program Leader, EACVF). The report benefited from comments from the following peer reviewers: Kenneth Simler (Senior Economist), Robertus A Swinkles (Economist), Clarence Tsimpo Nkengne (Senior Economist), Minh Cong Nguyen (Economist). The team benefitted from a close collaboration with the GSO and would particulary like to thank Do Anh Kiem (Director, Social and Environmental Statistics Department, GSO), Nguyen The Quan (Deputy Director of Social and Environmental Statistics Department, GSO) and Lo Thi Duc (Senior Statistician, Social and Environmental Statistics Department, GSO). The report was prepared with guidance from Salman Zaidi, (Practice Manager, GPV02). Sean Lothrop (Consultant), edited the report. TABLE OF CONTENTS ACKNOWLEDGEMENTS.............................................................................. III EXECUTIVE SUMMARY.................................................................................1 INTRODUCTION............................................................................................4 VIETNAM’S REMARKABLE SUCCESS STORY CONTINUES Poverty and shared prosperity trends: 2010-16....................................................6 Poverty declined everywhere for both ethnic minorities and the majority..........................................................................................................................6 Poverty reduction reflects significant improvement at all levels of welfare...................................................................................................................................7 But consumption growth is becoming less pro-poor over time......................8 Growth, rather than redistribution, drove the recent decline in poverty...................................................................................................................................9 Non-monetary indicators of welfare show improvements but gaps persist....................................................................................................................... 10 Economic Mobility: Chasing the middle class dream....................................... 12 The middle class is expanding as households move up the economic ladder............................................................................................................. 13 The risk of falling into poverty is low and declining.......................................... 14 Vietnam’s recipe of success........................................................................................... 17 Wage income growth drove poverty reduction...................................................18 Agriculture transformation contributed to poverty reduction in rural areas...........................................................................................................................20 Household businesses are important for attaining economic security.......21 CLIMBING THE LADDER: V POVERTY REDUCTION AND SHARED PROSPERITY IN VIETNAM LEAVING NO ONE BEHIND Who is left behind?......................................................................................................................................................................... 23 Poverty is increasingly concentrated in rural areas and among ethnic minorities.............................................. 23 The poor are constrained by a lack of human, physical, and financial capital....................................................... 25 Finishing the job and addressing new challenges.......................................................................................................... 30 Labor productivity and skills are key pillars to sustain high wage income growth............................................. 30 Adapting the social protection system to suit an export oriented wage economy will be necessary......... 32 Changing land use and strengthening land property rights will further unlock the agriculture potential of the poor and near poor...................................................................................................................................... 32 Equalizing opportunities in education is central to the poverty and shared prosperity agenda................... 32 Priorities for poverty reduction and shared prosperity............................................................................................... 35 References........................................................................................................................................................................................... 37 Annexes: Additional Tables and Figures............................................................................................................................... 38 LIST OF MAPS Map 1: Poverty Rate by District 2014...................................................................................................................................... 24 LIST OF TABLES Table 1: Poverty Trends by Region, 2010-16.............................................................................................................................7 Table 2: Trends in Inequality, 2010-16.........................................................................................................................................8 Table 3: Trends in Nonmonetary Welfare Indicators, 2010-16......................................................................................... 10 Table 4: Living Conditions Indicators by Economic Class, 2016...................................................................................... 13 Table 5: Transitions into and out of Poverty: 2014-16........................................................................................................ 14 Probability of Being Poor (Based on GSO-WB Boverty Line) in Terminal Period by Economic Table 6:  Status in Base Year: 2010-16........................................................................................................................................ 14 Table 7: Economic Mobility by Household Characteristics, 2014-16............................................................................ 15 Table 8: Income Decomposition of Poverty in Vietnam, 2014-16.................................................................................. 18 Table 9: Number and Bistribution of Poor People in Vietnam, 2010-16....................................................................... 23 VI CLIMBING THE LADDER: POVERTY REDUCTION AND SHARED PROSPERITY IN VIETNAM Table 10: Poverty Rate and Population Distribution by Topography, 2016................................................................... 24 Table 11: Average HouseholdAgriculture Income by Crop and Topography, 2016................................................... 27 Table 12: Income Decomposition of Changes in Middle Class in Vietnam, 2014-16................................................. 38 LIST OF FIGURES Figure 1. Poverty Rates Based on National and International Poverty Lines, 2010-16................................................6 Figure 2. Poverty Rates by Ethnic Group, 2010-16....................................................................................................................6 Figure 3. Regional Decomposition of Poverty Changes in Vietnam, 2012-16................................................................7 Figure 4. Poverty Dominance Curves, 2014-16..........................................................................................................................8 Figure 5. Per Capita Consumption Growth, 2010-16................................................................................................................8 The Relative Contributions of Consumption Growth and Distributional Equity to Changes in the Figure 6.  Poverty Rate, 2014-16........................................................................................................................................................9 Figure 7. Trends in Annual Consumption per Capita, 2010-16.......................................................................................... 11 Figure 8. Probability of Being Poorin 2016 Conditional on Economic Class in 2014................................................. 12 Figure 9: Population by Economic Class, 2010-16.................................................................................................................. 13 Figure 10. Trends in Economic Mobility, 2010-16................................................................................................................... 14 Figure 11. Framework for Income-poverty Decompositions............................................................................................. 17 Figure 12. Trends in Seasonally Adjusted Average Real Monthly Wages by Economic Sector, 2013-17............ 18 Figure 13. Trends in Seasonally Adjusted Average Real Monthly Wages by Industry, 2013-17............................. 18 Figure 14. Major Source of Household Income, 2010-16..................................................................................................... 19 Figure 15. Major Source of Household Income among Ethnic Minorities, 2010-16.................................................. 19 Figure 16. Decomposition of Poverty Changes by Major Income Source, 2013-17................................................... 19 Figure 17. Growth in Manufacturing Output and Employment, 2007 -2016............................................................... 20 Figure 18. Net Employment Creation by Sector: 2010-2016.............................................................................................. 20 Figure 19. Distribution of Households by Livelihood, 2016................................................................................................ 25 Figure 20. Composition of Wage Income by Poverty Status, 2010-16............................................................................ 25 Figure 21. Employment Profile by Education Level, 2014................................................................................................... 26 Figure 22. Trends in Returns to Education, 2011-14.............................................................................................................. 26 Figure 23. Profitability per Hectare by Crop Type and Topography, 2016..................................................................... 27 Figure 24. Land Use Choice by Crop Type, Topography and Poverty Status, 2016..................................................... 27 Figure 25. Share of Plots With Land User Certificates, 2016................................................................................................ 28 CLIMBING THE LADDER: VII POVERTY REDUCTION AND SHARED PROSPERITY IN VIETNAM Figure 26. Use of agricultural land as collateral, 2016........................................................................................................... 28 Figure 27. Net Job Creation by Firm Ownership and Manufacturing Subsector, 2014-2016................................. 30 Figure 28. Labor Productivity Growth, 2011-16...................................................................................................................... 30 Figure 29. Trends in Infrastructure Demand in Vietnam, 2008-......................................................................................... 31 Figure 30. Net School Enrollment by Welfare Status, 2016................................................................................................. 33 Figure 31. Net Enrollment by Ethnicity and Age, 2016......................................................................................................... 33 Figure 32. Per Pupil Spending on Tutoring and Study Support by Welfare Quintile and Ethnicity 2016........... 33 Figure 33. Trends in Poverty Gap and Poverty Gap Squared in Vietnam, 2010-16..................................................... 38 Figure 34. Major Source of Income in Urban Areas, 2010-16............................................................................................. 38 Figure 35. Major Sources of Income in Rural Areas, 2010 vs 2016.................................................................................... 39 Figure 36. Major Source of Income by Poverty Status 2016............................................................................................... 39 Figure 37. Average Annual Wage Per Worker, 2016 (VDN’ 000)......................................................................................... 39 Figure 38. Profitability by Crop Type and Poverty Status, 2016......................................................................................... 39 CLIMBING THE LADDER: 1 POVERTY REDUCTION AND SHARED PROSPERITY IN VIETNAM EXECUTIVE SUMMARY Vietnamese households are escaping poverty in middle-class. They have a high enough income to large numbers, and recent gains appear sustainable. cover their day-to-day necessities, absorb income shocks, and still have enough left for additional Poverty measured at the GSO-World Bank discretionary spending. These income classes are national poverty line1 declined by almost 4 growing rapidly, rising by over 20 percentage points percentage points since 2014, to 9.8 percent in between 2010 and 2017. An average of 1.5 million 2016. Notably, poverty among ethnic minorities Vietnamese joined the global middle class each year declined by 13-percentage points, representing the since 2014, confirming that households continue to largest drop in poverty among ethnic minorities climb the economic ladder after escaping poverty. in the past decade. Moreover, only 2 percent of The rise of the consumer class changes society’s individuals who were not poor in 2014 had fallen into aspirations and the focus of the poverty and shared poverty in 2016, suggesting that those who escape prosperity agenda shifts from combatting extreme poverty tend to remain out of poverty. Significant poverty to effecting broad improvements in the progress was observed in non-income dimensions, quality of life and supporting the further expansion from increases in enrollment in early childhood of the middle class. As these changes continue, education and post-secondary education to access consumption will become increasingly crucial to to improved water and sanitation. Overall, progress economic growth. in all indicators reflected significant improvements for everyone. Poverty reduction owed much Rapid job creation and an ongoing transition to to high average growth and not redistribution. wage employment are driving gains in poverty Though annual consumption per capita for the reduction and shared prosperity bottom 40 percent was very high (5.9 percent), it was 0.8 percentage points below growth in average A booming export sector and rising domestic consumption per capita. demand from the emerging consumer class helped create more than 3 million jobs between A sizeable economically secure class has emerged 2014 and 2016. Nearly 80 percent of these jobs and is growing were created in the manufacturing (50 percent), construction, retail and hospitality sectors, absorbing About 70 percent of Vietnam’s population can now a net outflow of 2 million workers out of agriculture. be classified as economically secure 2, including This marks a turning point in Vietnam’s structural the 13 percent who are now part of the global transformation, as employment in agriculture shrunk in absolute terms too, accompanied by rapid growth in wage employment in all sectors, including agriculture. Robust labor demand over this period 1  The GSO-World Bank poverty line in 2016 is VND 969,167 per boosted average monthly wages in the private sector person per month, equivalent to US$3.34 per day in 2011 by a cumulative 14 percent. Households in Viet Nam purchasing-power-parity (PPP) are therefore increasingly wage dependent. About 54 2  Economic security is defined as having a daily per capita percent obtained most of their income from wages in consumption of at least 2011 PPP $5.5, while the global middle 2016. Also, two in five people now have a paid job. The class is defined as having a daily per capita consumption of at least 2011 PPP $15. In Vietnam, these groups of households rise in wage incomes contributed to more than half of have less than 0.5 percent chance of falling back into poverty. the decline in poverty during 2014-16 and 40 percent Among those classified as the middle class, 75 percent of the increase in the share of people attaining have a washing machine, 98 percent live in a house build economic security. with concrete or bricks of which 55 percent have a private bathroom and kitchen. The average living area of 121m2 and 60 percent of the adults have post-secondary education. 