VIET NAM BI-ANNUAL POVERTY & EQUITY UPDATE Towards more inclusive cities June 2024 © 2024 International Bank for Reconstruction and Development / The World Bank This work is a product of the staff of the World Bank with external contributions. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Nothing herein shall constitute or be considered to be a limitation upon or waiver of the privileges and immunities of The World Bank, all of which are specifically reserved. Rights and Permissions This work is available under the Creative Commons Attribution 3.0 IGO license (CC BY 3.0 IGO) http://creativecommons.org/licenses/by/3.0/igo. Under the Creative Commons Attribution license, you are free to copy, distribute, transmit, and adapt this work, including for commercial purposes, under the following conditions: Attribution—Please cite the work as follows: World Bank. 2024. Viet Nam bi-annual poverty & equity update - June 2024 – Towards more inclusive cities. Washington, DC: World Bank. License: Creative Commons Attribution CC BY 3.0 IGO Translations—If you create a translation of this work, please add the following disclaimer along with the attribution: This translation was not created by The World Bank and should not be considered an official World Bank translation. The World Bank shall not be liable for any content or error in this translation. Adaptations—If you create an adaptation of this work, please add the following disclaimer along with the attribution: This is an adaptation of an original work by The World Bank. Views and opinions expressed in the adaptation are the sole responsibility of the author or authors of the adaptation and are not endorsed by The World Bank. Third-party content—The World Bank does not necessarily own each component of the content contained within the work. The World Bank therefore does not warrant that the use of any third-party-owned individual component or part contained in the work will not infringe on the rights of those third parties. The risk of claims resulting from such infringement rests solely with you. If you wish to re-use a component of the work, it is your responsibility to determine whether permission is needed for that re-use and to obtain permission from the copyright owner. Examples of components can include, but are not limited to, tables, figures, or images. All queries on rights and licenses should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; e-mail: pubrights@worldbank.org. Cover design: Ha Doan and Saengkeo Touttavong VIET NAM BI-ANNUAL POVERTY & EQUITY UPDATE Towards more inclusive cities June 2024 Contents Abbreviations................................................................................................................... vii Acknowledgments........................................................................................................... viii Overview........................................................................................................................... 1 Part 1. Poverty & Equity Update..........................................................................................................................................1 Part 2. Poverty and Inclusion in Urban Areas and Cities.................................................................................................3 PART 1. Poverty & Equity Update.................................................................................... 8 1. Poverty and equity developments.................................................................................. 10 A decade of successful poverty reduction in Viet Nam stalled during the COVID-19 pandemic...........................10 Pandemic-related slowdowns interrupted long-run poverty and inequality trends in the East Asia and Pacific region................................................................................................................................................................10 In 2022, progress in poverty reduction stalled...............................................................................................................11 Household impacts varied, with urban areas more negatively impacted than rural ones.......................................13 Differences in poverty trends across geographic regions are evident........................................................................15 The vulnerable have fewer coping strategies and experienced worsening conditions...........................................16 Inequality declined slightly as households at the higher ends of the distribution experienced relatively more adverse impacts than those at the bottom, particularly those in urban areas................................................18 2. Sources and dynamics behind changes in poverty.......................................................... 22 In 2022, labor incomes experienced small declines in either the intensive or extensive margins.........................24 Transfers offset some labor income losses, but these amounts were small on average...........................................25 Transition out of agriculture is linked with income and poverty dynamics..............................................................26 Short- and long-term trends will impact employment..................................................................................................28 3. Looking forward............................................................................................................ 32 With the return of economic growth, income and poverty rates are improving.......................................................32 Policies for the Next Mile...................................................................................................................................................34 Identify and address shocks..............................................................................................................................................35 Social protection needs to be modernized to guard against shocks...........................................................................35 Countercyclical fiscal policies..........................................................................................................................................37 Better and more secure jobs..............................................................................................................................................37 Annex A. Figures – Part 1................................................................................................... 38 Annex B. Household consumption ..................................................................................... 42 PART 2. Poverty and Inclusion in Urban Areas and Cities........................................... 44 1. Urbanization is important for the Next Mile................................................................... 46 While Viet Nam’s official urban population is growing, the share of people living in an urbanized setting could be larger.....................................................................................................................................................................47 A large share of Viet Nam’s urban population and growth are found in its five municipalities..............................49 CONTENTS 2. The urbanization of poverty........................................................................................... 54 Under certain conditions, urban poverty can rise more than rural poverty..............................................................55 While urban poverty rates are much lower than rural ones, they can vary by different measurement or classifications..................................................................................................................................................................56 3. Case Study: Mapping spatial inclusion in Ho Chi Minh City............................................ 58 City-level analysis is important to understand the gradient of urbanization and poverty......................................58 Survey-level indicators mask variation across a highly populated city.....................................................................58 A simple framework for within-city analysis..................................................................................................................63 Across HCMC, distances to amenities are shorter in peripheral areas of the city....................................................67 There are pockets of overcrowded and low-quality housing in HCMC......................................................................67 Flooding is a risk across many areas of HCMC...............................................................................................................69 A composite view illustrates a gradient of poverty and deprivations.........................................................................70 4. Data-informed policies in urban areas............................................................................ 74 Better data for within-city analyses.................................................................................................................................74 Inclusive policies for vulnerable groups in urban areas and cities.............................................................................76 Resilience amid climate change........................................................................................................................................77 Urban planning....................................................................................................................................................................77 Annex C. Figures – Part 2................................................................................................... 78 References........................................................................................................................ 79 List of Boxes Box 1.A. World Bank monitoring indicators.................................................................................................................. 20 Box 1.B. Household income dynamics............................................................................................................................ 29 Box 2.A. Urban and rural classification in Viet Nam..................................................................................................... 50 Box 2.B. DB and DOU classifications............................................................................................................................... 52 Box 2.C. Ho Chi Minh City is the largest city in Viet Nam, with a gradient of urbanization and characteristics.....60 Box 2.D. Data....................................................................................................................................................................... 65 Box 2.E. Migrants in cities................................................................................................................................................. 72 List of Figures Figure O.1. Recent changes in poverty in Viet Nam – three phases............................................................................. 1 Figure O.2. Growth incidence curves, household consumption.................................................................................. 2 Figure O.3. Poverty trends and projections...................................................................................................................... 3 Figure O.4. Urban share of the poor and population...................................................................................................... 4 Figure O.5. Share of urban population in Viet Nam 2022, various classifications...................................................... 4 Figure O.6. There is a gradient of poverty across urbanization categories................................................................. 5 Figure 1. Recent changes in poverty in Viet Nam – three phases............................................................................... 10 Figure 2. Strong poverty reduction and links to economic growth........................................................................... 11 Figure 3. Inequality in EAP, a varied story..................................................................................................................... 12 CONTENTS  Figure 4. 30 years of poverty trends in Viet Nam, 1992-2022....................................................................................... 12 Figure 5. If your household stopped earning income today, how long could you survive?................................... 13 Figure 6. Growth incidence curves, household consumption.................................................................................... 13 Figure 7. Poverty rates by urban and rural area, 2010-22............................................................................................. 14 Figure 8. How long can households economically survive if their incomes stopped today, by bottom-40 and top-60................................................................................................................................. 14 Figure 9. Distribution of the population, by economic classes.................................................................................. 