2 CLIMBING THE LADDER: POVERTY REDUCTION AND SHARED PROSPERITY IN VIETNAM Agriculture transformation still has considerable study support, as well as the low quality of schools power to reduce poverty in poor communities. This hints at reduced inter- generational mobility. Vietnam’s poorest households are concentrated in highlands and mountainous areas less Vietnam’s accelerating structural transformation known for agriculture dynamism, but there is has changed the economic landscape, and with it still untapped agriculture potential there. Sub- the poverty and shared prosperity agenda. optimal land use and cropping decisions account for The country pursued an export oriented model a larger share of the difference in agriculture income that successfully generated jobs. Now most between poor and nonpoor households than does households, both poor and non-poor, have a wage the local topography. Across lowland and highland income. The poverty and shared prosperity agenda areas, poor and nonpoor households cultivate is no longer about just moving people to wage jobs, similar amounts of land. However, they devote less but increasingly more about creating better wage of their land to more profitable industrial crops such jobs. It is much less about extreme poverty, and as coffee, black pepper or rubber, and devote more more about attaining economic security. At the same of their land on the less profitable rice or maize time, some old challenges persist that need to be production instead. This could be a result of lower addressed, like the gaps between ethnic minorities access to credit due to lacking land user certificates and the Kinh and Hoa. (“red books”), lower financial literacy and borrowing capacity and low technical skills. The poor have The analysis presented in this report reveals lower values of collateralized fixed assets and banks three strategic priorities for advancing poverty rarely make use of the lower collateral requirements reduction and shared prosperity in Vietnam. targeting such segments. Financial institutions also have a strong preference for collateralized loans with Boosting labor productivity and investing 1.  land. This limits the poor’s access to finance needed in infrastructure to sustain job creation to invest in perennial crops. Lower professional and and wage growth without losing management skills of farmers are evident in that competitiveness. With wages becoming central poor households tend be less productive when to households’ livelihoods, sustained welfare cultivating the same crops on the same types of land improvement hinges on future wage income as nonpoor households. growth and creation of better jobs. But recently wage growth outpaced labor productivity Existing inequalities persist growth. To boost labor productivity and sustain higher wage incomes, Vietnam needs to move Inequalities in opportunities entrench existing production up the value-chain and promote gaps between groups. While welfare has improved investment into higher productivity sectors to across the board, inequalities between groups are shift labor inflows into these sectors. This could not closing fast enough. Close to 45 percent of be achieved by: ethnic minorities still live in poverty. Thus, ethnic minorities who make up only 15 percent of the a. Attracting FDI into higher value agriculture, country’s population, constituted 73 percent of the manufacturing and services activities, poor in 2016. Their average per capita consumption while linking domestic SMEs to multi- was still less than 45 percent of the Kinh and Hoa. national corporations though information Gaps between the remaining poor and the non-poor exchanges, skills upgrade and technology in terms of access to upper secondary education transfer. and improved water and sanitation widened. With b. Keeping up investments in infrastructure a wage premium over secondary education ranging so that supply of transport, electricity, from 43 percent for professional trade training to 63 logistics and telecommunications keeps percent for a university degree, the poor and ethnic up with the high demand from a fast- minorities’ lower tertiary education attainment growing export sector and provide an excludes them from the most rewarding jobs. enabling environment for the country to Lower rates of educational enrollment are partially move up the value chain or into high value explained by household poverty itself, as lower- added sectors. income families are less able to invest in tutoring and CLIMBING THE LADDER: 3 POVERTY REDUCTION AND SHARED PROSPERITY IN VIETNAM 2.  mplementing education reforms designed I Spurring 3.  agriculture structural to equalize opportunities and develop transformation through changing farmland workforce skills. Rising private sector use patterns, strengthening land user wages in the face of abundant supply of rights, and improving skills of poor farmers. labor suggest that firms are competing for Addressing sub-optimal farmland use patterns a limited pool of competent laborers. A is key to unlocking the agriculture potential of significant share of hiring employers say that the poor by aligning land use with comparative job applicants lack the skills needed for the advantages of specific areas and farm income job, even for low-skilled jobs. Investments in generation objectives as outlined in the skills development would increase the pool of Agricultural Restructuring Plan. This requires a competent workers, facilitate the expansion of bolder shift of land from rice and maize to more value chains into more sophisticated activities profitable annual and perennial crops. Necessary and support the growth of new sectors. for achieving this is: Expanding access to high quality education Strengthening land user rights through a.  across groups will be vital not only to develop completing land titling to facilitate skills, but to reduce inequalities and increase household access to credit (using land as access to better paying jobs for all. Tutoring collateral) and enabling the poor to invest and differences between quality of schools in more profitable crops that require costly in poor and non-poor communities explain initial investments, intermediate inputs or the variation in academic achievement at hiring of labor. lower secondary level, which determines progression to tertiary education. This signals Improving farm management and b.  that inadequate teaching places poor children business skills of the poorer farmers, at a disadvantage. Necessary reforms include: often neglected by public extension and investment programs, is required to help Increasing instruction hours in schools to a.  boost agricultural productivity, thereby close the achievement gap that better-off reducing the productivity gap with less- households currently close through study poor farmers. support payments, but poor households cannot afford. b. Reforming the curricula to develop problem solving and critical thinking skills that most employers find lacking among workers. 4 CLIMBING THE LADDER: POVERTY REDUCTION AND SHARED PROSPERITY IN VIETNAM INTRODUCTION Vietnam has achieved remarkable success in of VND 969 167 per person i.e. equivalent to reducing poverty while controlling inequality. 2011 PPP $3.34 per person per day. This poverty The country’s broad-based growth reflects the line was determined in 2010 and has only been government’s focus on developing labor-intensive updated for changes in the cost of living since export sectors while investing heavily in human then. The Vietnamese government also uses a capital that saw the country exceed its peers multidimensional poverty line, which classifies (World Bank, 2016a). However, gains have been households as poor if they have an income per concentrated among the Kinh and Hoa ethnic capita of VND 900,000 in urban areas or VND 700,000 majority, while minority groups have not only in rural areas and are “deprived” in at least three of continued to experience poverty rates far above the 10 dimensions of nonmonetary poverty. Because the national average, but have seen slower progress too. thresholds used to determine the multidimensional During 2012-14, poverty among ethnic minorities poverty line are not comparable over time, the report declined by less than 2 percentage points, leaving uses the GSO-World Bank approach to assess long- close to 58 percent of ethnic minorities still living in term poverty trends. But the report also provides poverty. The country’s poverty agenda became more an update on complimentary non-monetary centered on issues of social exclusion. Recognizing dimensions of poverty too. this, the government implemented multiple programs aimed at reducing poverty among lagging The report is organized into two main sections. communities by more than 1.5 percentage points The first section reviews Vietnam’s progress in per year. reducing poverty and promoting share prosperity. It describes updated poverty and shared prosperity This report analyzes recent trends in poverty trends, the nature of economic mobility, and the and shared prosperity. It presents the findings of drivers of poverty reduction. The second section the 2016 Vietnam Household and Living Standards – titled leaving no one behind - is more forward- Survey (VHLSS), highlighting important progress looking, starting by identifying major constraints and identifying new challenges. The report defines faced by the poor, then proceeding to lay out monetary poverty according the GSO-World Bank challenges for moving the poverty and shared poverty line, amounting to monthly consumption prosperity agenda going forward. VIETNAM’S REMARKABLE SUCCESS STORY CONTINUES 6 CLIMBING THE LADDER: POVERTY REDUCTION AND SHARED PROSPERITY IN VIETNAM POVERTY AND SHARED PROSPERITY TRENDS: 2010-16 Poverty declined everywhere for both ethnic then plunged by more than 13 percentage points minorities and the majority from 57.8 percent to 44.6 percent between 2014 and 2016 (Figure 2). The decline in poverty among ethnic All measures of poverty reveal broad and minorities during 2014-16 is the steepest decline consistent (Figure 1). Measured at the GSO-World in poverty among them for the past two decades. It Bank national poverty line, the poverty headcount also marks the first instance when declining poverty rate fell from 20.8 percent in 2010 to 9.8 percent in among ethnic minorities drove the overall decline in 2016, having declined by nearly 4 percentage points poverty rates nationwide. in the two years between 2014 and 2016. Meanwhile, estimates based on the poverty line for lower middle Poverty declined everywhere. Between 2014 and income countries (at 2011 PPP $3.2 per person per 2016, poverty rates declined across all regions of Figure 1. Poverty Rates Based on National and Figure 2. Poverty Rates by Ethnic Group, 2010-16 International Poverty Lines, 2010-16 30.0 66.3 27.0 25.0 59.2 57.8 22.1 Poverty headcount rate (%) 20.0 20.7 18.6 44.6 17.2 15.0 13.5 13.6 10.0 9.8 6.0 5.4 12.9 5.0 9.9 3.8 6.3 1.6 3.1 0.0 2010 2012 2014 2016 1 2 3 4 Urban Rural Vietnam Ethnic Minorities Kinh & Hoa Source: Authors’ calculations based on the 2010-16 VHLSS. day), show a lower poverty rate of 8.6 percent in Vietnam, though to a varying degree. The Midlands 2016. Extreme poverty is all but eliminated, with only and Northern Mountains region and the Central 2 percent of the population living on less than 2011 Highlands region experienced the largest drops, with PPP $1.9 per day. The poverty gap, which measures regional poverty rates falling by 9.3 and 6.3 percentage how far consumption of the poor is below the points, respectively (Table 1). The gains in the Central poverty line, has also steadily declined, indicating Highlands were especially noteworthy, as the regional that poverty is becoming less severe among the poverty rate had barely declined during 2010-14. remaining poor (see Annex 1, Figure 33 ). This leaves Remarkably, there is no indication that the pace of no doubt that Vietnam’s remarkable success in poverty reduction is slowing in regions where poverty reducing poverty continues. rates are already low. Poverty more than halved, and one can say it was almost eliminated, in both the Red Minority groups have experienced an especially River Delta and Southeast regions which both had a significant decline in poverty rates in recent years. 