15 Figure 10. Poverty rates by region, 2010-22................................................................................................................... 16 Figure 11. Ethnic minorities experienced a larger absolute and relative increase in poverty than the Kinh...... 17 Figure 12. Average income gap to the poverty line....................................................................................................... 17 Figure 13. The prosperity gap, 2010-22........................................................................................................................... 17 Figure 14. Absolute changes in household consumption............................................................................................ 18 Figure 15. Inequality trends............................................................................................................................................. 18 Figure 16. Decomposition of the change in inequality................................................................................................ 19 Figure 17. Growth incidence curves, total household income.................................................................................... 22 Figure 18. Trends in household income by component, 2010-22............................................................................... 23 Figure 19. 2020-22 UMIC ($6.85/day 2017 PPP) poverty decomposition, by income sources................................ 24 Figure 20. Changes in household labor income, decomposed by intensive versus extensive margins................ 25 Figure 21. Private and public transfers increased, but not enough to compensate for losses in labor incomes................................................................................................................................................ 25 Figure 22. Growth and poverty trends near the pandemic period, selected EAP countries.................................. 26 Figure 23. Over time, more households receive labor incomes only in the form of wage income........................ 26 Figure 24. Distribution of households, by labor income diversification and region, 2022..................................... 26 Figure 25. Share of households participating in agriculture is declining................................................................. 27 Figure 26. LMIC and UMIC poverty rates, by household income sources, 2010-22................................................. 27 Figure 27. Average incomes, by primary sector and region 2016-22.......................................................................... 28 Figure 28. Employment trends in wage-earning households, by occupation.......................................................... 29 Figure 29. Employment share, by cohort and year....................................................................................................... 29 Figure 30. Share of household income by source and decile, 2022............................................................................ 30 Figure 31. Annual household income by source and decile, 2022.............................................................................. 30 Figure 32. Trends in net annual household income, by region................................................................................... 31 Figure 33. Increase in household income from 2010 to 2022, by region.................................................................... 31 Figure 34. Annual household income, by labor income diversification.................................................................... 31 Figure 35. Distribution of households, by labor income diversification................................................................... 31 Figure 36. Recovering labor incomes, but still lag behind pre-COVID trends.......................................................... 32 Figure 37. National trends in labor force size................................................................................................................ 32 Figure 38. Poverty trends and projections..................................................................................................................... 33 Figure 39. Household income trends, by region........................................................................................................... 38 Figure 40. Regional annual household income in 2010 and 2022............................................................................... 39 Figure 41. Decomposing monthly household income from farm production.......................................................... 40 Figure 42. Decomposing monthly household income from wages............................................................................ 41 Figure 43. Food expenditures, 2020 vs 2022................................................................................................................... 43 Figure 44. Relationship between urbanization, GDP, and poverty ............................................................................ 46 Figure 45. Viet Nam’s urban population is low compared to regional benchmarks ............................................... 47 Figure 46. Population, by urban-rural classification.................................................................................................... 48 Figure 47. Share of urban population in Viet Nam 2022, various classifications..................................................... 48 v Viet Nam bi-annual poverty & equity update - June 2024 CONTENTS Figure 48. Population trends across large cities in Viet Nam...................................................................................... 49 Figure 49. Distribution of population growth, by location (%)................................................................................... 49 Figure 50. Source of population change, 2009-19......................................................................................................... 50 Figure 51. Illustration of levels 1-3 administrative areas............................................................................................. 51 Figure 52. Distribution of urban and rural populations, by different classifications.............................................. 54 Figure 53. Share of population in urban areas is increasing, as is the urban poor.................................................. 55 Figure 54. Stylized illustration of poverty, by urban or rural classification.............................................................. 56 Figure 55. Share of the UMIC poor by urban classifications, Viet Nam 2016............................................................ 56 Figure 56. Gradient of poverty across urbanization categories.................................................................................. 57 Figure 57. Profiling by different classifications of urbanization................................................................................ 57 Figure 58. Household assets............................................................................................................................................. 59 Figure 59. Education spatial distribution....................................................................................................................... 60 Figure 60. Net population change from 2009 to 2019, by district............................................................................... 61 Figure 61. The 24 administrative districts of HCMC..................................................................................................... 62 Figure 62. HCMC districts, grouped and mapped......................................................................................................... 62 Figure 63. Distribution of HCMC’s population using the grid..................................................................................... 62 Figure 64. Population is lowest in the core, 2019.......................................................................................................... 62 Figure 65. HCMC districts vary by population and density, 2019............................................................................... 63 Figure 66. Framework........................................................................................................................................................ 63 Figure 67. Map of the median asset-based wealth scores, by district and grid........................................................ 64 Figure 68. Population distribution by proximity to amenities.................................................................................... 67 Figure 69. Average distance by quintile of the wealth index....................................................................................... 67 Figure 70. Housing features.............................................................................................................................................. 68 Figure 71. Housing quality................................................................................................................................................ 68 Figure 72. Distribution of the population in areas with exposure to flood risk....................................................... 69 Figure 73. Average number of deprivations, by grid cell............................................................................................. 71 Figure 74. Number of deprivations, by district.............................................................................................................. 72 Figure 75. % of the population that are recent migrants.............................................................................................. 72 Figure 76. Migrant population in HCMC (thousands)................................................................................................... 73 Figure 77. Share of HCMC population that are migrants (%)....................................................................................... 73 Figure 78. Deprivations by migrant or non-migrant..................................................................................................... 74 Figure 79. Population at risk of 10-year and 100-year floods, by district.................................................................. 78 List of Tables Table 1. Global poverty lines in PPP and VND............................................................................................................... 21 Table 2. Summary of economic class definitions.......................................................................................................... 22 Table 3. Household consumption, 2020 vs 2022............................................................................................................ 42 Table 4. Average health consumption............................................................................................................................. 43 Table 5. Criteria for urban wards, and district towns, or rural communes............................................................... 51 Table 6. Differentiation of urban areas........................................................................................................................... 53 Table 7. Selected indicators for HCMC........................................................................................................................... 65 Table 8. Deprivation thresholds for selected indicators.............................................................................................. 70 Table 9. Recommendations to mitigate urban poverty................................................................................................ 75 Towards more inclusive cities vi ABBREVIATIONS Abbreviations ASEAN Association of Southeast Asian Nations CHN China COVID-19 Coronavirus disease 2019 DB Dartboard methodology DOU Degree of Urbanization EAP East Asia and Pacific EM Ethnic Minority FAFH Food-away from home GDP Gross Domestic Product Ha Hectares HCMC Ho Chi Minh City HH Households IDN Indonesia ILSA Institute of Labour and Social Affairs IPL International Poverty Line Km Kilometer Km2 Kilometers squared LMIC Lower-Middle Income Country MOLISA Ministry of Labour, Invalids and Social Affairs NGO Non-governmental organization OECD Organisation for Economic Co-operation and Development  PHL Philippines PIP Poverty and Inequality Platform PL Poverty line PPP Purchasing power parities SDG Sustainable Development Goal THA Thailand UMIC Upper-Middle Income Country VASS Viet Nam Academy of Social Sciences. VHLSS Viet Nam Household Living Standards Survey VND Viet Nam Dong VNM Viet Nam VSS Viet Nam Social Security WB World Bank WBG World Bank Group WDI World Development Indicators vii Viet Nam bi-annual poverty & equity update - June 2024 ACKNOWLEDGMENTS Acknowledgments The bi-annual poverty update is a product of the Viet Nam Poverty and Equity team of the World Bank. This report was prepared by Judy Yang (Senior Economist), with support and contributions from Matthew Wai-Poi (Lead Economist) and Laura Takeuchi Rodriguez (Economist). The following individuals and teams provided valuable inputs to Part 1: • The construction of the consumption aggregate used for the World Bank global poverty measurement was conducted with cooperation and technical support from the Social Economics and Environment Department, General Statistics Office of Viet Nam. • Analysis of COVID-19 shocks in urban areas using the World Bank’s COVID-19 household surveys was conducted by Balasubramanyam Pattath (Consultant). • An adaptive social protection stress test for Ho Chi Minh City during the COVID-19 pandemic was conducted by: Nga Thi Nguyen (Social Protection Specialist), Kenichi Chavez (Senior Economist), Bao Ha Nguyen (ILSA, MOLISA). The following individuals and teams provided valuable inputs to Part 2: • Urban delineations for Viet Nam were provided by Shohei Nakamura (Economist). • Information on urban administrative classification in Viet Nam was collected by the Viet Nam Academy of Social Sciences. • Flood analysis of Ho Chi Minh City was conducted by Garrett Benz (Consultant, World Bank City Resilience Program). Comments on background papers that informed this analysis were received from Ross Eisenberg (Disaster Risk Management Specialist, Task Leader of the City Resilience Program), Steven Rubinyi (Senior Disaster Risk Management Specialist), and Calvin Kwon (Nonprofit and Global Organizations, ESRI). • Additional spatial analysis of Ho Chi Minh City was conducted by Luis Andres Herskovic (Assistant Professor at the School of Government of Universidad Adolfo Ibáñez, Chile). • Information about migrant experiences in urban areas was from a report supervised by David Baringo (Senior Social Development Specialist) and Giang Tam Nguyen (Senior Social Development Specialist): “Ethnic Minority Migrants in Major Cities of Vietnam: Challenges and Exclusion Dynamics 2022-2023.” Simon Drought provided editing services. Ha Doan designed the report. Towards more inclusive cities viii ix Viet Nam bi-annual poverty & equity update - June 2024 Overview Part 1. Poverty & Equity Update1 as well as households applying coping mechanisms in the early phase of the pandemic. Viet Nam did not experience a surge Recent poverty and equity dynamics in Viet Nam can be in COVID-19 cases until the spring of 2021, and lockdowns described across three distinct phases (Figure O.1). First, were in place until the autumn. By 2022, Viet Nam ranked at the start of the decade, there was rapid poverty reduction as having the second highest cumulative number of cases per accompanied by declining inequality. From 2010-14, both million people in ASEAN (Our World in Data). Households’ poverty and inequality fell due to large shifts in labor from ability to cope, a reliance on familial and friend networks agriculture into manufacturing and services jobs. Most of for support, use of savings, and smooth consumption also these non-farm jobs were still low-skilled, and agricultural delayed negative poverty trends. laborers could easily shift into these new jobs. This was Figure O1 followed by a period of poverty reduction, but accompanied Figure O.1. Recent changes in poverty in Viet Nam – by slightly rising inequality from 2014-20. During this period, three phases structural transformation continued, but as non-farm wages increased rapidly, farming income did not rise as quickly 6 Pecentage point change in poverty rate 4 and more recently has even declined. Lastly, in 2022, after 2 6 extended adverse impacts related to the COVID-19 pandemic, 0 5 -2 household consumption growth declined compared to 2020, 4 Annualized growth (%) -4 3 and income growth slowed. Dynamics during this distinctive -6 2 and short period from 2020 to 2022 is the focus of analysis in -8 -10 1 Part 1 of this report2. 2010-12 2012-14 2014-16 2016-18 2018-20 2020-22 0 Poverty reduction  overty reduction and P COVID-19 and declining small increase in inequality period -1 A decade of successful poverty reduction stalled during inequality -2 the COVID-19 pandemic3. In Viet Nam, poverty rates Growth Distribution declined in 2020 during the onset of the health crisis, but more negative impacts were seen later during 2021-22. A Note: Upper Middle income class poverty line, ($6.85/day 2017 PPP) stagnation in poverty reduction in 2022 was related to the Figure O4 Source: WB staff calculations using VHLSS. later arrival of COVID-19 outbreaks and severe lockdowns, 45% 100 40% 80 35% Poverty and equity dynamics discussed in this report are based on indicators computed % Population 1 30%by the World Bank for global poverty monitoring (see Annex B). 60 Urban share (%) The World Bank monitors welfare in Viet Nam using household consumption per capita, adjusted to international 2017 PPPs. The official 25% poverty indicator of Viet Nam is different, being a Multi-Dimensional Poverty Index comprised of 12 non-monetary dimensions and one 40 20% monetary dimension based on income (not consumption). The index was constructed by the General Statistics Office of Viet Nam. 15% 2 See World Bank 2022b for a discussion of poverty and equity trends during 2010-20. 20 10% 3 See World Bank 2021 for information about household experiences during COVID-19 in Viet Nam. 0 5% 0% 2010 2012 2014 2016 2018 2020 2022 Towards more inclusive cities 1 Share of LMIC poor Share of UMIC poor Urban (o Share of total population Urban ce Overview Figure O2 Figure O.2. Growth incidence curves, household Understanding income dynamics is key to understanding consumption poverty developments during the pandemic period. 2020-22 In the decade before COVID-19, household income grew 6 strongly at about 6-7 percent per annum and across the 5 entire distribution of households, before stagnating from 4 2020 to 2022. More than 90 percent of household income Annualized growth (%) 3 in Viet Nam is from three sources of labor income: wages, 2 farm production, and non-farm business income. During 1 20-22 the period 2020-22, all three types of labor income were 0 VID-19 negatively impacted either in the extensive or intensive riod -1 margins, meaning there was either a decline in the share of -2 households receiving income, or a fall in incomes among 1-National 2-Urban 3-Rural earners. At the same time, incomes from remittances or social assistance did not rise sufficiently to compensate for declines Source: WB staff calculations using VHLSS. in other income sources during this period. Figure O5 Household 100 consumption, the welfare metric used in this Growth returned post-COVID, but poverty and labor report for global poverty measurement and monitoring, indicators fall short of pre-COVID projections and 80 contracted from 2020 to 2022 in Viet Nam (Figure O.2). expectations. Post-COVID poverty projections show % Population Notably, 60 households spent less on eating out, non-essential renewed progress, but poverty rates remain higher compared food items like ice cream, and non-food items like clothing. to outlooks made before the pandemic. Poverty projections 40 With the avoidance of public spaces, more households made at the initial onset of COVID-19 predicted a stagnation consumed 20 food grown at home. Expenditures on education in poverty reduction in 2021, followed by declines in 2022 and and 0 health also declined as schools reduced hours and people onwards. Actual poverty rates in 2022 were higher than earlier avoided hospitals Official and elective DB care. However, DOU expenses like predictions. Labor income indicators in 2023 are recovering 2022 utilities still increased, especially Viet Nam petrol. Reduced income and to pre-COVID levels, but are still lower than expected oor economic activity C Urban (official) during ore (DB) the period are also seen Suburb (DB) through Town (DB) without COVID (Figure O.3). During the 2015-19 pre- lower Urbangrowth of deposits, center (DOU) retail Urban cluster goods (DOU) and Rural services in 2021 and (official) COVID period, labor incomes grew by nearly 10 percent per most of 2022. Household behavior during and after a crisis year. After a short period of declines in 2020 and 2021, labor typically includes more conservative spending and rebuilding incomes are once again growing, but are at lower levels than if of savings. A larger decline in consumption versus income, pre-COVID income growth trajectories were sustained. suggests both effects of falling income and precautionary savings in Viet Nam. Even with the ongoing recovery, experiences from the pandemic period are useful to reflect on the importance, Households along the entire distribution felt some timing, and complementarity of promotion and impact. Between 2020 and 2022, the share of the middle- protection policies. The ‘Next Mile’ is the journey to class in rural areas stagnated, while in urban areas the decline upper- and high-income country living standards. For a was more pronounced (52 to 43 percent). The poor were society like Viet Nam’s, this means creating more economic still affected, however, since they are vulnerable to even opportunities to build a strong middle class, while at the same small shocks. For households already below the poverty line, time expanding support to low-income and economically worsening conditions may go unmeasured if monitoring vulnerable households. Lessons from the exceptional period only the poverty rate. The poverty gap, indicating the depth from 2020 to 2022 call for the need to strengthen and time 2022 of poverty, also increased slightly in 2022. At the onset of promotion and protection policies. There is potential for (DOU) COVID-19, there were concerns about widening inequality. more shocks in the future that require strong social protection r (DOU) However, due to the nature of the pandemic, household and investments. A changing population profile, in terms of consumption declined relatively more in urban areas than an ageing population and more households located in urban rural areas, and the Gini coefficient slightly declined. areas, also requires updating strategies to tackle vulnerability. 2 Viet Nam bi-annual poverty & equity update - June 2024 Overview Figure O.3. Poverty trends and projections Note: Actual poverty rates are bi-annual, and poverty projections are annual. Projections made using March and October 2021 growth forecasts start in 2021. Poverty projections made using February 2024 growth forecasts start in 2023. Poverty projections based on methods described in Lakner, Mahler, Negre, and Prydz (2020). Source: World Bank staff calculations using VHLSS. As Viet Nam continues eliminating structural factors related Part 2. Poverty and Inclusion in to extreme poverty, future prosperity will be more interlinked Urban Areas and Cities with economic developments and the creation of good jobs. The structural transformation of the labor force, out of Experiences in Viet Nam during the pandemic agriculture and into manufacturing and services sectors, has highlighted challenges and the need for an increased been the main channel of upwards economic mobility for focus on social conditions in urban areas and cities. Due households. However, the transformation is not complete. to dense populations and shared public spaces, the health Moreover, the vulnerability of certain jobs was highlighted crisis spread faster, and lockdowns were more severe in urban during the pandemic, especially informal jobs and those settings. Disruptions in cities led to larger negative aggregate in urban areas. In the medium- and long-terms, despite a economic impacts, since they are commercial hubs and centers return of growth, policy makers should continue monitoring of economic activity. In addition, other types of shocks, such household developments. Global and regional economic as environmental disasters and flooding can lead to larger outlooks for 2024 are also subdued. As households rebound, damage in urban areas where there are more buildings and it is important to monitor the recovery to identify which infrastructure. The concept of the “urbanization of poverty” groups display longer-term scarring effects compared to has regained significance post-COVID. From 2010 to 2022, others. In particular, there are three groups and conditions to the urban population in Viet Nam increased from 30 to watch closely: 1) the already poor and if they become worse almost 40 percent (Figure O.4). At the same time, the share off, 2) the economically insecure who fall back into poverty, of the poor (based on World Bank poverty lines) located in and 3) the economically secure who may become insecure. urban areas also increased. As discussed in Part 1, the relatively Towards more inclusive cities 3 Overview larger adverse impacts in urban areas occurred recently during Viet Nam is rapidly urbanizing, and using different 2020-22, related to sharper shocks and economic disruptions measurement techniques, the share of the population experienced during the pandemic. living in an urban setting is even higher. In 2022, the urban share of the population was about 39 percent, but Some development challenges are more urgent in urban some research has shown that the urbanized population could areas than rural ones. For example, crowded housing be much larger. With the availability of satellite imagery, an conditions, higher living costs, congested traffic, lack of mobility urban area can be more consistently classified by its greenery, and access to jobs, or air pollution are more likely to be risks built-up area, or population density features. For example, Figure Figure O1 O1 Figure Figure O2 O2 and challenges to populations in urban areas than rural areas. using the Degree of Urbanization (DOU) methodology, Much of the urban growth in developing countries is referred the urban population share was modeled to be as high as 73 to as “pancake spread”4, where development is low lying. Cities percent (Figure O.5). The share of the urban population that in developing countries tend to expand flat and horizontally, reside in the urban center is lower than the official urbanization 6 but 6vertical layering is what is needed for agglomeration and rate, indicating that rapidly urbanizing 2020-22 suburbs and towns Pecentage point change in poverty rate 2020-22 Pecentage point change in poverty rate 4 4 to make 2 2 modern cities more productive. In Viet Nam, rapid are likely 6 6 to still be considered rural in official classifications. 