3-percentage point decline in poverty during 2014-16. The poverty rate among minority groups fell by just Progress in these regions drove the decline in urban 1.4 percentage points between 2012 and 2014, but poverty nationwide. SECTION I 7 VIETNAM’S REMARKABLE SUCCESS STORY CONTINUES Table 1: Poverty trends by region, 2010-16 Poverty Headcount Rate Distribution of the Poor 2010 2012 2014 2016 Change 2010 2012 2014 2016 Vietnam 20.7 17.2 13.5 9.8 -3.8 100.0 100.0 100.0 100.0 Rural 27.0 22.1 18.6 13.6 -5.0 91.4 90.6 90.6 94.7 Urban 6.0 5.4 3.8 1.6 -2.1 8.6 9.4 9.4 5.3 Regions Red River Delta 11.9 7.5 5.2 2.2 -3.0 13.7 9.9 9.0 5.2 Midlands and Northern Mountains 44.9 41.9 37.3 28.0 -9.3 28.6 33.4 35.6 40.2 Northern and Coastal Central 23.7 18.2 14.7 11.8 -2.9 25.9 23.7 23.3 26.7 Central Highlands 32.8 29.7 30.4 24.1 -6.3 9.5 10.0 13.7 16.2 Southeast 7.0 5.0 3.7 0.6 -3.1 5.2 4.7 4.6 1.0 Mekong Delta 18.7 16.2 9.8 5.9 -3.9 17.1 18.4 13.7 10.8 Source: Authors’ calculations from VHLSS 2010, 2012, 2014. The spatial pattern of poverty reduction during Figure 3. Regional Decomposition of Poverty 2014-16 was more balanced than in previous years, Changes in Vietnam, 2012-16 as gains in the poorest regions contributed the 120 most to poverty reduction. Decomposing changes 110 in the national poverty rates to account for changes Mekong Delta in poverty within regions and the effect of migration 100 between regions (Ravallion and Huppi, 1991) 90 reveals that reduction in poverty within the Mekong Southeast Contribution to poverty change (%) Delta and the Northern and Central Coastal areas 80 accounted for 55 percent of the reduction in poverty 70 Central Highlands during 2012-14, though these regions accounted for 40 percent of the population (Figure 3). During 60 Northern and Coastal Central 2014-16, declining poverty within the Midlands and 50 Northern Mountain region and the Central Highlands Midlands and Northern 40 Mountains region contributed 42 percent to the total reduction in poverty, even though these regions are home 30 Red River Delta to just 20 percent of the population. Over the same 20 period, the contribution of the Mekong Delta and Interaction e ect the Northern and Central coastal areas declined to 10 36 percent. Falling poverty rates in Vietnam’s poorest Population-shift e ect 0 regions have driven the overall decline in the national 2014-2016 2012-2014 rate observed in recent years. Source: Authors’ calculations from VHLSS 2012, 2014, 2016. Poverty reduction reflects significant the change in the distribution of the consumption improvement at all levels of welfare aggregate in real terms between 2014 and 2016 (Figure 4). The rightward shift in the distribution The observed decline in poverty rates reflects means that in 2016, Vietnamese at all levels of robust and broad-based income growth among welfare were spending more than in previous households at all income levels. Vietnam’s years, which helps to explain the decrease in annual per capita GDP growth rate averaged 6.4 poverty. Due to strong and consistent gains in percent during 2014-16. Per capita consumption consumption across welfare groups, the observed growth was equally high, rising by an average of shift in the distribution also implies that poverty 6.7 percent per year. Consumption growth was rates declined over this period, regardless of which robust across all welfare levels, as evidenced by poverty line is used. 8 CLIMBING THE LADDER: POVERTY REDUCTION AND SHARED PROSPERITY IN VIETNAM Figure 4. Poverty Dominance Curves, 2014-16 Figure 5. Per Capita Consumption Growth, 2010-16 100 6.7 6.2 90 5.9 5.6 Cumulative share of the population 80 4.8 70 4.2 60 2010 3.8 50 2014 40 2016 2.2 30 20 10 0 2010-12 2012-14 2014-16 2010-16 10,000 12,500 15,000 17,500 20,000 22,500 25,000 27,500 30,000 32,500 35,000 37,500 40,000 42,500 45,000 47,500 50,000 52,500 55,000 57,500 60,000 62,500 65,000 67,500 70,000 72,500 75,000 77,500 80,000 2,500 5,000 7,500 Annual per capita consumption growth (%) - BOTTOM 40 percent Annual consumption per capita in 2010 prices (VND '000) Annual per capita consumption growth (%) - VIETNAM Source: Authors’ calculations from VHLSS 2010-16 Source: Authors’ calculations from VHLSS 2010-16 But consumption growth is becoming less pro- Inequality now appears to be increasing, poor over time especially in the Central Highlands and the Mekong Delta. The Gini coefficient dropped from 39.3 in 2010 The rate of per capita consumption growth to 34.8 in 2014, but then rebounded to 35.3 in 2016 among households in the bottom 40 percent (Table 2). The increase in inequality occurred entirely has recently fallen below the national average. in rural areas, where the Gini coefficient increased by At nearly 6 percent per year, growth in per capita 0.8 points, while no change in inequality was observed consumption among households in the bottom in urban areas. Other measures of inequality, such as 40 percent was high, but still 0.8 percentage the Theil index, confirm an incipient rise in inequality, points below the national average and nearly a primarily in rural areas. The Central Highlands and the full percentage point below the growth rate for Mekong Delta regions both exhibited a significant households in the top 60 percent. However, due increase in inequality, with Gini coefficients rising to a highly pro-poor distribution in previous by as much as 2 Gini percentage points. Increasing years, consumption growth among lower-income inequality in these two regions and in the Northern households still outpaced the national average and Central Coastal region was responsible or the over the 2010-16 period (Figure 5). incipient rise in inequality. Table 2: Trends in Inequality, 2010-16 Gini Coefficient Theil Index 2010 2012 2014 2016 2010 2012 2014 2016 National 39.3 35.6 34.8 35.3 29.4 22.9 21.6 22.3 Urban 38.6 31.7 33.1 32.9 27.8 21.4 19.7 19.5 Rural 33.2 34.4 31.0 31.8 20.0 17.4 16.5 17.7 Region Red River Delta 40.1 34.4 33.6 32.8 29.7 20.9 20.3 19.3 Midlands and Northern Mountains 37.1 36.6 37.0 36.4 23.9 23.4 25.0 23.9 Northern and Coastal Central 34.0 33.3 33.2 33.9 20.9 19.6 19.6 20.9 Central Highlands 36.7 37.9 38.9 39.7 23.0 25.2 26.3 27.3 Southeast 39.8 33.3 31.1 30.9 31.6 20.5 18.0 17.4 Mekong Delta 31.7 30.3 28.7 30.6 17.8 17.6 14.5 17.3 Source: Authors’ calculations from VHLSS 2010, 2012, 2014, 2016 SECTION I 9 VIETNAM’S REMARKABLE SUCCESS STORY CONTINUES Growth, rather than redistribution, drove the Figure 6. The Relative Contributions of recent decline in poverty Consumption Growth and Distributional Equity to Changes in the Poverty Rate, 2014-16 As inequality rose during 2014-16, poverty reduction was a result of high growth of average 2.00 consumption rather than changes in the distribution of consumption. A growth-inequality 1.00 decomposition3 is used to show how much of the 0.00 reduction in poverty was a result of high growth Percentage pointv change in Poverty in average per capita consumption, assuming -1.00 no changes in inequality, and how much can be -2.00 attributed to changes in inequality keeping the average level of consumption per capita the same -3.00 as the base year. Estimates from the VHLSS 2014 -4.00 and 2016, suggest that growth was the sole driver of poverty reduction (Figure 6), while rising inequality -5.00 dampened poverty reduction. If inequality had -6.00 not changed, the overall poverty rate would have declined by an additional 1.1 percentage points, -7.00 Total Urban Rural and the rural poverty rate would have fallen by an additional 1.5 percentage points. Growth Redistribution Interaction Non-monetary indicators of welfare show Source: Authors’ calculations from VHLSS 2014, 2016. improvements but gaps persist is also on the rise. Two-thirds of three- to five-year-old Continued improvements in numerous children were enrolled in pre-primary education in nonmonetary welfare indicators underscore 2016, up more than 13 percentage points from 2010. Vietnam’s progress in improving the quality of Vietnam’s performance in these areas far exceeds the life. Electricity access has been near-universal for average for comparable countries in the region. some time now. In 2010, an estimated 98 percent of households was connected to the national electric As with consumption growth, gains in grid. Improvements have been made on other nonmonetary welfare indicators were in most indicators since 2010 Mobile connectivity is now cases, experienced by poor and nonpoor ubiquitous, as the share of households with at least households across all geographical regions. ECD one mobile phone rose from 73 percent in 2010 to enrollment increased by 15 percentage points among 93 percent in 2016. Access to improved sanitation children in the bottom 40 percent of households and increased from 70 percent to 83 percent, and access among rural children. Enrollment in ECD increased to an improved drinking water rose from 72 percent in all regions, ranging from 10 percentage points to 78 percent. Meanwhile, the share of households in the Southeast region to 18 percentage points in with an indoor piped water connection rose from 26 the Midlands and Northern Mountainous region. percent to 38 percent (Table 3). Access to improved sanitation increased by about 10 percentage points among households in the bottom Vietnam performed well on education and it 40 percent, broadly in line with the increase among continues to improve Upper secondary completion households in the top 60 percent. And access to rates increased, evidenced by the rise in the share of phone service rose by 16 percentage points among 20- to 24-year-olds with complete upper secondary households in the bottom 40 percent. education from 52 percent in 2010 to 59 percent in 2016. Enrollment in early childhood education (ECD) Households in the Midlands and Northern Mountainous regions, however did not make as much progress on upper secondary education 3  or a description of the growth-inequality decomposition F and access to water. Upper secondary completion methodology, see: Datt and Ravallion, 1992. rates in the region remained broadly unchanged at 43 percent between 2010 and 2016. However, 10 CLIMBING THE LADDER: POVERTY REDUCTION AND SHARED PROSPERITY IN VIETNAM Table 3: Trends in Nonmonetary Welfare Indicators, 2010-16 ECD Tertiary Piped Improved Improved Telephone Completion Water Water toilet 2010 2016 2010 2016 2010 2016 2010 2016 2010 2016 2010 2016 Male 53.6 66.1 47.7 50.9 22.4 35.1 69.6 76.2 68.7 82.5 79.3 94.6 Female 52.9 67.4 56.0 66.6 37.0 47.2 78.2 83.6 74.1 85.4 75.0 87.7 Ethnic Minorities 44.6 64.5 26.2 31.5 5.7 10.8 43.4 46.9 23.1 45.3 51.0 85.0 Kinh & Hoa 55.2 67.3 56.8 66.0 29.2 42.6 76.0 83.2 77.0 89.4 81.9 94.0 Not Poor 59.2 69.1 59.0 64.2 30.4 40.8 76.4 81.3 78.1 87.7 84.1 94.4 Poor 38.7 53.2 18.7 16.2 6.7 7.1 50.9 40.2 33.5 30.5 49.0 72.6 Top 60 percent 62.2 74.4 64.1 71.9 35.7 49.3 79.6 86.1 84.0 93.4 87.6 96.6 Bottom 40 percent 42.4 57.9 24.9 33.4 9.4 18.2 58.2 63.8 45.6 65.0 60.8 86.0 Rural 49.0 64.2 44.8 53.2 8.7 20.5 63.4 69.9 60.4 77.0 74.4 90.9 Urban 63.8 72.7 69.3 71.4 66.2 75.8 91.2 95.6 92.2 96.4 86.8 96.7 Red River Delta 68.2 81.5 72.8 80.7 27.6 46.6 61.4 73.1 87.4 97.8 81.1 91.6 Midlands and Northern 60.8 78.0 43.3 43.2 13.1 15.3 58.8 60.4 52.2 69.5 69.7 93.4 Mountains Northern and Coastal Central 54.5 66.6 55.3 63.0 21.1 31.1 79.5 81.4 72.1 83.4 75.3 90.1 Central Highlands 38.7 52.3 41.2 49.0 12.8 18.0 78.6 81.9 51.2 66.6 78.8 89.4 Southeast 54.7 64.7 51.4 57.8 43.8 56.0 93.7 97.6 90.2 95.7 84.8 97.4 Mekong Delta 33.0 44.1 29.9 43.0 27.3 42.8 64.2 74.9 44.5 67.8 77.1 94.2 Vietnam 53.3 66.7 51.9 58.8 26.2 38.2 71.9 78.1 70.1 83.3 78.2 92.8 Source: Authors’ calculations from VHLSS 2010, 2012, 2014, 2016. completion rates may rise in the near term, since and nonpoor households widened by 13 percentage upper secondary enrollment has increased by 5 points, and by 2016 access rates among nonpoor percentage points. The share of households with a households were nearly three times higher than piped water connection or access to an improved access rates among poor households. water source rose by less than 2 percentage points during 2010-16, with similarly limited progress Gender gaps are emerging in upper secondary observed among households above and below the education as female students outperform their poverty line. male peers. Enrollment rates for girls and boys are broadly equal from the ECD to lower secondary While access to upper secondary education levels and have increased at a similar pace. In 2010, and water and sanitation improved for both net enrollment rates in upper secondary education poor and nonpoor households, gaps between were also equal at about 34 percent, but by 2016 the them widened over time The disparity in upper rate for female students had risen to 43.5 percent, secondary enrollment rates between children living while the rate for male students had reached just 39.6 in poor and nonpoor households widened to more percent. Similarly, the upper secondary completion than to 24 percentage points in 2016. Meanwhile, rate among females aged 20-24 was 67 percent in the gap in access to improved toilets between poor 2016, compared to 51 percent for male students. SECTION I 11 VIETNAM’S REMARKABLE SUCCESS STORY CONTINUES Overall, all population subgroups are Figure 7. Trends in Annual Consumption per experiencing substantial gains, but welfare Capita, 2010-16 indicators for some, especially minorities, 30000 continue to lag far behind. Both monetary and nonmonetary indicators are improving among 25000 Consumptio per capita in 2010 Prices ('000) ethnic minorities, but not fast enough to catch up with the Kinh and Hoa (Figure 7). Average per 20000 capita consumption of ethnic minorities was only 41 percent of the average per capita consumption 15000 of the Kinh and Hoa in 2010 and was still less that 45 percent in 2016. As the economy grows, the 10000 absolute gap between ethnic minorities and the Kinh and Hoa has increased. Poverty is also significantly 5000 deeper among poor ethnic minority households than among poor Kinh and Hoa households. Similar 0 disparities are evident in education, and gaps at the 2010 2012 2014 2016 upper secondary level in 2016 mirrored gaps at the All Ethnic Minorities Poor Ethnic Minorities lower secondary level a decade earlier. Thus, even as All Kinh & Hoa Poor Kinh & Hoa society progresses, those at the bottom remain there. Despite recent progress, targeted measures will be Source: Authors calculations from VHLSS, 2010, 2012, 2014, 2016 necessary to ensure that poverty rates among ethnic minorities converge with the national average. 12 CLIMBING THE LADDER: POVERTY REDUCTION AND SHARED PROSPERITY IN VIETNAM ECONOMIC MOBILITY: CHASING THE MIDDLE CLASS DREAM Panel data reveals the progress of Vietnamese Living conditions or life style choices of households as they move up the economic Vietnamese we classify as part of the global ladder.4 To demonstrate this progress, households middle class, are much better and represent the are grouped into five economic classes based on dream of many. In 2016, about 55 percent of their daily consumption per capita in 2011 PPP middle-class households lived in a house with a dollars. Following international norms5, these five private bathroom and kitchen (Table 4). Nearly all classes are defined as: (i) the extremely poor, who middle-class houses had concrete or brick walls live on less than $1.90 per day, (ii) the moderately and used septic/semi-septic tanks. Their average poor, whose per capita consumption ranges living area was 120 m2. About 75 percent of the from $1.90 to $3.20 per day, (iii) the economically middle class had a washing machine, 58 percent vulnerable, who consume $3.20 - $5.5 per person had an air conditioner, and 57 percent had a per day, (iv) the economically secure, consuming computer. By comparison, just 5 percent of the $5.50 - $15 per person per day, and (v) the global economically vulnerable lived in a house with a middle class, who live on more than $15 per person private bathroom or kitchen, 47 percent had a per day. Households in the last two groups are septic tank, and their average living space was 64 referred to as the “consumer class,” since they have m2. Fewer than 8 percent had a washing machine, enough income to cover daily expenses, absorb fewer than 2 percent had an air conditioner, and income shocks, and consume some amount of non- fewer than 3 percent had a computer. On average, necessity goods and services. Economic mobility the middle class spent three times as much on can be analyzed by examining the movement nonfood items (excluding durable goods) than the of households between these groups, as well as economically vulnerable, Moreover, 60 percent movements in and out of poverty. of adults in middle-class households had a post- secondary education, twice national average. Though based on international norms, these groups resonate in Vietnam. Households in the first Figure 8. Probability of being poor in 2016 two groups are classified as poor based on the GSO- conditional on economic class in 2014 World Bank national poverty line. Panel data for 2014- 16 show that the estimated risk of a household in the consumer class falling into poverty is almost zero Probability of being poor in 2016 79.6 (Figure 8). Thus $5.5 per capita per day is a reasonable threshold for defining economic security in Vietnam. 41.6 4 The VHLSS includes a rolling panel of households, in which 50% of the households in one round (e.g. VHLSS 2014) are revisited as part of the sample for the next (the VHLSS 2016 7.0 in this case). This survey is a rooftop survey, without any track- 0.5 ing of individuals or split households. This design may lead to 0.0 attrition bias, depending on migration patterns. We test for Economic Middle Economically Economically Moderately Extreme the existence of such bias by looking at key outcomes for the Class in 2014 class secure vulnerable Poor Poor panel sample against the full sample, and find no evidence of 2014-2016 attrition bias. 5 These thresholds are based on international income classifica- Source: Authors’ calculations from VHLSS 2014, 2016. tions. See: World Bank, 2017a. SECTION I 13 VIETNAM’S REMARKABLE SUCCESS STORY CONTINUES Table 4: Living Conditions Indicators by Economic Class, 2016 Economic Class Extreme Moderately Economically Economically Middle Vietnam Poor Poor vulnerable secure class Living area (m2) 51 57 64 84 121 84 Piped water or borehole 8.3 20.3 41.2 68.1 85.7 62.3 Septic/Semi-Septic Tank 4.2 14.1 46.5 78.9 95.7 70.8 Concrete or brick wall house 30.8 43.0 70.8 89.9 97.8 84.2 Lives in villa or a house with private 0.0 1.4 5.2 23.0 55.4 23.2 bathroom and kitchen Has a computer 0.0 0.8 2.5 21.8 56.0 22.1 Has air conditioners 0.0 0.4 1.6 15.8 58.6 18.8 Has a washing machine 0.0 0.7 7.5 38.1 74.7 35.5 Has a water heater 0.0 0.2 7.9 29.6 57.6 27.9 Share of adults with post- 3.9 6.9 14.6 30.2 60.4 30.3 secondary education Nonfood expenses (VND '000) 1,267 2,230 3,831 7,519 17,196 7,967 Source: Authors’ calculations from VHLSS 2016. The middle class is expanding as households class is “possible” in this classification. About 65 move up the economic ladder percent of households that were either extremely poor or moderately poor in 2014 had moved up Vietnam’s consumer class is growing rapidly. the economic ladder by 2016. Among households The share of households classified as economically that were economically vulnerable, 52 percent secure increased from less than 50 percent in 2010 had become economically secure or joined the to 70 percent in 2016. These include the 13.3 percent middle class by 2016. Meanwhile, not more than 9 of households that are part of the global middle percent fell to a lower economic class among each class (Figure 9), up from 7.7 percent in 2010. Most of these groups. These figures clearly indicate that of that increase occurred between 2014 and 2016, households’ welfare is on an upward trajectory, when 3 million people joined the global middle consistent with the robust consumption growth class. More importantly, the population classified as observed in recent years. economically insecure or poor is rapidly declining Figure 9. Population by Economic Class, 2010-16 from half the population in 2010 to 30 percent in 2016. Not only are households managing to 100 7. 7 7. 9 9.6 escape poverty, they can quickly progress out of 90 13.3 the economic insecurity to a place in the consumer 80 class. The shrinking share of economically vulnerable Share of the population 41. 5 households suggests that economic security is 70 47. 6 within reach of most of the population. 60 54.2 57. 0 50 Economic mobility among the same households over time provides robust evidence of upward 40 32. 0 movement in Vietnam (Figure 10). About 28 30 28. 9 percent of the population moved into a higher 24.4 20 21.1 economic class between 2014 and 2016, 63 percent stayed in the same class, while only 9 percent fell 10 into a lower economic class. Thus, three times as 0 many people moved up the economic ladder than 2010 2012 2014 2016 those who moved down. Upward mobility is even Extreme Poor Moderately Poor higher when the top economic class is excluded, Economically vulnerable Economically secure as this class represents the upper bound from Middle class which no further economic movement to a higher Source: Authors’ calculations from VHLSS 2010, 2012, 2014, 2016. 14 CLIMBING THE LADDER: POVERTY REDUCTION AND SHARED PROSPERITY IN VIETNAM Figure 10. Trends in Economic Mobility, 2010-16 2010 – 2012 VHLSS Panel 2014-16 VHLSS Panel Vietnam 13.5 58.5 28.1 Vietnam 8.9 63.2 27.9 Middle class 39.9 60.1 Middle class 26.5 73.5 Economically secure 15.7 76.2 8.1 Economically secure 8.8 77.4 13.8 Economically 10.5 47.3 42.2 Economically 5.1 42.9 52.0 vulnerable vulnerable Moderately Poor 5.9 37.3 56.8 Moderately Poor 4.8 30.3 64.9 41.0 59.0 Extreme Poor Extreme Poor 34.8 65.2 Share of the population (%) Share of the population (%) Sliders Stayers Climbers Sliders Stayers Climbers Source: Authors’ calculations from VHLSS 2010, 2012, 2014, 2016 The risk of falling into poverty is low and declining poverty by 2016 (Table 5). The mass movement of households out of poverty, combined with Table 5: Transitions into and out of Poverty: the small share falling below the poverty line, 2014-16 suggest that fewer Vietnamese households are in 2016 chronic poverty, and those who escape poverty mostly sustain their gains. Poverty eradication 2014 Not Poor Poor Total can now be regarded as a realistic goal, given Not Poor 97.9 2.1 100 how few households who escape poverty end up Poor 49.9 50.1 100 falling back and how close the remaining poor households are to the poverty line Source: Authors’ calculations from VHLSS 2014, 2016 panel households. Over time, the risk of falling into extreme poverty has substantially declined. This trend Table 6: Probability of Being Poor (Based on reflects the shrinking share of economically GSO-WB Poverty Line) in Terminal Period by vulnerable households, as well as the low Economic Status in Base Year: 2010-16 downward mobility of households in each of the top four economic classes. This is made clear from Poor at end of panel period a comparison of transitions between the 2010-12 Economic class in 2010-2012 2012-2014 2014-2016 and 2014-16 panels in Figure 10 above. Further base year evidence comes from comparing movements Extreme Poor 51 57 64 in and out of poverty by economic class in the Moderately Poor 8.3 20.3 41.2 base year using panel data (Table 6). Between Economically 4.2 14.1 46.5 2010 and 2012, just 4 percent of the population vulnerable fell into poverty, and this this share declined Economically secure 30.8 43.0 70.8 to 2 percent between 2014 and 2016. Similarly, Middle class 0.0 1.4 5.2 while 13 percent of households classified as economically vulnerable in 2010 had fallen All non-poor 4.0 4.4 1.6 below the poverty line in 2012, just 7 percent Source: Authors’ calculations from VHLSS 2010, 2012, 2014, 2016. of economically vulnerable households in 2014 had fallen into poverty by 2016. Almost all the In recent years, very few non-poor households households classified as economically secure in have fallen below the poverty line. While half of 2014 remained non-poor in 2016. households that were poor in 2014 had moved out of poverty by 2016, only 2 percent of households Upward mobility has increased across all that were not poor in 2014 had fallen into demographic groups. In the past, gains among SECTION I 15 VIETNAM’S REMARKABLE SUCCESS STORY CONTINUES Table 7: Economic mobility by household characteristics, 2014-16 Transition, 2014-16 Economic Class in 2016 Sliders Stayers Climbers Economically Middle class Secure or middle class Male 9.0 63.1 27.9 72.0 12.6 Female 8.6 63.5 27.9 77.7 21.8 Ethnic Minorities 13.8 49.4 36.8 23.5 2.7 Kinh & Hoa 8.1 65.4 26.5 81.3 16.7 Household head education level Some primary 10.2 54.4 35.4 53.3 6.3 Completed primary 9.2 61.9 29.0 70.1 8.4 Completed lower secondary 10.0 65.3 24.6 76.4 10.7 Completed post-secondary 6.0 70.4 23.6 92.7 35.1 Rural 9.8 59.6 30.7 65.7 7.5 Urban 7.2 70.4 22.4 88.7 29.3 Red River Delta 9.9 67.9 22.2 81.8 18.7 Midlands and Northern Mountains 12.2 55.6 32.2 43.6 6.5 Northern and Coastal Central 7.0 64.9 28.1 70.8 10.5 Central Highlands 11.7 58.5 29.8 58.2 8.5 Southeast 5.9 68.0 26.1 91.0 28.2 Mekong Delta 9.3 57.6 33.1 75.4 9.6 Livelihoods Non-agriculture labor income only 7.1 72.3 20.6 91.0 31.1 Nonwage agriculture only 11.1 54.8 34.2 62.1 6.7 Agriculture wage and nonwage income 11.5 55.8 32.7 36.6 2.3 Agriculture plus non-agriculture wages 8.9 56.6 34.5 63.8 5.8 Agriculture plus household business 9.4 65.1 25.5 78.6 11.9 Agriculture and both non-agriculture wages & 8.6 68.1 23.3 72.4 8.4 household business Remittances & Transfers only 16.6 49.8 33.6 83.1 17.3 Major source of income Crop cultivation 12.6 57.8 29.6 56.6 4.9 Other agriculture 8.4 54.8 36.8 58.4 5.6 Household business 5.3 70.0 24.7 88.6 22.1 Wages 9.6 63.3 27.1 74.1 15.5 Remittances 8.5 62.0 29.6 72.2 11.8 Transfers 13.7 48.0 38.3 62.4 10.9 Other income 1.4 77.7 21.0 79.0 37.7 Source: Authors’ calculations from VHLSS 2014 -2016 panel households’ data. 16 CLIMBING THE LADDER: POVERTY REDUCTION AND SHARED PROSPERITY IN VIETNAM ethnic minorities, farming households, and economically secure in 2016, of which a third of households located in the Midlands and Northern them are classed as the middle class. On the other Mountain region tended to lag the national hand, only 37 percent of people supplementing average. However, between 2014 and 2016, their family farm income with agriculture wages more than 30 percent of households in these could be classified as economically secure. Only demographic groups had moved up the economic 23 percent of ethnic minorities were classified ladder (Table 7). In many cases, upward mobility as economically secure, in contrast to 88 percent among these groups exceeded the national of the Kinh and Hoa. Economic security, seems average. However, they were also subject to an the preserve of the more educated, and mostly increased risk of downward mobility compared urban based population. About 93 percent of to other groups. For example, upward mobility people living in households headed by a person was 10 percentage points higher among ethnic with post-secondary education are classified as minorities than among the Kinh and Hoa, but economically secure, with 35 percent being in downward mobility among ethnic minorities was the global middle class. In urban areas, close to also 5 percentage points higher. The same can be 89 percent of the population are economically said for exclusively family farming households secure, with 29 percent being in the middle class. and those living in Midland and Northern The Southeast region leads the way in economic Mountainous areas. But on balance, more people security (91 percent) while the Midland and from these groups moved up the economic ladder Northern Mountainous regions fall short (just 44 than sliding down. percent). Despite, recent progress in escaping poverty, lagging groups fall short of attaining Some groups were more likely than others to economic security. achieve economic security. Households who made the full transition out of agriculture are more likely to attain economic security (Table 7). About 91 percent of such households are classified as SECTION I 17 VIETNAM’S REMARKABLE SUCCESS STORY CONTINUES VIETNAM’S RECIPE OF SUCCESS Vietnam’s model of export led growth in labor The contribution of labor income and its intensive sectors paid off. Jobs were created subcomponents to poverty reduction and and wage incomes grew. To understand Vietnam’s upward mobility can be quantified using success in raising living standards, a simplified income-poverty decompositions7 (Figure 11). analytical framework for examining changes in Components of labor income analyzed include the households’ income-generating capacity is applied. share of adults (an indicator of the availability of This considers that changes in household income- labor), the proportion of employed adults in the generating capacity are determined by changes in: (i) household, average wages per worker, and average the stock of household labor and other assets such as non-wage earnings per adult from agriculture Figure 11: Framework for Income-Poverty Decompositions Non-labor income per adult (1 n ∑iЄA yi ) n NL a n Income per capita Share of adults ( na ) Consumption per capita Labor Income Share of occupied per adult n adults ( no ) a Number of household members (n) Income per adult Labor income per occupied adult (1 n ∑iЄA yi ) n L a Propensity to consume & measurement error (θn) Source: Adapted from Azevedo et al. 2013. land, (ii) the employment rate of household’s labor, and household businesses. Non-labor income is (iii) the returns to household labor – both wages and separated into remittances, social transfers and profits from agriculture or household businesses, other income. Table 8 below presents estimates of and (iv) transfers received by the household.6 A the contribution of these factors to poverty changes combination of the first three makes up labor based on two poverty lines corresponding to the income and reflect human capital accumulation and GSO-WB poverty line and the 2011 PPP $5.5 per day labor market dynamics, while the fourth represents poverty line denoting the threshold for economic nonlabor income. This is in part influenced by security. Results for the global middle class line are government taxes and government transfers. shown in Annex 1, Table 12. 6 This framework is based on Busolo et al., 2014. This methodology was proposed by Barros et al. (2006) and 7  adapted by Azevedo et al. (2013). 