5 urbanization 0 0 has led to sprawling development, declining Globally, 5 there is a strong correlation between urbanization, -2 -2 4 returns to agglomeration, lower labor productivity, and GDP, 4 and poverty. Higher urbanization rates are correlated Annualized growth (%) Annualized growth (%) -4 -4 3 3 increasing -6 -6 congestion in major cities . Higher urban poverty 5 with both higher levels of economic development and lower -8 2 2 rates-8 can occur when crowding or congestion outweigh benefits poverty rates. Viet Nam’s official urbanization rate is low -10 -10 1 1 from density 2010-12 and 2010-12 agglomeration. 2012-14 2012-14 2014-16 2014-16 2016-18 2016-18 2018-20 2018-20 2020-22 2020-22 given its GDP and poverty levels, but a higher urbanization 0 0 Poverty Poverty reduction reduction Poverty Poverty reduction reduction and and COVID-19 COVID-19 rate is more -1 in line with global averages. and declining and declining small small increase increase in inequality period in inequality period -1 inequality inequality -2 -2 Distribution GrowthDistribution Growth 1-National 1-National 2-Urban 2-Urban 3-Rural 3-Rural Figure O.4. Urban share of the poor and population Figure O.5. Share of urban population in Viet Nam O4 O4 Figure Figure 2022, various O5 O5 Figure classifications Figure 45% 45% 100 100 40% 40% 80 80 35% 35% % Population % Population 30% 30% 60 60 Urban share (%) Urban share (%) 25% 25% 40 40 20% 20% 15% 15% 20 20 10% 10% 0 0 5% 5% Official Official DB DB DOUDOU 0% 0% 2010 2010 2012 2012 2014 2014 2016 2016 2018 2018 2020 2020 2022 2022 Viet Nam Viet Nam ShareShare poor poor of LMIC of LMIC ShareShare of UMIC of UMIC poor poor Urban rban (official) Core C U(official) ore (DB) (DB) (DB) (DB) Suburb Suburb T(DB) Town own (DB) ShareShare of population of total total population UrbanUcenter rban center (DOU) (DOU) UrbanUcluster rban cluster (DOU) RuralR (DOU) ural (official) (official) Source: WB staff calculations using VHLSS 2022 and applying Dartboard (DB) Source: WB staff calculations using VHLSS. and Degree of Urbanization (DOU) classifications from Nakamura et al.(2023) O6 O6 Figure Figure 30 30 4 Pancake development is low and outward, while pyramid development expands outwards but also fills in pockets in urban centers and notably 25 25 with tall buildings. 5 For a 20 20review of urbanization trends from a planning perspective, see World Bank 2020b Urbanization at a Crossroads, and World Bank 2011 Viet Poverty rate (%) Poverty rate (%) Nam Urbanization Review. 15 15 10 10 4 Viet Nam bi-annual poverty & equity update - June 2024 5 5 0 0 6 2020-22 Pecentage point change in poverty rate 4 6 Overview 2 0 5 -2 4 Annualized growth (%) The -4rate of urbanization is an important consideration -6 to reduce3 urban poverty. Across different types of urban areas, for policy -8 direction in a Next Mile context6. Due to the urban2 centers consistently have the lowest poverty rates. well-designed -10 social and poverty reduction policies, Viet However, 1 poverty rates outside the center are much higher. For 2010-12 2012-14 2014-16 2016-18 2018-20 2020-22 0 following the DOU method, 2022 UMIC poverty had reduction Nam has Poverty great success Pin tackling poverty overty reduction and through COVID-19 example, -1 and declining and targeting rural development ethnic small increase period and minorities in inequality rates in urban clusters were about 18 percent, compared to less inequality -2 mountainous areas. In particular, National Targeted than 5 percent in the urban center (Figure O.6). In addition, Programs (NTPs) have been Growth quite successful. However, Distribution of the urban center about 70 percent1-National 2-Urban population is in the 3-Rural less and less of Viet Nam’s population is reliant on rural Red River Delta and Southeast regions where the capital livelihoods and farm income, with more living in urban Hanoi and the southern economic engine Ho Chi Minh City Figure O4 Figure O5 areas. As more people live in urban areas and cities, and as (HCMC) are located, respectively. Understanding the degree the country 45% becomes denser, conditions in urban areas will of urbanization 100 is important, as whether people live in urban have a larger role to play in poverty reduction and upward 40% or rural areas is meaningful for policies and strategies if these 80 35% mobility. Sustaining poverty reduction will require economic areas reflect distinct patterns of development. But, there are 30%management and provision of services in ever denser effective 60 differences in living standards between those who live clearly % Population Urban share (%) and crowded 25% areas. in a city center versus a suburb or town. 40 20% Urban15%areas are generalized as wealthier with low Due 20to the heterogeneity of areas that are “urban” and 10% but there is a gradient of poverty across levels poverty, the structure of governance, city-level analysis can be an 0 of urbanization. 5% There are different degrees of urbanization important way to study urban livelihoods. From a policy Official DB DOU – ranging 0% from urban centers or cores to suburbs, clusters, perspective, it can be valuable to examine living standards 2010 2012 2014 2016 2018 2020 2022 Viet Nam towns – that can hide heterogeneity in living conditions of by municipalities rather than urban areas, in general. There Share of LMIC poor Share of UMIC poor Urban (official) Core (DB) Suburb (DB) Town (DB) “urban” areas and confound the extent, nature, and solutions are five city-level municipalities in Viet Nam (Can Tho, Urban center (DOU) Urban cluster (DOU) Rural (official) Share of total population Danang, Haiphong, Hanoi and HCMC), accounting for about 40 percent of the official urban population in 2019. Figure O.6. There is a gradient of poverty across Many policies are made, and resources are managed at the urbanization categories Figure O6 municipality-level. Thus, in some cases, city-level analysis would also be more useful and practical than nationwide 30 urban analysis for planning and decision-making. When 25 discussing the benefits from agglomeration, challenges to 20 service delivery or financing, these conversations also make Poverty rate (%) sense at a city level. 15 10 A principal difference between rural and urban areas is population density, meaning more granular data 5 is needed to adequately analyze urban conditions. 0 Granular within-city analysis is usually more challenging 2018 2020 2022 due to data constraints. But when feasible, analysis is much Urban (official) Core (DB) Urban center (DOU) more informative about the distribution of socio-economic Suburb (DB) Town (DB) Urban cluster (DOU) conditions. As a case study, spatial within city-level analysis of HCMC, as the largest municipality of Viet Nam, was Note: World Bank Upper middle income poverty line is $6.85/day 2017 PPP, conducted for this report. Within-city analysis revealed that using 2018 classifications. concentrations of informal housing were similar to areas Source: WB staff calculations using classifications from Nakamura et al. 2023. with concentrations of recent migrants, and broad flood 6 See the World Bank 2022b Viet Nam Poverty & Equity Assessment – From the Last Mile to the Next Mile, for discussion of Last Mile and Next Mile framework. Towards more inclusive cities 5 Overview risk, among other trends. HCMC is a good example of a and to support successful urban development strategies. To diverse metropolis with a gradient of rural-to-urban areas. better inform urban social policies for the Next Mile, granular By official accounts, almost 80 percent of the population data is needed to design policies that are more targeted, rather in HCMC reside in its urban districts, with the remainder than area-based covering large domains common in rural in rural districts. However, even in its urban districts, there areas. Urban policies require more precision for determining are variations in housing characteristics, education and transportation planning, location of schools and hospitals as demographics. This is especially the case with large-sized well as allocation of budgets. districts. In the case of HCMC, several districts are home to more than half a million people each. With the availability of Preserving gains and developing opportunities in urban city-level data, household conditions in HCMC are described areas are important for Viet Nam’s Next Mile to upper- with added granularity. middle and high-income country status. Over the last decade, rapid economic growth was broadly inclusive and Viet Nam has a high population density, and the livelihoods in Viet Nam improved dramatically and in a quality of urbanization is more important than the progressive manner. Given the substantial success in poverty extent of urbanization for continued development to reduction, the poverty and equity agenda in Viet Nam today high-income status. Benefits from urbanization can be less is no longer only about raising minimum living standards impactful if urban areas are not well planned, public services and tackling extreme poverty. It is also about creating new are inadequate for growing populations, or there is inequity in and sustainable economic pathways for a more aspirational access to services and utilities. While analyses more commonly population. Development and economic opportunities in discuss aggregate conditions of urban areas, and comparisons urban areas are key to sustaining upwards economic mobility between urban and rural areas due to more available data, for the millions who have left poverty and now seek even within-city analysis is key to ensuring that cities are livable higher economic gains. 6 Viet Nam bi-annual poverty & equity update - June 2024 PART 1.  PART 1. Poverty & Equity Update 7 7 The dynamics discussed in this report are based on poverty and equity indicators computed by the World Bank for global poverty monitoring, including Sustainable Development Goal 1 (see Annex B). The World Bank monitors welfare in Viet Nam using household consumption per capita, adjusted to international 2017 PPPs. Viet Nam’s official poverty indicator is a multi-dimensional indicator comprised of 12 non-monetary dimensions and one monetary dimension based on income (not consumption). Trends based on the World Bank’s monetary consumption-based measures of poverty and equity will differ from non-monetary measures or measures using income instead of consumption. Monitoring is bi- annual due to the availability of expenditure data in the Viet Nam Household Living Standards Survey (VHLSS) (see Annex B for a summary of this method). 8 June 2024 - Viet Nam bi-annual poverty & equity update Part 1 discusses recent short-term poverty and equity dynamics (2020-22), anchored in longer-term trends. Globally, the COVID-19 pandemic resulted in unprecedented increases in poverty. Viet Nam’s progress in poverty reduction was also affected. The first section discusses changes in 2022’s poverty rates and inequality. Incomes are the key to household poverty dynamics, and section 2 examines recent trends among labor and non-labor income sources, as well as the role of each in changes in poverty. Lastly, section 3 highlights projections and outlooks. Dynamics in 2022 were unique given the health crisis, but likely to be short-lived as the Vietnamese economy recovers. However, the experience provides important lessons on balancing and timing fiscal and social policies that can support households in the short- and long-terms. 1. Poverty and equity developments 2. Sources and dynamics behind changes in poverty 3. Looking forward Towards more inclusive cities 9 PART 1. POVERTY & EQUITY UPDATE 1. Poverty and equity Pandemic-related slowdowns developments interrupted long-run poverty and inequality trends in the East Asia A decade of successful poverty and Pacific region reduction in Viet Nam stalled In the East Asia and Pacific (EAP) region, where poverty during the COVID-19 pandemic reduction is strongly linked to economic growth, Recent poverty reduction trends in Viet Nam can be slowdowns during COVID-19 impacted the pace of summarized in three distinct phases, with Part 1 of poverty reduction, including in Viet Nam (Figure 2, this report focusing on the third exceptional phase panel A). Among the larger developing economies in the during 2020-228. Changes in poverty can be decomposed region, Viet Nam has historically stood out with high rates into different components, with Figure 1 illustrating changes of poverty reduction and economic growth. However, the in poverty by growth and redistribution factors. In the first pandemic was an unprecedented shock to the global economy phase, at the start of the decade, there was rapid poverty that stalled progress in both, and in some cases, trends were reduction accompanied with declining inequality. From even reversed (World Bank, 2020a, 2021a, 2022a, 2023). 2010-14, both poverty and inequality fell due to large shifts In the high-growth EAP region, poverty increased in some in labor from agriculture into off-farm manufacturing and countries, or progress slowed where poverty reduction was services jobs. Most of these non-farm jobs were low-skilled, once rapid (Figure 2, panel B). Changes in poverty dynamics and agricultural laborers could easily shift into these new were more common at higher poverty lines, since COVID-19 jobs. This was followed by a period of continued poverty tended to impact urban areas more than rural ones, and reduction, but accompanied by a slight rise in inequality from households in the middle and upper ranges of the distribution 2014-20. During this second phase, structural transformation felt relatively more adverse impacts. For example, based on continued, but as non-farm wages increased rapidly, farming the World Bank’s Upper-Middle Income Country (UMIC) income did not rise as quickly and more recently has been poverty line valued at $6.85/day 2017 PPP, poverty reduction declining. Lastly, in 2022, after extended adverse impacts paused in China, Indonesia, the Philippines, and Viet Nam. related to COVID-19, poverty reduction stalled as household consumption declined, and income growth tapered. Figure 1.Recent Figure 1. Recent changes changes in poverty in poverty in Viet in Viet Nam Nam, 3 phases phases – three Lower Middle income class poverty line Upper Middle income class poverty line ($3.65/day 2017 PPP) ($6.85/day 2017 PPP) Growth Distribution Growth Distribution Percentage point change in poverty rate Percentage point change in poverty rate 1 4 0.5 0 2 -0.5 0 -1 -1.5 -2 -2 -4 -2.5 -3 -6 -3.5 -8 -4 -4.5 -10 2010-12 2012-14 2014-16 2016-18 2018-20 2020-22 2010-12 2012-14 2014-16 2016-18 2018-20 2020-22 Poverty reduction and  overty reduction and P COVID-19 Poverty reduction and P  overty reduction and COVID-19 declining inequality small increase in inequality period declining inequality small increase in inequality period Source: WB staff calculations using VHLSS. 8 See World Bank 2022b. Viet Nam Poverty & Equity Assessment – From the Last Mile to the Next Mile, for discussion on dynamics from 2010-20. 10 Viet Nam bi-annual poverty & equity update - June 2024 1. Poverty and equity developments  Figure 2. Strong poverty reduction and links to economic growth Notes: WB Upper-Middle income country (UMIC) poverty rates ($6.85/day 2017 PPP). Welfare is measured using household consumption per capita. Breaks indicate changes in methodology and non-comparability across the data series. The regional series ranges from 1985-2022, data availability varies by country. Source: World Bank Poverty & Inequality Platform (PIP), World Development Indicators (WDI). Economic growth has a less consistent relationship Urban areas were mainly affected by lockdowns, and Viet with inequality, than it does with poverty (Figure Nam’s export-intensive sectors also suffered from lower global 3, panel A). Large volumes of literature have studied the demand. Comparing incomes from Q1-2020 to Q1-2023, relationship between inequality and economic growth, most rural incomes showed higher growth than urban ones. with inconclusive results (World Bank, 2006). Most large economies in EAP do not have a Gini coefficient above 0.4, In 2022, progress in poverty which is the World Bank’s threshold for high inequality. reduction stalled9 While inequality was rising slowly in Viet Nam before the pandemic, it was low compared to other large EAP economies In 2022, progress in poverty reduction stalled based on (Figure 3, panel B). At the outset of COVID-19, there were poverty measurement using household consumption. concerns about widening inequality in both monetary and Poverty reduction between 2018 and 2020 were statistically non-monetary indicators (World Bank, 2021a). Yet in 2022, significant, but changes between 2020 and 2022 are not. At inequality measured using household consumption declined the higher Upper-Middle income (UMIC, $6.85/day 2017 from 0.37 in 2020 to 0.36 in 2022. The decline in inequality PPP) poverty line, there is a larger absolute change in poverty is attributable to larger negative impacts experienced by those due to the nature of the pandemic, and who was more likely in the middle and upper ranges of the welfare distribution. to be negatively impacted. 9 For global poverty monitoring, the World Bank measures poverty for Viet Nam based on household consumption per capita. Household reporting in 2022 may still reflect some residual experiences and spending in 2021 due to recall periods of consumption questions. However, this is not unique to the 2022 survey year, and applies across all survey years (see Annex B). Towards more inclusive cities 11 PART 1. POVERTY & EQUITY UPDATE Figure 3. Inequality in EAP, a varied story Notes: Welfare is measured by household consumption. Breaks indicate changes in methodology and non-comparability in the poverty data series. The regional series ranges from 1985-2022, data availability varies by country. Source: World Bank Poverty & Inequality Platform (PIP), World Development Indicators (WDI). Figure 4. 30 years of poverty Figure 5. If your household stopped trends in Viet Nam, 1992-2022 earning income today, how long could you Lockdowns in urban and on? survive areas factory closures during Figure 4. 30 years of poverty trends in Viet Nam, 1992-2022 waves of COVID-19 more negatively impacted households 100 4500 in the middle and higher ends of the income distribution. 90 4000 100most severe lockdown occurred in the southern commercial The GDP per capita (constant 2015 US$) 80 3500 hub90 of HCMC in the autumn of 2021. Labor income from 28 29 80 70 3000 wages is the primary source of income growth for households. Share of households (%) 70 Poverty rate (%) 60 2500 The health crisis 14slowed growth in this income 12 source and 50 60 2000 coupled 50 with continuing declines in farm production income 40 24 1500 among 40 32 near-poor, an aging population and low the poor and 30 20 1000 amounts 30 of transfers to compensate for falls in labor income, 10 500 consumption-based 20 poverty reduction stagnated. 35 26 0 0 10 Delayed negative 2021impacts in 2022 were related to the 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 2020 2022 0 December - January 2022 April - May 2022 Extreme poverty ($2.15/day 2017 PPP) LMIC PL ($3.65/day 2017 PPP) later arrival ofLessCOVID-19 than one month and subsequent 1-3 months lockdowns, UMIC PL ($6.85/day 2017 PPP) GSO Multi-dimensional poverty GDP per capita (constant 2015 US$) as well as households 3-6 months applying copingMore than mechanisms 6 months in the early phase of the pandemic . While other nations were 10 Note: WB poverty rates are measured using household consumption per capita. The World Bank monitors global poverty using three poverty lines, for relevance across a recovering or adjusted to handling outbreaks, Viet Nam did range of countries and contexts. First, the World Bank’s International poverty line not experience its largest COVID-19 surge until the spring of ($2.15/day 2017 PPP) is used for monitoring SDG 1.1. The Lower-Middle ($3.65/ 2021. By the end of 2022, Viet Nam had the second highest day 2017 PPP) and Upper-Middle ($6.85/day 2017 PPP) income country poverty lines were derived as the median of national poverty lines from countries in these income groups to reflect minimum needs in those contexts (Jolliffe and Prydz 2016). 10 See World Bank (2021a) about household experiences during Source: WB staff calculations, WDI, PIP, GSO. COVID-19 in 2020 and 2021. 12 Viet Nam bi-annual poverty & equity update - June 2024 1. Poverty and equity developments  cumulative number of cases per million in the ASEAN may not emerge until coping strategies are exhausted, or region (Our World in Data). Another potential reason for income experience significantly large declines. Despite most delayed impacts could be related to a lag between household lockdowns occurring in late 2021 and borders reopening in conditions and economic cycles due to households’ ability to March 2022, a larger share of urban households reported only Figure cope and smooth in bad times. stopped 5. If your household consumption Negative trends having sufficient savings for less than one month’s spending earning income today, how long could in the spring of 2022 compared to late 2021 (Figure 5). you survive on? Figure 5. If your household stopped earning income today, how long could you survive? Household impacts varied, with 4500 urban areas more negatively 4000 100 impacted than rural ones GDP per capita (constant 2015 US$) 90 3500 28 29 80 Due to the types of economic shocks that occurred during 3000 Share of households (%) 70 2500 14 12 the pandemic and the localization of economic impacts, 60 2000 urban areas where more negatively impacted than rural 50 24 ones. During the 2020-22 period, households experienced 1500 40 32 1000 30 historically low consumption growth compared to the last 500 20 35 decade (Figure 6). Across the distribution, welfare as measured 26 by household consumption per capita declined more in 0 10 urban households than rural households. Nationally, average 2020 2022 0 December 2021 - January 2022 April - May 2022 day 2017 PPP) Less than one month 1-3 months consumption growth was negative, with larger declines at the sional poverty 3-6 months More than 6 months top rather than at the bottom of the distribution. Changes in Note: Urban households only. poverty were more pronounced at the UMIC than the LMIC Source: World Bank Viet Nam COVID-19 household phone surveys. poverty line, also reflecting that negative impacts were felt more along the middle and higher ends of the distribution (Figure 7). The share of UMIC poor from urban areas rose to 20 percent in 2022, compared to 13 percent in 2020. Figure6. Figure Growthincidence 6.Growth curves,household incidencecurves, householdconsumption consumption Figure 6. Growth incidence curves, household consumption 2010-22 2010-22 2020-22 2020-22 55 55 44 44 33 33 (%) growth (%) (%) growth (%) Annualized growth Annualized growth 22 22 Annualized Annualized 11 11 00 00 -1 -1 -1 -1 -2 -2 -2 -2 1-National 1-National 2-Urban 2-Urban 3-Rural 3-Rural 1-National 1-National 2-Urban 2-Urban 3-Rural 3-Rural Note: Welfare is household consumption per capita in 2017 PPP. Source: WB staff calculations using VHLSS. Towards more inclusive cities 13 PART 1. POVERTY & EQUITY UPDATE Within urban areas, the bottom-40 and ethnic minorities shock (a negative event that reduced household income), while were more likely to report lower economic stability than having less savings and safety nets than better-off households. better-off households. As COVID-19 cases increased to For example, in April 2022, 47 percent of the bottom-40 record levels in 2021, vulnerable groups within urban areas in urban areas reported that if their incomes stopped, they were the most strapped for income. During 2022, vulnerable would only have sufficient savings to survive for less than groups in urban areas were more likely to report experiencing a one month, compared to 33 percent of the top-60 (Figure 8). Figure 7. Poverty rates by urban and rural area, 2010-2022 Figure 7. Poverty rates by urban and rural area, 2010-22 LMIC poverty rate ($3.65/day 2017 PPP) UMIC poverty rate ($6.85/day 2017 PPP) 60 60 50 50 Poverty rate (%) Poverty rate (%) 40 40 30 30 20 20 10 10 0 0 2010 2012 2014 2016 2018 2020 2022 2010 2012 2014 2016 2018 2020 2022 Rural Urban Rural Urban Note: Welfare is household consumption per capita in 2017 PPP. Source: WB staff calculations using VHLSS. Figure 8 Figure 8. How long can households economically survive if their incomes stopped today, by bottom-40 and top-60 Bottom-40 Top-60 100% 100% 90% 20 90% 26 29 31 80% 80% 12 70% 10 70% Share of households (%) Share of households (%) 14 12 60% 60% 21 50% 27 50% 25 40% 40% 33 30% 30% 47 20% 38 20% 33 10% 10% 24 0% 0% December 2021 - January 2022 April-May, 2022 December 2021 - January 2022 April-May, 2022 Less than one month 1 - 3 months Less than one month 1 - 3 months 3 - 6 months More than 6 months 3 - 6 months More than 6 months Note: Urban samples only. Source: World Bank COVID-19 household monitoring surveys, Rounds 6 and 7. 