18 CLIMBING THE LADDER: POVERTY REDUCTION AND SHARED PROSPERITY IN VIETNAM Table 8: Income Decomposition of Poverty in Vietnam, 2014-16 GSO-World Bank National Poverty Line Economic security line Vietnam Rural Urban Ethnic Kinh & Vietnam Rural Urban Ethnic Kinh & Minorities Hoa Minorities Hoa Propensity to consume -0.8 -1.7 0.7 -2.9 -0.7 -1.8 -3.5 0.6 -1.2 -1.7 Share of Adults 0.2 0.3 0.4 -0.3 0.3 -0.2 -0.4 0.1 0.8 -0.4 Share of wage workers 0.2 0.2 -0.6 -0.8 0.1 -0.3 -0.7 -1.0 -0.2 -0.8 Average wages per worker -1.8 -2.2 -1.3 -3.9 -1.7 -3.3 -3.2 -3.4 -3.1 -3.7 Average business earnings -0.4 -0.5 -0.7 -0.9 -0.5 -1.4 -1.3 -1.9 -1.1 -1.7 Average crop income 0.4 0.6 0.2 0.0 0.5 0.0 0.2 0.1 0.4 -0.1 Average other farm income -0.5 -0.8 0.1 -1.8 -0.3 -0.6 -0.9 0.1 -1.3 -0.8 Remittances -0.8 -0.8 -0.7 -1.5 -0.7 -1.0 -1.1 -0.7 -1.0 -1.0 Transfers -0.2 -0.2 -0.1 -0.6 -0.1 -0.2 -0.2 -0.3 -0.1 -0.2 Other income 0.1 0.2 -0.2 0.1 0.0 0.1 0.1 -0.1 0.2 0.0 Total change -3.7 -4.9 -2.1 -12.6 -3.2 -8.8 -10.9 -6.5 -6.5 -10.5 Source: Authors’ calculations from VHLSS 2014, 2016. Notes: Numerical differences in the overall change arise from differences in the number of observations used in the analysis as with missing income information for at least one of the income aggregates are dropped. Wage income growth drove poverty reduction of the decline in the poverty rate during 2014-16, or about half of the total reduction in poverty observed Rising wage income made the largest contribution over the period (Table 8). Rising wages accounted to both poverty reduction and the increase in for about 38 percent of the growth in economically economically secure households. Real wages rose secure households nationwide and more than 52 a cumulative 8 percent during 2014-16 (Figure 12), percent in urban areas. Wage growth was the most promoting poverty reduction and shared prosperity important driver of household income growth in both in Vietnam. The growth of average wages per worker rural and urban areas, and for both ethnic minorities accounted for an estimated 1.8 percentage points and the Kinh and Hoa. Figure 12. Trends in Seasonally Adjusted Average Figure 13. Trends in Seasonally Adjusted Real Monthly Wages by Economic Sector, 2013-17 Average Real Monthly Wages by Industry, 2013-17 6,000 5,500 State Non-State FDI Agric Industry Services 5,500 5,000 4,500 5,000 4,000 4,500 3,500 4,000 3,000 3,500 2,500 3,000 2,000 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 2013 2014 2015 2016 2017 2013 2014 2015 2016 2017 Source: Estimates based on earning data from GSO Labor Force Survey Quarterly Reports, 2013-2016. SECTION I 19 VIETNAM’S REMARKABLE SUCCESS STORY CONTINUES Figure 14. Major Source of Household Income, amount of wage income. Declining poverty among 2010-16 those already earning most of their income from wages 60 contributed 50 to 44 percent to overall poverty reduction, 51.79 53.52 with 45 another 45.71 11 percent 42.9 coming from households 43.9 Share of population 50 49.74 46.2 switching 40 of income, mostly to their major source40.52 38.81 Share of population 35 40 wages The latter effect was driven (Figure 16).34.11 31.01 32.04 by a 30 30 25 Figure 20 15. Major Source of Household Income 20 18.96 17.69 15.02 among 15 Ethnic Minorities, 2010-16 12.68 10 10 5 60 50 0 0 45 45.71 2010 2012 201451.79 201653.52 1 2 42.9 3 4 43.9 Share of population 50 49.74 40 40.52 46.2 Crop cultivation 38.81 Wages Crop cultivation Wages Share of population 35 34.11 40 31.01 32.04 30 Authors’ calculations from VHLSS 2010, 2012, 2014, 2016. Source:30 25 20 20 18.96 15 17.69 Wages grew faster in the private 15.02sector (Figure 12.68 10 10 12). Average monthly wages in the domestic 5 0 sector wages grew the fastest during private 0 2010 2012 2014 2016 1 2 3 4 2014-16, rising by 14 percent and outpacing wage Crop cultivation Wages Crop cultivation Wages growth in the FDI sector which rose by 11 percent. Wages increased more in the industry sector, by shift among ethnic minority households, for which the 11 percent, followed by the agriculture sector (9 switch to mostly wage income contributed 12 percent percent) while wages in the service sector grew by to total poverty reduction in contrast to one percent 5 percent (Figure 13). Due to robust wage growth among the Kinh and Hoa across all industries in both the domestic and FDI sectors, rising wage income drove household Figure 16. Decomposition of Poverty Changes income growth for all groups. by Major Income Source, 2013-17 Wages are now the largest source of income 110 among Vietnamese households. This underlined 100 the subtle change in livelihoods. Between 2010 and 2016, the share of households that earned most of 90 their income from nonagricultural wages increased 80 by 7 percentage points overall (Figure 14). The share Contribution to change in poverty (%) 70 of ethnic minority households obtaining most of their income from wages increased too, almost by 60 13 percentage points, to 44 percent in 2016 (Figure 50 15). Livelihoods were transformed primarily by 40 growing wage income among households that already received some, rather than an increase in the 30 number of households that received wage income 20 for the first time. Indeed, the share of households 10 that received wage income rose by just 2 percentage points between 2010 and 2016. While the transition 0 to wage employment drove down poverty rates -10 among ethnic minorities, many minority workers are Vietnam Ethnic Minorities Kinh & Hoa engaged in low-wage activities, thus most minority Other income Transfers households have not yet achieved economic security. Remittances Wages Household business Other agriculture Sectoral decompositions of changes in poverty8 Crop cultivation Interaction e ect indicate that most of the decline in poverty was Population-shift e ect among households that already earned some Source: Authors’ calculations from VHLSS 2014, 2016 8 Ravallion and Huppi, 1991 20 CLIMBING THE LADDER: POVERTY REDUCTION AND SHARED PROSPERITY IN VIETNAM Figure 17. Growth in Manufacturing Output Figure 18. Net Employment Creation by Sector: and Employment, 2007 -2016 2010-2016 16.0 2,000 14.0 1,500 12.0 1,000 Number of jobs 10.0 500 8.0 - 6.0 (500) 4.0 (1,000) 2.0 (1,500) 0.0 2010 2011 2012 2013 2014 2015 Sơ bộ 2007 2008 2009 2010 2011 2012 2013 2014 2015 Sơ bộ 2016 2016 Other services Hotel & Catering Manufacturing GDP Manufacturing Employment Wholesale and retail trade Other industry Construction Manufacturing Agriculture, forestry and shing Source: GSO, 2017. Robust labor demand supported rapid wage family-farm labor supply in rural areas resulted in an growth. A booming export sector boosted labor increase in wage agriculture jobs. That too contributed demand for both low and high skilled workers. to rising agriculture wages. Manufacturing output grew by a total of 13.3 percent during 2014-16, and the sector added a net 1.4 The net movement of labor out of agriculture was million jobs. The manufacturing sector’s employment followed by a rise in remittances. Though fewer rose in tandem with output growth, suggesting that than five percent of households receive most of the increase in labor drove the overall growth of the their income from remittances, remittances still play sector (Figure 17). Labor demand was also buoyant in an important role in helping households escape the construction, retail and hospitality sectors, which poverty. Rising remittances accounted for more than added a combined 700,000 jobs (Figure 18). All four 21 percent of the decline in poverty during 2014-16, sectors together accounted for 80 percent of total jobs equal to the contribution of rising agriculture income. created during the period, with the manufacturing sector representing more than half of all new jobs. A Agriculture transformation contributed to poverty forthcoming report on jobs in Vietnam (Word Bank, reduction in rural areas 2018) finds that entry of new firms and growth of the more productive existing firms drove the rapid job Increased income from non-crop cultivation helped creation in Vietnam. reduce rural poverty rates, especially among ethnic minorities. Income-poverty decompositions Vietnam began to see a net movement out of suggest that rising non-crop farm income contributed agriculture to more productive sectors with higher about 1.8 percentage points to the total reduction wages. Despite robust job creation, the total number of in poverty rates among ethnic minorities and about employed people grew by less than one percent during 0.8 percentage points to the decline in rural poverty 2014-16. Instead of absorbing unemployed workers, the rates. Non-crop income was especially important fast-growing manufacturing, construction, retail and for those ethnic minority households that attained hospitality sectors drew labor away from agriculture economic security, accounting for 20 percent of the (Figure 18), reducing employment in the sector by more increase in the share of ethnic minorities classified as than 4 percent per year since 2014. This saw higher economically secure between 2014 and 2016 —second growth in wage employment in urban areas. The rising only to the contribution of wage growth. This is part of share of adults engaged in wage employment in urban Vietnam’s agricultural transformation story, which has areas contributed 0.6 percentage points, or 28 percent, also reflected in the country’s success in growing non- to the decline in urban poverty. The decline in unpaid traditional agriculture exports (World Bank, 2016b). SECTION I 21 VIETNAM’S REMARKABLE SUCCESS STORY CONTINUES Household businesses are important for attaining In sum, Vietnam’s export-oriented growth economic security model proved highly successful in reducing poverty and promoting economic security. Nationwide, household business income was the Robust job growth in the export-oriented second-most important contributor to economic manufacturing sector created half of the jobs security, after wage growth. It accounted for 1.4 in the country, boosted demand for labor and percentage points increase in the share of people accelerated the structural transformation that classified as economically secure in Vietnam. In urban saw a net movement of labor out of agriculture. areas, this contribution rose to 1.9 percentage points. With rising incomes, the construction, retail and Household businesses had a more modest effect hospitality sectors were also booming, creating on poverty rates, as most households that received a vibrant labor market. A combination of rising income from household businesses were already wages and a shift toward wage employment raised above the poverty line. The share of households wage incomes and drove poverty reduction. An with nonfarm business income remained broadly increase in non-crop agriculture gave an extra constant at around 35 percent, but the average push to reduce poverty and contributed to greater income from household businesses rose by 79 economic security in rural areas, while rising percent. Thus, growth in household business income household-business income further bolstered was driven by increased profitability rather than new economic security, especially in urban areas. formations of household businesses. LEAVING NO ONE BEHIND SECTION II 23 LEAVING NO ONE BEHIND WHO IS LEFT BEHIND? Poverty is increasingly concentrated in rural total population, they now represent 73 percent areas and among ethnic minorities of the poor population. The overrepresentation of ethnic minorities among the poor reflects the Though rapidly declining, there are still a lot dramatic decline in poverty among the Kinh and of poor people remaining in Vietnam. Their Hoa, and slower progress among minorities until number declined from about 18 million poor recently. In 2010, 8.4 million poor people were people in 2010 to around 9 million in 2016 (Table ethnic minorities, and they represented 47 percent 9). But that means Vietnam’s poor population of the poor population. Another 9.5 million poor Table 9: Number and Distribution of Poor People in Vietnam, 2010-16 Number of poor people Share of poor people (%) Year Vietnam Rural Ethnic Kinh & Hoa Rural Ethnic Kinh & Hoa minorities minorities 2010 17,889,556 16,342,568 8,354,993 9,534,563 91.4 46.7 53.3 2012 15,341,951 13,905,071 7,803,869 7,538,082 90.6 50.9 49.1 2014 12,432,678 11,258,372 7,430,997 5,001,681 90.6 59.8 40.2 2016 9,123,737 8,637,695 6,653,882 2,469,855 94.7 72.9 27.1 Source: Authors’ calculations from VHLSS, 2010, 2012, 2014, 2016. exceeds the entire population of neighboring Lao people were ethnic Kinh and Hoa. By 2010, however, PDR. Of course, Vietnam is now home to more just 2.5 million poor Kinh and Hoa remained. Thus, than 84 million non-poor people, 64 million of three in every four poor Kinh and Hoa in 2010 which are economically secure—more than the escaped poverty, while eight in every ten poor total population of Thailand. The number of poor ethnic minorities in 2010 were still poor in 2016. people is dwarfed by those who are economically Irrespective of ethnicity, poverty in Vietnam is rural. secure by a long shot. Still, there are a lot of them A full 95 percent of poor people lived in rural areas remaining. in 2016, although the rural population accounts for just 68 percent of the total population. Poverty in Vietnam is synonymous with remoteness. A district level poverty map of There is an interplay between ethnicity, Vietnam shows that poverty is mainly concentrated topography and poverty in Vietnam. The remote, in mountainous areas (Map 1). Together, the mountainous areas where the poor are concentrated Midlands and Northern Mountains and the are heavily populated by ethnic minorities. For Central Highlands regions comprise 20 percent of example, about 73 percent of the population in high Vietnam’s total population, yet these areas were mountain communes are ethnic minorities, while home to 56 percent of the poor population. By more than 96 percent of the population in coastal contrast, the Red River Delta and Southeast regions and inland delta communes are Kinh and Hoa. The comprise nearly 40 percent of the population, but ethnic minorities population is overwhelmingly only 6 percent of the poor population. concentrated in rural mountainous communes, where more than 80 percent of them live. Only 11 The poor are heavily concentrated in rural areas percent of ethnic minorities live in urban areas. and among ethnic minorities. About 6.6 million In contrast, 35 percent of the Kinh and Hoa are of Vietnam’s remaining 9 million poor people are urban and another 45 percent lives in coastal and members of ethnic minorities (Table 9). Although inland delta rural communes. Poverty rates for ethnic minorities make up just 15 percent of the 24 CLIMBING THE LADDER: POVERTY REDUCTION AND SHARED PROSPERITY IN VIETNAM both ethnic minorities and the Kinh and Hoa are than the incidence of poverty among the Kinh and higher in mountainous areas. However, in low and Hoa (Table 10). Thus high poverty among ethnic high mountains where the data allows for within minorities do not just reflect their geographical location comparison, the incidence of poverty location, but differences between them and the Kin among ethnic minorities is as much as 6 times more and Hoa as well. Map 1: Poverty Rate by District 2014 Source: World Bank Staff Estimates based on the Population