14 Viet Nam bi-annual poverty & equity update - June 2024 1. Poverty and equity developments  Figure 9 Figure 9. Distribution of the population, by economic classes Rural areas Urban areas 100% 100% 80% 80% Share of the population (%) Share of the population (%) 60% 60% 40% 40% 20% 20% 0% 0% 2010 2012 2014 2016 2018 2020 2022 2010 2012 2014 2016 2018 2020 2022 Middle-class Economically secure Economically insecure Middle-class Economically secure Economically insecure Near poor Extreme poor Near poor Extreme poor Note: See Box 1.A for definitions of economic class. Source: WB staff calculations using VHLSS. The increase in economic insecurity from December 2021 to surveys, round 7). At the UMIC poverty line ($6.85/day 2017 April-May 2022 was similar in both the bottom-40 and top-60 PPP), poverty rates in the Southeast region climbed from 0.28 groups. From December 2021 to April-May 2022, the share percent in 2020 to 1.06 percent in 2022. of those who could survive less than one month increased by an additional 9 percentage points in both groups. The medium-term increases in poverty in the Mekong Delta region are statistically significant and likely due to The size of the middle-class declined in 2022. The size more than just economic factors related to COVID-19. of the middle-class declined both in absolute population as The rise in poverty from 2018 to 2022 – based on the higher well as a share of the population (Figure 9). The share of the World Bank UMIC poverty line – is related to longer-term middle-class declined in both rural and urban areas, but much challenges in this southern region, also known as Viet Nam’s more in urban areas. Between 2020 and 2022, the share of the rice basket. The stagnation and reversal of poverty reduction middle-class in rural areas stagnated, while in urban areas it fell mid-decade is linked to lower household farm production from 52 percent in 2020 to 43 percent in 2022. The change in incomes that were not offset by growth in other labor income the structure of economic classes also varied by regions. The sources, especially for households in the middle of the largest decline in the middle-class in urban areas was in the distribution. In 2020, severe droughts coincided with a non- Southeast region where HCMC is located. In this region, 65 statistically significant increase in regional poverty rates. In percent of residents were middle-class in 2020, dropping to 53 2022, poverty rates in the Mekong Delta rose a second time. percent in 2022. While poverty rates also rose slightly in 2022 in some other regions due to pandemic-related factors, the continuing Differences in poverty trends rise in the Mekong is likely also due to persistent challenges across geographic regions are related to climate change and farming, that are not as severe in other regions. In a labor-intensive agricultural-reliant evident region, impacts from climate change, aging workforce and Poverty rates in 2022, based on both the World Bank out-migration are culminating into measurable increases in LMIC and UMIC poverty lines, stagnated across most poverty and vulnerability, lower household farm production regions (Figure 10). HCMC, located in the Southeast income, and a shrinking labor force in the agriculture sector. region, experienced strict lockdowns in 2021 that were later While the agricultural labor force is contracting throughout lifted in October. Residual effects from these restrictions the country, there are larger negative impacts in the Mekong lasted into 2022. In the Southeast region, from April-May Delta due to both the importance of agriculture and 2022, as many as 43 percent of urban households reported relatively weaker non-farm opportunities compared to other that if they stopped earning incomes, their savings would last regions in Viet Nam. for less than one month (World Bank COVID-19 household Towards more inclusive cities 15 PART 1. POVERTY & EQUITY UPDATE rates 10 Figure Figure 10. Poverty by region, 2010-22 LMIC poverty rate ($3.65/day 2017 PPP) UMIC poverty rate ($6.85/day 2017 PPP) 70 70 60 60 50 50 Poverty rate (%) Poverty rate (%) 40 40 30 30 20 20 10 10 0 0 2010 2012 2014 2016 2018 2020 2022 2010 2012 2014 2016 2018 2020 2022 Midlands and Northern Mtns Red River Delta Midlands and Northern Mtns Red River Delta Northern and Coastal Center Central Highlands Northern and Coastal Center Central Highlands Southeast Mekong Delta Southeast Mekong Delta Note: Welfare is household consumption per capita in 2017 PPP. Source: WB staff calculations using VHLSS. The vulnerable have fewer coping At higher poverty lines, the gap in poverty rates strategies and experienced between ethnic minorities and the majority Kinh widened. The share of the population in ethnic households worsening conditions declined slightly from 16 percent in 2010 to 13 percent For the poor, even small changes in consumption can in 2022. Despite strong poverty reduction among ethnic lead to falls deeper into poverty or result in insufficient minorities over the long-run, ethnic minority poverty consumption and nutrition. Thus, any amount of reversal rates are still much higher than among the Kinh. Their in progress is of concern. The vulnerable were still affected chronically higher poverty rates can be attributed to higher during COVID-19, since they can be vulnerable to even rates of participation in farming, residence in more remote small shocks, while richer households have better coping areas, as well as lower education and literacy in the majority mechanisms. For poor households, primary strategies to cope Vietnamese language. At the LMIC poverty line ($3.65/day with negative shocks and lower incomes during the health 2017 PPP), the poverty rate of the Kinh has remained under crisis included reducing food and non-food consumption 2 percent since 2016, while the ethnic minority poverty rate or accepting assistance from either non-governmental has exceeded 20 percent. At the UMIC poverty threshold, organizations (NGOs) or the government. Tapping savings poverty rates among the Kinh rose – but by a smaller tended to be a coping strategy more readily available to richer amount compared to the nearly 4 percentage point increase households. As a strategy to smooth consumption, relying experienced by ethnic minorities (Figure 11). on savings is less likely for poor households that tend to lack assets and savings as well as access to formal banking and Changes in the poverty rate do not account for poor saving channels (World Bank, 2021a). Not only did monetary- households that fall further. For households that are based poverty increase, learning losses occurred and longer- already below the poverty line, worsening conditions may go term household economic mobility may have been affected unnoticed since the poverty rate (as a measure) reflects the as households invested less in education or family businesses share of the population who live below the poverty line, but (OECD, 2023; World Bank, 2021a). not how far the poor are from the poverty line. 16 Viet Nam bi-annual poverty & equity update - June 2024 1. Poverty and equity developments  Figure Figure 11. Ethnic minorities experienced a larger 11 and relative increase in poverty than the Kinh absolute LMIC poverty rate ($3.65/day 2017 PPP) UMIC poverty rate ($6.85/day 2017 PPP) 90 90 80 80 70 70 60 60 Poverty rate (%) Poverty rate (%) 50 50 40 40 30 30 20 20 10 10 0 0 2010 2012 2014 2016 2018 2020 2022 2010 2012 2014 2016 2018 2020 2022 Ethnic minority Kinh majority Ethnic minority Kinh majority Source: World Bank staff calculations using VHLSS. Figure 1212 Figure Figure 1313 Figure Figure 12. Average income gap to the poverty line Figure 13. The prosperity gap, 2010-22 0.9 0.9 6 6 0.8 0.8 5 5 0.7 0.7 Average income gap (2017 PPP) Average income gap (2017 PPP) 0.6 0.6 4 4 Prosperity Gap Prosperity Gap 0.5 0.5 3 3 0.4 0.4 0.3 0.3 2 2 0.2 0.2 1 1 0.1 0.1 0 0 0 0 2010 2012 2010 20142016 2012 2014 2016 2018 2018 2020 2020 2022 2022 2010 2012 2010 2012 2014 20142016 2018 2020 2016 2018 2020 2022 2022 IPL IPL LMIC LMIC UMIC UMIC Rural Rural Urban Urban VietViet Nam Nam Note: Among the poor defined by various poverty lines, the average income gap Note: Household consumption per capita in 2017 PPP. to the poverty line. Source: World Bank staff calculations using VHLSS. Source: World Bank staff calculations using VHLSS. The poverty gap is a useful additional source of The global prosperity gap11 is a new World Bank information to measure the average gap from the poverty measure that monitors how much income or line. Across all three poverty lines, the gap increased in 2022 consumption would have to multiply to reach a set (Figure 12), indicating that the depth of poverty is higher as well. prosperity standard. Intuitively, the gap is how much 11 See Box 1.A. for information about the prosperity gap. Towards more inclusive cities 17 PART 1. POVERTY & EQUITY UPDATE household incomes or consumption would have to multiply In Viet Nam, household consumption was adversely to reach typical minimum thresholds in high-income impacted at both ends of the distribution, and countries, $25 per day per person (2017 PPP) (Kraay et inequality declined. Household consumption declined al., 2023). In 2010, household consumption in Viet Nam along the entire welfare distribution, but relatively more would have had to multiply by four to reach the prosperity for those at the top of the distribution. Average household standard. One desirable feature of the prosperity gap is that it consumption among the top 10 of the population was 10- gives more weight to dynamics among the poorest. Progress 11 times greater than the bottom 10. In 2022, the drop in in narrowing the gap was consistent throughout the decade, consumption of the top 10 percent was more pronounced and the gap reduced to 2.6 in 2020. However, momentum than among the bottom 10 percent. Absolute gaps have towards poverty reduction and narrowing the prosperity gap been widening up to 2020, before reducing again during both stalled during COVID-19 (Figure 13). 2022 (Figure 14), driven primarily by the larger reduction in consumption among households at the top of the Inequality declined slightly as distribution. Inequality in Viet Nam is still considered not households at the higher ends high, with 0.4 the World Bank threshold for high inequality. During the period of interest in this report, inequality of the distribution experienced declined slightly from 2020 to 2022 (Figure 15). relatively more adverse impacts than those at the bottom, particularly those in urban areas Figure 14. Absolute changes in household Figure 15. Inequality trends Figure 14 consumption Figure 15 50 0.4 45 Consumption (per capita per day, 2017 PPP) 40 0.3 35 30 Gini index 25 0.2 20 15 0.1 10 5 0 0 2010 2012 2014 2016 2018 2020 2022 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 2020 2022 Bottom 10 Bottom 40 Top 60 Top 10 Note: Real household consumption per capita. Note: Gini Index. Source: WB staff calculations using VHLSS. Source: World Bank staff calculations. 18 Viet Nam bi-annual poverty & equity update - June 2024 1. Poverty and equity developments  Figure 16 Figure 16. Decomposition of the change in inequality 0.04 0.04 0.03 0.03 0.02 0.02 0.01 0.01 0.00 0.00 -0.01 -0.01 -0.02 -0.02 -0.03 -0.03 -0.04 -0.04 -0.05 -0.05 -0.06 -0.06 -0.07 -0.07 Urban Region Ethnicity Urban Region Ethnicity Urban Region Ethnicity and rural and rural and rural 2010-14 2014-20 2020-22 Changes in within group inequality Changes in group population shares Changes in the relative average income between groups Total change Note: Household consumption per capita. Mookherjee & Shorrocks (1982) decomposition of GE(0). Source: World Bank staff calculations. Figure 16 illustrates the sources of changes in inequality The sources of changes in inequality highlight the over three phases. There are clear differences in inequality disparate nature of who was impacted by COVID-19 and trends as measured by the generalized entropy index12. the economic slowdown. In the period 2020-22, the relative Changes in inequality can be decomposed into three effects: income effect by urban/rural areas was inequality reducing, 1) changes in the relative average income between groups, 2) pointing to a convergence in average consumption between changes in group population shares and 3) changes within urban and rural areas, much larger than the convergence group inequality. In the first two phases and for the three between regions. While some convergence in relative incomes types of groups being analyzed (area, region, and ethnicity), between geographic groups was apparent, relative incomes both the first and third effects move in the same direction, between ethnic minority and Kinh groups widened. This is contributing to net increases or decreases in inequality. consistent with the widening gap in poverty rates by ethnicity However, for the period 2020-22, these components moved from 2020 to 2022, shown earlier in Figure 11. in different directions, counterbalancing each other, and there was a smaller decline in inequality. The population shares played no significant role in any period, except by ethnicity in the third period. 12 The generalized entropy index is also a measure of dispersion, but is decomposable. As such, it is used in this report here instead of the Gini Index. Towards more inclusive cities 19 PART 1. POVERTY & EQUITY UPDATE Box 1.A. World Bank monitoring indicators Poverty measurement for global monitoring There are many concepts of poverty, each with its own merits. The World Bank’s bi-annual Poverty and Shared Prosperity Report, Piecing Together the Poverty Puzzle, expanded the menu of indicators used by the institution to measure globally comparable poverty rates, including higher absolute poverty lines, relative poverty lines, and a multidimensional poverty measure (World Bank, 2018a). Countries also set their own national poverty lines, determined by national governments that are best suited to a country’s conditions. Some countries explicitly measure happiness, such as in Bhutan, or poverty based on a consensus measure, as in Tonga. In many developing countries, the basis of monetary poverty measurement is household consumption rather than household income. This is often chosen because for many poor households, informal incomes can be lumpy and look quite different when measured in one week compared to another, varying with the agricultural season, the amount of business a home enterprise conducts, or the number of hours of casual labor they can sell. Consumption is often more regular, as savings or borrowing is used to smooth out lumpy incomes. Thus, consumption is a close reflection of their average welfare or living standards. In higher-income countries, national aggregates are often based on household current income. As populations live farther from the concept of minimum needs, households are more likely to have stable formal incomes, and thresholds reflect what is regarded to be needed to maintain middle-class living standards. For Viet Nam, global poverty monitoring is based on welfare measured by household consumption (Ravallion, 2015; Deaton and Zaidi 2002; Mancini and Vecchi 2022). Welfare is measured by valuing a basket of commodities or goods and services that a household consumes based on expenditure data from household surveys. Households are assumed to gain value from consuming these goods and services, which increases their welfare or their utility function. The World Bank monitors global poverty using three poverty lines, for relevance across a range of countries. First, the World Bank’s International poverty line ($2.15/day 2017 PPP) is used for monitoring SDG 1.1. The Lower-Middle ($3.65/day 2017 PPP) and Upper-Middle ($6.85/day 2017 PPP) income country poverty lines were derived as the median of national poverty lines from countries in these income groups (Joliffe and Prydz 2016). Global poverty and national poverty measurements should be treated separately and are used for different purposes. The global poverty line is used primarily to measure progress on global goals and to make cross-country comparisons. Today, poverty in Viet Nam is low, and higher global poverty lines should also be used for monitoring. For the very poor, any small adverse shock can lead to long-term or permanent reversals in welfare, and the use of lower poverty lines (Lower-Middle income (LMIC) poverty line ($3.65/day 2017 PPP)) is useful to monitor those in extreme poverty. At the same time, Viet Nam is nearly an upper middle-income country, and the share of the population that is poor by World Bank’s LMIC poverty rate is now less than 5 percent, providing monitoring of a small percentage of the population. COVID-19 illustrated that even those that are not poor can be hit by economic shocks and fall into poverty. During the pandemic, the urban and non-poor populations also experienced income losses and economic shocks. For monitoring higher aspirations, as well as larger shares of the population, it is valuable to utilize a higher poverty line, namely the UMIC poverty line ($6.85/day 2017 PPP). It is informative to convert the value of global poverty lines to local currency to understand its relevance to the local context. For the 2022-25 Socio-Economic Development Plan, Viet Nam’s multi-dimensional poverty index includes a monthly income poverty line of VND1.5 million per capita for households in rural areas and VND2 million per capita for households in urban areas. While income and consumption are different concepts, this income line is closer to the global UMIC poverty line of $6.85/day 2017 PPP. 20 Viet Nam bi-annual poverty & equity update - June 2024 1. Poverty and equity developments  Box 1.A. World Bank monitoring indicators (contd) Table 1. Global poverty lines in PPP and VND 2017 PPP 2017 VND 2017 VND 2017 VND per day per day per month per year International Poverty Line (IPL) $2.15 16,786.37 510,585.31 6,127,023.75 Lower-Middle income (LMIC) poverty line $3.65 28,497.78 866,807.62 10,401,691.49 Upper-Middle income (UMIC) poverty line $6.85 53,482.14 1,626,748.55 19,520,982.65 Note: All values are in per capita terms. The 2017 PPP conversion factor for Viet Nam is VND7,807.61. This means that ~VND7,807 can purchase $1 of a representative basket taking into account purchasing power. Source: Construction of the Viet Nam 2022 Consumption Aggregate note. Prosperity gap The prosperity gap is a new indicator to monitor shared prosperity (Kraay et al., 2023). This new indicator was developed with the objective to measure shared prosperity, but to reward growth among the poorest. It measures the average shortfall in household income or consumption from a standard of prosperity set at $25 per person per day (2017 PPP). It is defined as the average factor by which income or consumption need to be multiplied to bring everyone to this prosperity standard. The prosperity gap is based on the ratio (z/y_i), where z= $25 per day is the prosperity standard, and y_i is the income of individual i. For a person whose income is $2.50 per day, this ratio is 10, meaning that their income must increase by a factor of 10 to reach the prosperity standard. Similarly, for a person with income equal to $5 per day, this ratio is five, meaning their income would need to increase by a factor of five. For a person with income equal to $25 per day, the ratio is equal to one, meaning no increase is needed to reach the prosperity standard. The prosperity gap is simply the average of these ratios across the entire population, and is the average factor by which incomes must be multiplied to attain the prosperity standard of $25 per day. This measure unambiguously gives greater weight to poorer people. The poorest person with daily income equal to $2.50 gets twice the weight of a person with double the income (at $5.00), and 10 times the weight of a person with 10-fold higher income (at $25). Summarized from: https://blogs.worldbank.org/en/developmenttalk/prosperity-gap-proposed-new-indicator-monitor-shared- prosperity Definition of economic classes To highlight the evolution of households along the entire welfare distribution, this paper uses the economic class definitions defined in the World Bank’s Riding the Wave report (World Bank, 2017). These economic class thresholds were developed for economies in the East Asia and Pacific region based on global poverty lines at the time. Originally thresholds were based on 2011 PPPs. Lower economic classes are bounded by the World Bank’s $3.10, and $5.50/day 2011 PPP poverty lines. The middle-class threshold of $15.00/day 2011 PPP is similar in value to other thresholds for large regional analyses (Ferreira et al., 2013; World Bank, 2018a). In high-income countries, middle-class lines can be even higher (e.g., $50.00/day), but upper ends of the welfare distribution do not translate well in household surveys, where the tail distributions are the most problematic. For this report, the economic class thresholds are updated to 2017 PPP using USA CPI. Towards more inclusive cities 21 PART 1. POVERTY & EQUITY UPDATE Box 1.A. World Bank monitoring indicators (contd) The World Bank’s regional and global poverty thresholds are intended to be used for cross-country comparisons, but these thresholds are also suitable for examining the welfare story of Viet Nam. While some of the lower-level World Bank poverty lines are comparable in monetary value to the MOLISA monetary dimension of poverty thresholds, these definitions do not reflect government definitions of economic classes. Rather, they are created to help illustrate the discussion on welfare dynamics across the entire distribution of households. Table 2. Summary of economic class definitions Definition Value in 2011 Inflated value Approximate Notes of economic PPP per day to 2017 PPP value VND [1] classes and per capita using CPI Poor (<$-3.1) $3.5 626.5-1,055 The upper threshold is the moderate poverty line traditionally used by the World Bank in analyzing trends in developing East Asia and Pacific. Economically ($3.1-$5.5) $3.50-$6.00 1,055-1,813 Between the World Bank’s LMIC and UMIC poverty lines. vulnerable Economically ($5.5-$15) $6.00-$16.50 1,813-4,946 Above the World Bank UMIC poverty line, but not yet secure middle-class (see below). Middle-class $15+ $16.50+ 4,946+ Those living on more than $15.00 a day. This threshold is broadly consistent with the values used by other studies (World Bank, 2018a). Notes: [1] Units: Jan 2020 VND in thousands, monthly, and per capita. Source: Summarized from World Bank (2017). Figure 17. Growth Figure 17 curves, total household 2. Sources and dynamics incidence income behind changes in poverty 2010-22 2016-18 2012-14 2018-20 2014-16 2020-22 Understanding changes to household income is a key 12 to understanding dynamics in poverty13. Similar to consumption, household income had its lowest growth 10 during 2020-22 compared to any other period over the last decade (Figure 17). For much of the last decade, household 8 incomes were growing at above 6 percent per annum Annualized growth (%) throughout multiple periods and for most households. 6 Shared growth helped reduce poverty dramatically and also in a progressive manner. Rising labor incomes (primarily wages) have been the primary sources of growth for 4 households (Figure 18, and see Figure 39 in Annex A for trends by region). From 2010 to 2022, annual household 2 net income increased by nearly VND50 million (2017 VND prices), driven by rising wages. 0 13 Note: Household income per capita in 2017 PPP terms. Dynamics are based on cross-sectional data, panel data is not available Source: WB staff calculations using VHLSS. after 2018. 22 Viet Nam bi-annual poverty & equity update - June 2024 2. Sources and dynamics behind changes in poverty  Other sources of non-labor income – including private Income decompositions show the role of different transfers, public transfers, and pensions – are either factors, relative to others, towards changes in poverty. much lower in value, or are received by a much smaller The primary drivers of poverty reduction are summarized in set of households. While pension incomes are higher among Figure 19. While the poverty-increasing effects from declining those who receive them, they are also the least common source employment and farming income were not necessarily of income. Private transfers (including remittances14) are the larger in 2020-22 than previous years, labor incomes had a most common source of income cited by households, since much smaller poverty reduction impact to compensate and they include gifts received throughout the year, including completely offset the decline in other sources of income. These annual Tet holidays. However, they have traditionally been factors are broadly consistent across all regions. One exception a small income source compared to labor incomes. Private is the Southeast region, where a decline in non-farm business transfers increased in 2022, likely because households received income was also poverty-increasing. This is consistent with more support from friends and family to cope during business closures from longer and stricter lockdowns in this COVID-19. While the share of households that received social region, home to the country’s commercial hub Ho Chi Minh assistance also increased considerably in 2022, the average City. Due to population aging and demographic changes, amounts did not – reflecting the low amounts of COVID-19 including out-migration, the share of employment is also a cash transfer support. factor in the rise in poverty in some regions. Figure 18. Trends in household income by component, 2010-22 Note: Annual net household incomes are averaged among households that report having this source of income, in other words, does not include zeros. Deciles based on consumption. A more detailed description of household income trends can be found in Box 1.B. “Other” includes wedding and funeral gifts, financial income, and rental income. Source: WB staff calculations using VHLSS. 14 It is not possible to distinguish the context of private transfers, or if amounts are remittances from household members working elsewhere or support from other family members. Towards more inclusive cities 23 PART 1. POVERTY & EQUITY UPDATE Figure 19. 2020-22 UMIC ($6.85/day 2017 PPP) poverty decomposition, by income sources Note: Shapley decomposition by components of welfare. Source: WB staff calculations using VHLSS. In 2022, labor incomes or extensive margins (Figure 20). In wage and non-farm experienced small declines in income categories, average incomes among recipients still increased (intensive margins), but there were also fewer either the intensive or extensive households reporting income from these sources (extensive margins margins). Since the start of the data series in 2010, 2022 was In 2022, three sources of labor income accounted for the first year where there was a small decline in the share of about 90 percent of household income across the entire households that reported receiving wage income. Regions in distribution of households (see Box 1.B). Total labor the south, the Mekong Delta and Southeast, saw the lowest income is the sum from three sources: 1) wage incomes15, growth in wages from 2020 to 2022 (see Figure 42 in Annex 2) net family farm production income, and 3) net non-farm A for regional wage decomposition). Incomes from non- family business incomes. Only 4.5 percent of households farm business continued to grow, but this income source is in Viet Nam do not have labor income from any of these more common among wealthier households and has a smaller categories. impact on poverty reduction. The share of households earning farm income fell across all regions, continuing a long trend of All three forms of labor incomes in 2022 experienced exits from agriculture. Net farm incomes among remaining some degree of negative impacts in either the intensive farming households also declined slightly. 