Inter-Census, 2014 and VHLSS, 2014 Note: The boundaries, colors, denominations and other information shown on any map in this work do not imply any judgement on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Table 10: Poverty Rate and Population Distribution by Topography, 2016 Poverty Headcount Population composition Population distribution Rate (%) with location (%) across locations (%) Ethnic Kinh & Hoa Ethnic Kinh & Hoa Ethnic Kinh & Hoa Minorities Minorities Minorities Urban 20.4 0.7 4.5 95.5 8.7 34.7 Rural Communes Coastal 25.4 4.1 3.4 96.6 0.9 4.8 Inland delta 12.1 3.8 2.8 97.2 6.3 41.4 Hills/midlands 9.4 1.7 6.8 93.2 1.7 4.4 Low mountains 34.4 5.3 28.9 71.1 23.0 10.7 High mountains 57.0 10.4 72.8 27.2 59.4 4.2 Source: Authors’ calculations based on the 2010, 2012, 2014, and 2016 VHLSS. SECTION II 25 LEAVING NO ONE BEHIND Even though based in regions with less dynamic equal to about 64 percent of average monthly wage agriculture, the remaining poor have primarily in the industrial and construction sectors and about 57 agricultural based livelihoods. Over 96 percent percent of the average monthly wage in the services of the poor population derives at least some sector. Consequently, poor households’ greater reliance income from agriculture (Figure 19). Nearly half on agricultural wages relative to other types of wage depend exclusively on agriculture, with 27 percent income contribute to their low total income. earning their living solely out of their family farms and another 23 percent mixing family farming The poor are constrained by a lack of human, and agriculture wage work. The rest derive their physical, and financial capital livelihood from a mix of agricultural income and nonagricultural wages (33 percent), agricultural Underlying the concentration of poverty among income and household businesses (8 percent), remote, rural and agricultural households are or a combination of the three (6 percent). Poor fewer productive assets. The poor are disadvantaged households are far less likely than nonpoor by their low education attainment, financial capacity households to derive income from a household and to some extent, unfavorable topography or business. In this, the poor substantially differ from limited access to land. About 57 percent of adults in the non-poor, 38 percent of whom receive income poor households have primary education or less, and from a household business. More than a third fewer than 7 percent have post-secondary education. of the non-poor only earn their income outside The poor have limited access to financial services. agriculture too, compared to fewer than 4 percent of poor households. Figure 19. Distribution of Households by Figure 20. Composition of Wage Income by Poverty Livelihood, 2016 Status, 2010-16 100% Remittances & Transfers only 100% 9 8 90% 90% 6 11 Agriculture with household business 80% 80% 70% Income share (%) 70% Agriculture with wage & nonwage Share of the population 33 60% 29 nonagriculture income 60% 50% Agriculture with non- agriculture wages 50% 4 40% 40% 11 23 30% Agriculture with wages (agric) 30% 20% Non wage agricultureonly 10% 20% 34 27 0% 10% Non agriculture labour income only Not poor Poor 0% 3 Agriculture Low end manufacturing Non-Poor Poor Low end services High end manufacturing High end services Others Source: Authors’ calculations from VHLSS 2010, 2012, 2014, 2016. Many of the poor today already have a wage income, Only 19 percent of adults in the poorest two quintiles but mostly from the lower end of the wage scale. In had an account at a formal financial institution in 2014 2016, about 62 percent of poor households received and only 27 percent of rural adults held an account at income from wages—23 percent from agriculture and a formal financial institution.9 Most of the poor reside 39 percent from non-agricultural sectors. Wages were in hills and mountainous areas. These have access to the primary source of income for 44 percent of the more but less productive land. In the coastal and inland poor (see Annex 1, Figure 36). But the poor mostly earn delta communes, the poor have significantly less land. wages from employment in low-productivity sectors They cultivate 20 percent less land than the non-poor and less-skilled occupations. About 40 percent of their in these areas. These three factors– low education, wage income is derived from agriculture, and another financial capabilities and amount and topography of 40 percent came from low end manufacturing. Less land – determine households’ earnings potential and than 10 percent is earned from services. By contrast, drive livelihood outcomes that so separate the poor the service sector accounted for almost half of all wage from the non-poor. income among nonpoor households (Figure 20). The average monthly wage in the agricultural sector is 9 FINDEX 2014. 26 CLIMBING THE LADDER: POVERTY REDUCTION AND SHARED PROSPERITY IN VIETNAM Figure 21. Employment Profile by Education Figure 22. Trends in Returns to Education, 2011- Level, 2014 14 100 % change in hourly wages relative to lower 90 80 66% 70 secondary education Share workers (%) 60 50 43% 40 34% 29% 30 20% 20 10 10% 0 2011 2012 2013 2014 Never attended school Trade vocational school Upper Secondary University or higher Some primary Lower secondary Vocational school College Primary University or higher College Professional vocational Secondary vocational Basic vocational Upper secondary Farming Non-farm self employment Wage without contract Wage with contract Source: Demombynes and Testaverde, 2017. The lack of education reduces access to better declines as the slope of the cultivated land increases. employment or productive opportunities among Excluding rice, which is also grown on terraces in the poor, resulting in low earnings. Participation highland areas, the average per hectare profitability in wage employment is a clear example. Nearly all of cereals in coastal areas is about three times the adults in poor households had secondary education average in mountainous regions. However, the or less and in 2014. Only 1 in 4 workers with that average per hectare profitability for industrial crops education level had a wage job, least of all one with a increases as the slope of the land rises because high- contract (Figure 21). By contrast, about 85 percent of value cash crops such as coffee thrive in highlands all workers with a college or university degree had a and low mountain areas compared to coastal areas. wage job. The correlation between wage employment On these types of crops, profitability per hectare in and educational attainment remains robust even high mountains are at par with plains and midland after factoring in characteristics such as age, location, areas, and significantly higher than in coastal areas gender, and ethnicity 10. Those with a university degree (Figure 23). have a 50 percent chance of working in a wage job compared to people with only secondary education Sub-optimal land use and crop choice explain but are of a similar age, gender, ethnicity and living in much of the differences in agriculture income the same areas. Due to their low levels of education, between the poor and non-poor. Given the variation even those with a wage job earn substantially less. of profitability for different types of crops by and Among wage workers in 2014, those with a college or topography, households in inland and coastal areas university degree earned 43 to 66 percent more than can maximize their incomes by shifting more of their workers with lower secondary education (Figure 22). land to cereals production. Those in midland and That largely explains why average wages per worker mountainous areas could boost their earnings by among poor households were only 30 percent that of using more of their land to grow perennial or industrial nonpoor households (see Annex 1, Figure 37). crops. However, poor households in midland and mountainous areas allocate their land less efficiently The concentration of poor farming households than nonpoor households (Figure 24). For instance, in highland and mountainous areas diminishes the poor in low mountains devote twice as much land their productivity but not entirely. The per as the non-poor on production of cereals, and half as hectare profitability of cereal crops significantly much on the production of industrial crops. A similar suboptimal allocation can be seen in high mountain 10  See occupational choice regressions in Demombynes and and midland areas. The higher income earned by the Testaverde, 2017. non-poor from production of industrial crops account for the difference in agricultural income between SECTION II 27 LEAVING NO ONE BEHIND poor and nonpoor farmers (Table 11). This suboptimal Only in plains do the poor earn more per hectare allocation is a more pronounced disadvantage than the non-poor in the production of starches and for ethnic minorities who are concentrated in industrial crops, but they have less land. Overall, Figure 23. Profitability per Hectare by Crop Figure 24. Land Use Choice by Crop Type, Type and Topography, 2016 Topography and Poverty Status, 2016 14 16.00 12 14.00 12.00 10 Amount per m2 (VND '000) 10.00 8.00 8 6.00 6 4.00 2.00 4 0.00 Industrial Crop Industrial Crop Industrial Crop Industrial Crop Industrial Crop Starchy Starchy Starchy Starchy Starchy Rice Rice Rice Rice Rice 2 0 Rice Starchy Industrial Crop Coastal Plain Mid-land Low mountain High Mountain Coastal Plain Mid-land Low mountain High mountain Not Poor Poor Source: Authors’ calculations from VHLSS 2016. Table 11: Average Household and Agriculture Income by Crop and Topography, 2016 Topography Rice Starchy crops Industrial Fruit All Crops Crop Coastal 5,066 3,338 2,299 3,742 14,445 Non-poor Plain 16,983 4,902 3,260 4,990 30,134 Mid-land 7,055 2,398 9,054 988 19,495 Low mountain 7,657 4,356 13,149 5,257 30,419 High mountain 5,815 8,753 29,339 3,107 47,015 Coastal 3,705 4,477 367 - 8,562 Plain 6,961 1,227 967 2,104 11,259 Poor Mid-land 8,265 483 389 748 9,886 Low mountain 5,699 6,191 4,044 2,740 18,674 High mountain 7,637 8,333 5,410 579 21,959 Source: Authors’ calculations from VHLSS, 2016. mountainous areas. They do not necessarily own less poor households tend to cultivate less-profitable land in these areas, but are clearly less likely than the crops, and they tend to produce less output than Kinh and Hoa to grow perennial crops. nonpoor households for the same crops. This could be a result of lesser attention from public extension Poor households also earn less profit per hectare and other agricultural departments in development than nonpoor households cultivating the same of farm management and business skills of ethnic crop groups on the same types of land. Nonpoor minority farmers, who require more time and households earn twice as much per hectare for sometimes different training approaches than more industrial crops in midland and high mountain commercial farmers. Smaller public investments in areas and 71 percent more in low mountain areas irrigation and other infrastructure in these areas (Annex 1, Figure 38). They also earn between 50 and may also be a contributing factor. 62 percent more than the poor for starch crops too. 28 CLIMBING THE LADDER: POVERTY REDUCTION AND SHARED PROSPERITY IN VIETNAM Underlying drivers of land-use patterns, crop (Nguyen et. al., 2017). The same study found that choice and productivity have a greater negative households with more agricultural labor also impact on incomes among poor households devote more of their land to perennial crops. This than topography. Both international evidence and suggests that the lack of financial capacity to hire research in Vietnam reveal that supply of household labor may deter households’ from growing fixed labor, agricultural skills, and intermediate inputs crops. In this context, constraints to access to determine cropping choices. Because poor finance have a binding effect on households’ land households often lack access to financial capital, use patterns. they are less likely to grow crops that require expensive intermediate inputs, that take a long time A lack of formal land title limits access to to yield returns (e.g. rubber), or that require a greater finance among poor households. The VHLSS quantity of labor or more specialized skills than the 2014, finds that 23 percent of agricultural land household itself can provide. Consequently, low cultivated by households was not covered by land- education levels contribute to suboptimal crop user certificates. In contrast to the Mekong Delta choices among poor households. A recent World where farmers have land user certificates for 91 Bank report on global agriculture found that skilled percent of their agriculture land, about 30 percent farmers are the most likely to innovate and adopt of farmers’ land in the Central Highlands and 25 new technologies (World Bank, 2017b). In some percent in Midland and Northern Mountainous areas, the shift of land to industrial crops can also areas did not have land user certificates (Figure be constrained by land use/production masterplans 25). About 15 percent of plots with land titles were and generally weaker convening power of the poor used as collateral for loans, compared to just 1.3 communities to adjust the masterplans or reclassify percent of plots without titles (Figure 26). Financing paddy land to other crop land. institutions in Vietnam have a strong preference for land use rights as collateral and there are only Low financial capacity contributes to suboptimal limited channels to share risks in agriculture with land choices among poor households. An uncollateralized lending. Banks rarely use lower analysis based on panel data in Vietnam has collateral requirements as specified by Decree 55, shown that households with large houses (i.e., designed to support those without land. Therefore, high-value collateral) tend to invest more in the the inability of those without land titles to use production of perennial crops, and those that land as collateral hampers their ability to access possess agriculture equipment (e.g., a tractor) credit and consequently their capacity to invest in tend to cultivate more of the main regional crop cultivation of perennial crops. Figure 25. Share of Plots with Land User Figure 26. Use of Agricultural Land as Collateral, Certificates, 2014 2014 100.0 Mekong Delta 90 90.0 Share of plots used as collateral (%) 80.0 Southeastern Area 84 70.0 60.0 Central Highlands 62 50.0 40.0 Northern and Coastal Central 75 30.0 20.0 Midlands and Northern 78 10.0 Mountainous 0.0 Red River Delta 70 Has land title on plot No land tile on plot Land not used as collateral Land used as collateral Source: Authors calculations from the VHLSS, 2014. SECTION II 29 LEAVING NO ONE BEHIND These findings suggest that the key constraints Consequently, even in highlands and mountainous faced by poor households are low levels of areas where perennial crops are more profitable than education, a lack of financial capacity, and to cereals, poor households tend to focus on producing some extent, marginal land quality. Low education cereals. A strategy that effectively alleviated consigns the poor to low paying jobs, and inhibits these constraints and enabled poor households the production of crops requiring specialized skills. in highlands and mountainous areas to shift to Low financial capacity confines poor households to perennial and other cash crops could have a highly crops that require modest inputs and mature quickly. positive impact on poverty and shared prosperity. 30 CLIMBING THE LADDER: POVERTY REDUCTION AND SHARED PROSPERITY IN VIETNAM FINISHING THE JOB AND ADDRESSING NEW CHALLENGES Vietnam’s accelerating structural transformation Moreover, household’s increased reliance on wage has changed its economic landscape, and with income intensifies their exposure to macroeconomic it the poverty reduction and shared prosperity fluctuations especially to external shocks in Vietnam’s agenda The country pursued an export-oriented case. Finally, the high wage premiums for tertiary growth model that catalyzed job creation. Now 70 education suggest that income inequality could rise percent of households earn at least some of their rapidly if inequalities in opportunity are not addressed. income from wages. But as wage income become ubiquitous, most lower-income households already Labor productivity and skills are key pillars to have a wage income. Consequently, the poverty sustain high wage income growth reduction and shared prosperity agenda is no longer about just moving people to wage jobs, but Vietnam now needs to create better jobs and increasingly more about improving the quality of sustain wage income growth without hurting those jobs. Alleviating extreme poverty remains its competitiveness or value proposition to important, but attaining economic security is a investors. Recent wage growth reflected robust rising priority. At the same time, some old challenges labor demand, driven by the country’s success in persist that need to be addressed, like the gaps attracting investors into the export-oriented sectors, between coastal and mountainous regions, and especially electronics and textiles and garments. between ethnic minorities and the Kinh and Hoa. These sectors created the most jobs (Figure 27). Foreign firms created more than 90 percent of new Wage growth has been a driver of poverty reduction jobs in the electronics sector and more than 45 and greater upward mobility. Sustaining wage percent of new jobs in textiles. These firms face high growth will be crucial to advance shared prosperity, international competition and they are footloose. but this will entail new challenges. For example, wage They were attracted, in part, by Vietnam’s low wage growth can undermine competitiveness in labor- rates, therefore rising wages could weaken its intensive sectors, slowing overall economic growth. competitive advantage. Figure 27. Net Job Creation by Firm Ownership Figure 28. Labor Productivity Growth, 2011-16 and Manufacturing Subsector, 2014-2016 100% 8 80% Economywide Manufacturing Contribution to net jobs creation 7 60% 6 40% Annual growth rate (%) 20% 5 0% 4 -20% 3 -40% 2 -60% 1 -80% -100% 0 Textile and Electronics Other All garments manufaturing manufacturing -1 -2 Domestic Private State Owned 2011 2012 2013 2014 2015 Prel. Foreign Owned 2016 Source: Authors’ calculation based on the 2016 VHLSS. Source: GSO, 2017 SECTION II 31 LEAVING NO ONE BEHIND As wages rise, labor productivity becomes education, and it may be competing for skilled increasingly important to competitiveness. Wage workers with other subsectors, even as 9 million less- growth will not erode competitiveness if it is matched skilled workers remain engaged in unpaid household by a commensurate increase in labor productivity. labor. Competition for a small pool of skilled labor However, labor productivity in the manufacturing is likely causing the growth of wage rates to exceed sector is failing to keep pace with wages (Figure 28). productivity gains. A shift to more productive sectors Between 2014 and 2016, labor productivity in the thus needs to be complimented by investment in skills sector grew by 1 percent, while median monthly development to keep labor costs affordable and to wages rose by 11 percent. This disparity indicates provide the requisite skills to produce in these sectors. that the underlying cost of labor is rising much faster than labor productivity growth in the manufacturing Ensuring that investment in infrastructure keeps sector. A widening productivity gap could in future, up with increased demand from high output slow the growth of wages, employment, or both. growth is essential for continued creation of better jobs, especially in the export sector. There is always Moving up the value chain or expanding into high a risk that growth will be throttled by infrastructure productivity subsectors, combined with investing bottlenecks if Vietnam does not invest enough in in skills are necessary to sustain wage growth. infrastructure. During the 2008-2015 period, both Productivity varies widely across manufacturing power demand and containerized freight grew by subsectors. Rebalancing production towards the most 12 percent per year, while total amount of goods productive subsectors would allow a net labor inflow transported grew by more than 8 percent per year into these sectors, raising average wages without (Figure 29). Significant infrastructure investments are increasing the cost of labor. However, Vietnam’s needed to keep up with this demand. Just keeping surplus labor pool is dominated by lowly educated up with electricity demand will require doubling of workers who lack the skills employers demand. the current capacity of 35GW for example. Yet these Consequently, a shortage of skilled workers may significant infrastructure needs must be addressed be driving up wages. The movement of labor from under tight fiscal constraints, necessitating the need other manufacturing subsectors into the electronics for private sector participation in infrastructure and textiles and garments subsectors suggests financing and delivery. This is currently hampered that either competition for labor is high, and rising by a challenging legal and regulatory environment wages are pulling workers out of lower-productivity and cumbersome approval processes (World Bank, subsectors, or other sectors are investing in labor- 2016a). Thus, regulatory reforms and effective saving technology. Preliminary evidence points to the frameworks for risk sharing are necessary for former. Wages are higher in the electronics subsector ensuring reliable electricity supply, efficient and a net labor inflow into electronics appears to be logistics and transportation which are needed not driving up wages for other subsectors. Electronics only to maintain growth, but for Vietnam to move primarily employs workers with post-secondary up the value chain too. Figure 29. Trends in Infrastructure Demand in Vietnam, 2008-15 12.6 12.2 8.4 5.9 GDP Total tons Containerized tons Power demand transported transported Source: GSO, 2016 32 CLIMBING THE LADDER: POVERTY REDUCTION AND SHARED PROSPERITY IN VIETNAM Adapting the social protection system to suit an Aside from the legal constraints of converting export oriented wage economy will be necessary paddy land to other crop (or livestock or aquaculture) land, limited access to financial Job creation was driven by labor demand not just resources and technical skills contributes to in the export sector – but two consumer driven suboptimal land use among poor rural households. sectors, intensifying the country’ vulnerability In the highlands and mountainous regions, many to global demand shocks. Households rode the households could significantly increase their income previous shock due to their dependency on the by devoting more land to perennial crops. However, informal sector—including household agriculture. cultivating the most profitable crops involves high At the height of the crisis, fewer than one-third of costs, long lead times, and significant technical skill. workers had a wage job, but this share has since Easing financial and human capital constraints could risen to 41 percent, and most households now rely help poor farmers in highland areas make more on wages for the bulk of their income. As wage efficient use of their land. Lack of finance is not only a employment continues to rise, and the capacity of barrier for mobilizing upfront investments in perennial the informal sector to serve as an employer of last crop cultivation, it plays in the risk aversion among the resort diminishes, future external shocks that could poorest farmers. A lack of resources to fall back on in stagnate wages or lead to job losses will have a more an emergency discourages cash-crop production and disruptive impact on household welfare. inhibits specialization. For this reason, Vietnam needs to strengthen Strengthening property rights and adapting foundations for managing external shocks by extension services have the potential to unlocking establishing a social protection system suitable agriculture potential. Lack of land titles that could be for a wage economy. More than 80 percent of new used as collateral for obtaining loans contributes to the labor market entrants are employed in a wage job, financial constraints faced by the poor. Completing while retirees are predominantly farmers. Within a land certificates issuance addresses this barrier. The few years, wage workers will significantly outnumber lack of skills could be addressed through improving nonwage workers. The growing share of wage the messaging offered by extension workers and workers increases households’ exposure to both the establishing collaborative groups/cooperatives, domestic and idiosyncratic external shocks, while the critical to connecting producers with food processors shrinking informal sector will be less able to absorb and exporters. Still, without other complementary excess labor from the formal sector if labor-market public investments, for example, in applied research conditions deteriorate. Unemployment insurance and of seedling materials, water saving technologies, and active labor market policies become more critical to other support services, the shift to perennial crops for social protection in a wage economy, as they help the poor farmers will be slow. households manage income shocks and stabilize the business cycle. Equalizing opportunities in education is central to the poverty and shared prosperity agenda Changing land use and strengthening land property rights will further unlock the agriculture A top priority is getting the remaining poor into potential of the poor and near poor better paying jobs – basically formal wage jobs and non-traditional crops in agriculture. The best Even as we speak of a post agricultural wage jobs in Vietnam require tertiary education so the economy on the horizon, agriculture holds poor are hurt by their low levels of education. Most the potential to lift the remaining poor out of enter the labor market with secondary education poverty. The transformation of the agricultural and only 6 percent make it to university. But returns sector remains unfinished, and human, physical and to secondary education, and even upper secondary financial capital constraints inhibit the ability of poor education are low. With big premiums for university, rural households to take advantage of emerging college and professional vocational training, the opportunities. Improving the efficiency of land use poor’s lack of progression to tertiary education hurts can unlock the economic potential of smallholder their earning capacity. farmers and support poverty reduction and shared prosperity, even in the country’s remote highlands Unfortunately, disparities in education persist, and mountainous regions. consigning the poor and near poor to low SECTION II 33 LEAVING NO ONE BEHIND earning jobs from the time they enter the labor that differences in school quality explained one- market. School enrollment rates among children third of the variation in average math scores. Within from households in the bottom income quintile the same school, students from all backgrounds lag the rates for children from households in the benefited equality from an increase in school quality, richest quintile across all grade levels (Figure which is not always the case in other countries. These 30). These differences widen toward the end of findings indicate that differences in school quality lower secondary school, at which point a third of between poor and nonpoor communities, rather children from households in the bottom quintile than unequal treatment of students within schools, have dropped out. By age 19, fewer than one-fifth tend to drive disparities in educational attainment. Figure 30. Net School Enrollment by Welfare Figure 31. Net Enrollment by Ethnicity and Age, Status, 2016 2016 100 100 90 90 80 80 School enrollment (%) 70 70 School enrollment (%) 60 60 50 50 40 40 30 30 20 20 10 10 0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 Age Age Ethnic Minorities Kinh & Hoa Poorest Quintile Richest Quintile Nguồn:. Tính toán của các tác giả từ KSMS, 2016. of 19-year-olds from households in the bottom Figure 32. Per Pupil Spending on Tutoring and quintile remain in school. By contrast, over two- Study Support by Welfare Quintile and Ethnicity, thirds of students from households in the top 2016 quintile continue with their education up to college or university. Similar differences are observed for ethnic minorities’ children, who also drop off after 4, 882₫ lower secondary education (Figure 31). Given the high wage premium for tertiary education, these Am ount per year ('000) disparities are slowing social mobility. 3, 441₫ 2, 829₫ Differences in quality of education partly drive 2, 310₫ these inequalities. The Young Lives Study tracked two cohorts of children, one from the age of two 1, 621₫ and one from the age of eight. It found that poor academic achievement contributes to dropout rates 745₫ 811₫ at the lower secondary level. Among students who completed lower secondary education, those who had good or excellent test scores when they were 12 Poorest Second Third Fourth Richest Ethnic K in h & Hoa years old had a 36 percent lower probability of not Q uintile Q uintile M inorities continuing to the upper secondary level compared to students from similar backgrounds. In addition, W elfare Q uintile Ethnicity 45 percent of students with poor test scores failed to even complete lower secondary school (Le Thuc Source: Authors calculations from VHLSS, 2016 Duc and Tran Ngo Minh Tam, 2013). The study found 34 CLIMBING THE LADDER: POVERTY REDUCTION AND SHARED PROSPERITY IN VIETNAM Household socioeconomic status also to the curricula and learning times are necessary to contributes to educational disparities, close this gap. Additionally, evidence suggests that especially in terms of the resources households drop outs at secondary schools are higher for children devote to tutoring, study support and extra in communes further off from the district center, payments. Households in the top income quintile pointing to challenges in access to tertiary education spend more than 6.5 times as much per student for pupils in remote areas. Their remoteness increases on tutoring and study support than households the cost for post-secondary education, which could in the bottom quintile (Figure 32). Kinh and Hoa be mitigated by raising education aid. households spend about 3.5 times more per student than ethnic minority households. There is In summary, efforts to promote shared prosperity evidence that tutoring is a significant predictor of will increasingly focus on enabling households pupil’s academic grades, even after controlling for to achieve economic security while addressing household background or school characteristics. persistent welfare disparities across groups. Parental education also plays an important role, shared prosperity will require creating more and as students whose mothers have completed better jobs that are accessible to all workers, rich lower secondary education have a greater chance and poor. This can be achieved in the context of of progressing to tertiary education, even after the current export oriented model by moving up controlling for test scores and other factors (Le the value chain and promoting investment in more Thuc Duc and Nguyen Thi Thu Hang, 2016). Overall, productive sectors of the economy to increase students from households in the bottom quintile labor inflows these sectors. The development of are the most likely to drop out of school. These more sophisticated economic activities will, in turn, differences translate into disparities in educational require investment in workforce skills to ease wage attainment, slowing intergenerational mobility pressures arising from competition for a limited and exacerbating inequality. pool of workers and provide the skills needed for the economy to move up the value chain. Expanding Vietnam therefore needs to address issues with access to high-quality education, especially in the quality of education – that makes tutoring so poor and underserved regions, will be critical not central for academic achievement. The importance only for developing skills, but to reducing existing of tutoring to pupils’ academic achievement at lower inequalities in access to opportunities between the secondary education signals inadequate teaching poor and the non-poor. and places poor children at a disadvantage. Reforms SECTION II 35 LEAVING NO ONE BEHIND PRIORITIES FOR POVERTY REDUCTION AND SHARED PROSPERITY Living standards in Vietnam are improving, from services, the highest-paid sector. Vietnam’s poverty rates are declining, and the consumer labor market places a high premium on college class is rapidly expanding. The incidence of and university education, and the better jobs are poverty has begun to fall sharply for ethnic accessed mainly by workers with post-secondary minorities as well, revealing the potential for education. The poor’ low levels of education thus further poverty reduction among disadvantaged contribute to their low earnings both in and groups. Meanwhile, households that have outside agriculture. However, existing inequalities escaped poverty face a low risk of falling back. in access to post-secondary education, entrenches Just 2 percent of members of households that this disadvantage. were above the poverty line in 2014 had fallen into poverty by 2016. Economic mobility has been Rapid wage growth and agriculture overwhelmingly oriented upwards, and the share transformation drove poverty reduction. of the population classified as economically secure Increasing wages accounted for more than half expanded by 20 percentage points between 2010 of the reduction in poverty observed between and 2016. Overall, 70 percent of the population is 2014 and 2016, while income from non-crop now classified as economically secure, including cultivation and remittances each contributed 20 the 13 percent who are part of the global middle percent. Wage growth also propelled the rise of class. The rise of the consumer class is changing the consumer class, supported by an increase in society’s aspirations and altering the focus of the household business income. Within agriculture, poverty reduction and shared prosperity agenda. a shift away from traditional cereals crops cultivation was a driver of household income Vietnam’ success in reducing poverty was growth, and differences in land use among on the back of a rising consumer class and smallholder farmers drives earnings differences most importantly, growing wage incomes between poor and nonpoor households. These buoyed by increased labor demand in the trends reflect the centrality of wages to livelihoods export sector. More than 3 million wage jobs and shared prosperity in Vietnam, as well as the were created between 2010 and 2016, half in potential for further agricultural transformation to the manufacturing sector and the other half in reduce poverty in rural areas. To sustain poverty construction, retail, and hospitality. This spurred reduction and transition to economic security, the movement of 2 million workers out of the Vietnam should keep creating jobs, sustain wage agricultural sector, marking a turning point in income growth, equalize opportunities to good the country’s structural transformation. Robust jobs and unlock the potential in agriculture. labor demand boosted average monthly wages in the private sector by 14 percent, which in turn Achieving these objectives calls for a focus on the put upward pressure on wages in other sectors. following strategic priorities for socio-economic Agricultural wages rose by 9 percent, and wage development of the country. employment in the agricultural sector expanded. Vietnam is now primarily a wage economy, as Boosting labor productivity and investing (i)  more than 54 percent of households earn most in infrastructure to sustain job creation and of their income from wages and a full 70 percent wage growth without losing competitiveness. of households receive at least some amount of wage income. Most of the poor households now With wages becoming central to households’ also receive a wage income, but more than half of livelihoods, sustained welfare improvement those wages come from agriculture, the lowest- hinges on future wage income growth and paid sector, while less than 10 percent come creation of better jobs. But recently wage growth 36 CLIMBING THE LADDER: POVERTY REDUCTION AND SHARED PROSPERITY IN VIETNAM outpaced labor productivity growth. To boost differences between quality of schools in labor productivity, raise wage incomes and poor and non-poor communities explain continue creating wage jobs, Vietnam needs the variation in academic achievement at to move production up the value-chain and lower secondary level, which determines promote investment into higher productivity progression to tertiary education. This sectors to shift labor inflows into these signals that inadequate teaching places sectors. This can be achieved by: poor children at a disadvantage. This could be addressed by increasing teaching hours • Attracting FDI into higher value, agriculture, in school. manufacturing and services sectors Revision the curricula and pedagogical •  Supporting growth of the domestic private •  approach - Teaching and testing should small and medium enterprises through place more emphasis on developing information and skills upgrading to link problem solving and critical thinking, them to multi-national corporations the skills that employers deem lacking in I ncreasing investments in infrastructure so •  Vietnam. that supply of transport, electricity, logistics and telecommunications keeps up with the (iii) Spurring agriculture structural high demand from a fast-growing export transformation through changing farmland sector, and provide an enabling environment use patterns, strengthening land user rights, for the country to move up the value chain or and improving skills of the poor farmers. into high value added sectors. Addressing sub-optimal farmland use patterns Implementing education reforms designed (ii)  is key to unlocking the agriculture potential of to equalize opportunities and develop the poor by aligning land use with comparative workforce skills. advantages of specific areas and farm income generation objectives, as outlined in the Rising private sector wages in the face of Agricultural Restructuring Plan. This requires abundant supply of labor suggest that firms a bolder shift of land from rice and maize to are competing for a limited pool of competent more profitable annual and perennial crops. laborers. Indeed, a significant share of hiring Key to achieving this is: employers say that job applicants lack the skills needed for the job, even for low-skilled • Strengthening land user rights through issuing jobs. Investments in skills development would land titles which could help increase household increase the pool of competent workers, access to credit (using land as collateral), facilitate the expansion of value chains into enabling the poor to invest in more profitable more sophisticated activities and support the crops that require costly initial investments, growth of new sectors. Expanding access to intermediate inputs or hiring of labor. high quality education across groups will be Improving farm management and business •  vital not only to develop skills, but to reduce skills of the poorer farmers, often neglected by inequalities and increase access to better public extension and investment programs, paying jobs for all. This can be achieved by: could help boost agricultural productivity, thereby reducing the productivity gap with Reforming the structure of the school day •  less-poor farmers. to increase instruction hours - Tutoring and   SECTION II 37 LEAVING NO ONE BEHIND REFERENCES Azevedo, Joao Pedro; Gabriela Inchauste; Sergio Olivieri; Jaime Saavedra; Hernan Winkler. 2013. Is Labor Income Responsible for Poverty Reduction? A Decomposition Approach. World Bank Policy Research Working Paper WPS6414, World Bank. Washington, DC. USA Barros, Ricardo Paes, Mirela de Carvalho, Samuel Franco and Rosane Mendoça. 2006. “Uma Análise das Principais Causas da Queda Recente na Desigualdade de Renda Brasileira.” In: Revista Econômica 8 (1): 117–147. Rio de Janeiro: Universidade Federal Fluminense. http://www.uff.br/revistaeconomica/V8N1/RICARDO.PDF Bussolo, Maurizio and Lopez-Calva, Luis Felipe., 2014. Shared Prosperity: Paving the Way in Europe and Central Asia,” Europe and Central Asia Regional Framework for Shared Prosperity. The World Bank. Washington, DC Datt, G. and M. Ravallion (1992) Growth and Redistribution Components of Changes in Poverty Measures: A Decomposition with Applications to Brazil and India in the 1980s. Journal of Development Economics 38, 275–295. Demombynes and Testaverde, 2017. Employment Structure and Returns to Skill in Vietnam: Estimates Using the Labor Force Survey. Background paper. World Bank. Le Thuc Duc and Nguyen Thi Thu Hang, 2016. Inequality in Educational Opportunities and Outcomes: Evidence from Young Lives Data in Vietnam. Country Report. Young Lives Study. London. Le Thuc Duc and Tran Ngo Minh Tam, 2013 Why Children in Vietnam Drop out of School and What They Do After That. Working Paper 102. Young Lives Study. London Nguyen T. R,, Nguyen L. D., RATTIYA S. L., ULRIKE G., 2017. Determinants of Farmers’ Land Use Decision- Making: Comparative Evidence from Thailand and Vietnam. World Development Vol. 89 pp 199-213 Ravallion, M., and M. Huppi (1991) Measuring Changes in Poverty: A Methodological Case Study of Indonesia during an Adjustment Period. The World Bank Economic Review 5:1, 57–82. World Bank. 2016. Vietnam Systematic Country Diagnostic: Sustaining Success – Priorities for Inclusive and Sustainable Growth. World Bank. Washington DC World Bank and IPSARD. 2016. Vietnam Development Report 2016: Transforming Vietnamese Agriculture: Gaining More from Less. Washington D.C. and Hanoi. World Bank. 2017a. Riding the Wave: An East Asian Miracle for the 21st century. Washington, D.C.: World Bank Group. https://hubs.worldbank.org/docs/ImageBank/Pages/DocProfile.aspx?nodeid=27585372 World Bank. 2018 (forthcoming). Vietnam Future Jobs. World Bank. Washington DC World Bank Group. 2017b. Future of Food: Shaping the Food System to Deliver Jobs. World Bank. Washington, D.C. 38 CLIMBING THE LADDER: POVERTY REDUCTION AND SHARED PROSPERITY IN VIETNAM ANNEXES: ADDITIONAL TABLES AND FIGURES Table 12: Income Decomposition of Changes in Middle Class in Vietnam, 2014-16 Vietnam Rural Urban Ethnic Kinh & Minorities Hoa Propensity to consume -0.1 -1.0 1.2 -0.4 -0.1 Share of Adults 0.0 0.2 -0.5 0.3 0.0 Share of wage workers -0.4 0.0 -0.5 0.0 -0.4 Average wage -1.7 -0.9 -4.0 -0.6 -2.0 Average business earnings -1.3 -0.9 -2.7 -0.5 -1.7 Average crop income -0.1 -0.1 -0.2 0.0 -0.1 Average other farm income -0.3 -0.1 -0.1 -0.2 -0.2 Remittances -0.6 -0.2 -1.2 -0.3 -0.6 Transfers 0.0 -0.1 -0.1 0.0 -0.1 Other income -0.2 0.2 -0.7 0.0 -0.1 Total change -4.5 -2.9 -8.9 -1.7 -5.3 Source: Authors’ calculations from VHLSS 2014, 2016. Notes: Numerical differences in the overall change arise from differences in the number of observations used in the analysis as with missing income information for at least one of the income aggregates are dropped. Figure 33. Trends in Poverty Gap and Poverty Gap Figure 34. Major Source of Income in Urban Areas, Squared in Vietnam, 2010-16 2010-16 7.0 Areas 70 6.0 62.12 62.26 63.53 5.9 60 57.6 5.0 4.5 50 Poverty rate (%) Share of population 4.0 3.7 40 3.0 2.6 30 29.35 2.0 25.91 26.28 25.53 1.0 20 0.5 0.3 0.3 0.2 0.0 10 2010 2012 2014 2016 0 2010 2012 2014 2016 Squared Poverty Gap Poverty Gap Household business Wages Source: Authors’ calculations from VHLSS 2010, 2012, 2014, 2016. SECTION II 39 LEAVING NO ONE BEHIND Figure 35. Major Sources of Income in Rural Areas, Figure 36. Major Source of Income by Poverty 2010 vs 2016 Status, 2016 50 5.0 4.3 Other income 45 40 Transfers Share of the population Share of population 35 44.2 30 Remittances 54.5 25 20 Wages 2.8 15 12.5 Household business 10 5 20.3 Other agriculture 0 Crop cultivation Other agriculture Household Wages 33.1 7.1 business Crop cultivation 10.5 2010 2016 Not Poor Poor Source: Authors’ calculations from VHLSS 2010, 2012, 2014, 2016. Figure 37. Average Annual Wage Per Worker, 2016 (VND’ 000) 42,100 42.714 25,096 20,740 13,581 14,528 6,135 4,067 Agricculture with Agricculture with Agricculture with Agricculture with wages (agric) non-agriculture wages wage & nonwage household business nonagriculture income Non-Poor Poor Source: Authors’ calculations from VHLSS 2016. Figure 38. Profitability by Crop Type and Poverty Status, 2016 16 Not Poor 14 Poor 12 Amount per m2 (VND '000) 10 8 6 4 2 0 Coastal Plain Mid-land Low mountain High mountain Coastal Plain Mid-land Low mountain High mountain Coastal Plain Mid-land Low mountain High mountain Rice Starchy Title Industrial Crop Source: Authors’ calculations from VHLSS 2016.