15 Total household wage income includes earnings from first, second, and third jobs that all working household members were engaged in over the last 12 months. Due to a mix of factors – including a large agricultural workforce, informality, and classification of informal activities such as family businesses – earnings are not always reported as wage income. Lottery sellers are a particularly unique occupation, as workers in this sector often report they are employed, but not earning wages as they earn income from commissions on tickets sold and thus report business income. Even with these distinctions, there is large variation in the type and quality of work associated with wage employment. Wage income can be derived from any sector (agriculture, manufacturing, and services) and obtained through informal or skilled contract work. Reporting wages from the agriculture sector is uncommon, as the majority of households report it as part of self-employed family farming. 24 Viet Nam bi-annual poverty & equity update - June 2024 2. Sources and dynamics behind changes in poverty  Figure 20 20. Changes in household labor income, decomposed by intensive versus extensive margins Figure A. Wage B. Farm C. Non-Farm 16 16 16 Change in household income Change in household income Change in household income 11 11 11 (2017 VND, millions) (2017 VND, millions) (2017 VND, millions) 6 6 6 1 1 1 -4 -4 -4 -9 -9 -9 -14 -14 -14 2 6 8 0 4 2 2 6 8 0 4 2 2 6 8 0 4 2 -1 -1 -1 -2 -1 -2 -1 -1 -1 -2 -1 -2 -1 -1 -1 -2 -1 -2 10 14 16 18 12 20 10 14 16 18 12 20 10 14 16 18 12 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 Average earnings % HH earning income Average earnings % HH earning income Average earnings % HH earning income Note: Average annual household income, in 2017 VND, millions. Absolute changes in average household income can be decomposed into two factors, 1) intensive margins: changes in average earnings among households who earn it, and 2) extensive margins: changes in the share of households receiving that source of income. This paper uses the rate decomposition method by Gupta (1993). Source: WB staff calculations using VHLSS. Figure 21 Figure 21. Private and public transfers increased, but not enough to compensate for losses in labor incomes A. Public transfers B. Private transfers 16 16 Change in household income Change in household income (2017 VND, millions) 11 11 (2017 VND, millions) 6 6 1 1 -4 -4 -9 -9 -14 -14 2 6 8 0 2 2 6 8 0 4 4 2 -1 -1 -1 -2 -1 -2 -1 -1 -1 -2 -1 -2 10 14 16 18 20 10 14 16 18 12 12 20 20 20 20 20 20 20 20 20 20 20 20 20 Average earning % HH earning income Average earning % HH earning income Source: WB staff calculations using VHLSS. Source: WB staff calculations using VHLSS. Transfers offset some labor Regional poverty trends near the pandemic period show income losses, but these amounts that countries able to spend and protect were able to manage poverty, even amid economic contractions. The were small on average positive impacts and importance of social protection and Public and private transfers increased in 2022, but by other policy strategies implemented during the COVID-19 small amounts. More transfers were private rather than public crisis are now apparent. Growth elasticities showed that while (social assistance). The share of households receiving social some countries experienced a decline in economic growth, assistance income increased, but average amounts received by poverty rates still fell (Figure 22). In countries like Indonesia households were still small (Figure 21, panel A). With strong and Thailand, economic growth contracted, but at the same family networks, private transfers are the most common form time, the government spent heavily on cash transfers. In 2021, of household income, reported by more than 80 percent of Thailand spent about 18.4 percent of GDP on COVID-19 households. However, in absolute value it is also one of the related support compared to 5 percent in Viet Nam (World lowest. Private transfers rebounded after a period of decline, Bank, 2021a and 2023). In Viet Nam, the share of households and increases were seen in all regions (Figure 21, panel B). that received assistance in 2022 was much higher than in Towards more inclusive cities 25 PART 1. POVERTY & EQUITY UPDATE the past, indicating an expansion to new households that Transition out of agriculture is received pandemic-related support, but the average amounts linked with income and poverty of assistance received were small. A range of policies were put dynamics into place during COVID-19, but there was more spending on firms than for households in Viet Nam (World Bank, 2023). The movement from farm income to higher wage income is a key driver of Viet Nam’s successful poverty reduction Figure 22. Growth and poverty trends near the and progressive patterns of upwards economic mobility. pandemic period, selected EAP countries While fewer families were earning incomes from family farms, Figure 22 most regions experienced a comparable rise in the share of 4 VNM 2020-22 households that received wage income (mainly in non-farm Annualized change in poverty rate ($6.85/day 2017 PPP) 2 PHL 2018-21 sectors). The number of households earning a combination of THA 2019-20 labor incomes from wage and farm incomes grew until 2018 IDN 2020-21 0 (Figure 23). From 2020, the number of households earning CHN 2019-20 only wage incomes increased, further continuing the exit out IDN 2019-20 -2 of farming as a primary livelihood strategy. -4 The share of households no longer receiving labor income from farming is declining in all regions of Viet -6 Nam. Nationally, 66 percent of households reported farming VNM 2018-20 as an income source in 2010, declining to 46 percent by 2022. -8 The most dramatic decline was in the Red River Delta region; -10 from 68 percent of households engaged in farming in 2010, -8 -6 -4 -2 0 2 4 Annualized change in GDP in per capita (%) 6 down to 39 percent in 2022. Since 2018, there has been a faster rate of exits from farming activities, with the largest Note: Poverty rates based on the World Bank’s UMIC poverty line ($6.85/day declines in crop and livestock activities (Figure 25). The 2017 PPP). Source: World Bank staff calculations using PIP, WDI. government also aims for further structural transformation Figure 23. Over time, more households receive labor Figure 24. Distribution of households, by labor incomes only in the form of wage income income diversification and region, 2022 Figure Figure 2323 Figure Figure 2424 8 8 100100 Number of households (millions) Number of households (millions) 90 90 Share of households (%) Share of households (%) 7 7 80 80 70 70 6 6 60 60 5 5 50 50 40 40 4 4 30 30 20 20 3 3 10 10 2 2 0 0 on c e on rm e Fa m o y Fa on y Fa m o ly on anly d W n-fa nm d W e onm ag ly no ge a ly g nd W e ann-f nm d ag d arm W nd m dn ,f m , -ff , m -fa , rm To al l go arm on aarrm ta N or in om N -fa om r nl rm nl t rm n a n ar ag r ar r an age far 1 1 To no -fa ag o -efa e a fa -fa o W eo rm ab inc W n an d n en r Fa W ol r n an a N labo W 0 0 o 20102010 2012 2014 2016 2012 2014 2016 2018 2018 20202020 2022 2022 N labor No No income labor income Non-farm Non-farm only only Farm Farm only only Central Central Highlands Highlands Mekong Mekong Midlands Delta Midlands Delta Northern andand Northern MtnsMtns Farm andand Farm non-farm Wage non-farm only Wage only Wage andand Wage non-farm non-farm Northern andand Northern Central Central Coast CoastRedRed River Delta River Delta Southeast Southeast Wage andand Wage farmfarm Wage, farm, Wage, andand farm, non-farm non-farm Note: Households receiving different combinations of labor income may also Source: WB staff calculations using VHLSS. receive non-labor income. Source: World Bank calculations using VHLSS. 26 Viet Nam bi-annual poverty & equity update - June 2024 2. Sources and dynamics behind changes in poverty  into modern sectors, with a target for 25 percent of the labor from only family farming activities. These households tend force to be engaged in the primary agricultural sector by 2025 to be older and have lower levels of education. Given the within reach (27 percent in 2023). Conversely, households in profile of these aging farming households, social protection the poorer regions are more likely to have labor income only policies may be more appropriate than promotion policies from farming. to support this group. Households reliant on farming for labor income have The two regions in Viet Nam with the highest rates of the highest poverty rates. Households with labor incomes poverty and labor shares in agriculture are the Central only from family farming or a combination of wage and Highlands and the Midlands and Northern Mountains farming, have the highest poverty rates (Figure 26). In 2022, regions. Comparisons of these regions are insightful, with about 10.8 percent of households received labor income different paces of poverty reduction and dependence on agriculture. In 2010, the Midlands and Northern Mountains stood out with poverty rates that surpassed any other region, Figure 25. Share of households participating in but now has a poverty rate more similar to the Central agriculture is declining Figure 25 Highlands region (Figure 10). Much of the success can be related to stronger diversification out of agriculture compared 45 Rice to the latter, with the former seeing the largest increase in the 40 Livestock share of households receiving income from wages, which are Staples By- products generally higher than farm income. In 2010, 56 percent of 35 Share of households (%) households in the Midlands and Northern Mountains region 30 Fruit had income from wages, increasing to 67 percent in 2022. 25 On the other hand, compared to other regions, the Central Industrial crops Highlands had the lowest total household wage income, and 20 Forestry the smallest share of households with at least one person with 15 Aquaculture a formal contract. While farm incomes in Central Highlands 10 are higher than in Midlands and Northern Mountains, 5 Agricultural land there is a higher dependence on agriculture in the former. In Agricultural services addition, one-in-three workers is economically active in the 0 2010 2012 2014 2016 2018 2020 2022 coffee sector alone. Figure 26 Source: WB staff calculations using VHLSS. Figure 26. LMIC and UMIC poverty rates, by household income sources, 2010-22 Higher poverty Average poverty Lower poverty 70 70 70 60 60 60 50 50 50 Poverty rate (%) Poverty rate (%) Poverty rate (%) 40 40 40 30 30 30 20 20 20 10 10 10 0 0 0 2010 2012 2014 2016 2018 2020 2022 2010 2012 2014 2016 2018 2020 2022 2010 2012 2014 2016 2018 2020 2022 Wage and farm Farm only Farm and non-farm Total No labor income Non-farm only Wage, farm, and non-farm Wage and non-farm Wage only Note: Categorization based only on receipt of various types of labor income. Households can receive other non-labor sources of income. Source: World Bank calculations using VHLSS. Towards more inclusive cities 27 PART 1. POVERTY & EQUITY UPDATE Agriculture is also an important part of the economy Short- and long-term trends will in the Mekong Delta. The share of the labor force in impact employment agriculture is a bit higher than one-third, the third highest among all regions. One aspect that stands out is that due Among wage-earning households, declines in to the higher concentration of large agri-businesses, labor employment in services were most common in 2022 incomes in the delta from farming are more likely to be in the (Figure 28). From 2020 to 2022, the share of households form of wages, rather than family farm income (Figure 27). with wage incomes declined in most regions except for the Based on the 2023 Enterprise Census, 35 percent of all large Northern and Central Coast and the Midlands Northern enterprises specialized in food products are located in the Mountains. The largest declines in wage employment were delta. Relatedly, labor and wage incomes in the agricultural unsurprisingly in regions with the largest cities, the Red sector in the delta are higher than in most other regions. River Delta (Hanoi) and Southeast (HCMC). While drops However, average wages in the manufacturing and services in employment rates were small on average, there were sectors are the lowest in the delta compared to other regions additional impacts from occupational downgrading during in Viet Nam. The Mekong Delta is also experiencing adverse the pandemic. The underemployment rate was the highest in environmental impacts from climate change. Declining farm Q3-2021 at 4.5 percent, but dropped to 2 percent nationally incomes coupled with lower wages in non-farm sectors are by Q2-2023 (GSO, 2023). Across the six regions, the push factors for migration out of the region. The declining highest underemployment rate in 2023 was in the Mekong share of households that report receiving private transfers Delta region. could also be a signal that out-migration is becoming more permanent or that entire households are leaving the Mekong (see Figure 39 in Annex A). Figure 27. Average incomes, by primary sector and region 2016-22 Note: Labor income includes wage and non-wage incomes. Source: WB staff calculations using LFS. 28 Viet Nam bi-annual poverty & equity update - June 2024 2. Sources and dynamics behind changes in poverty  Figure 28. Employment trends in wage-earning Figure 29. Employment share, by cohort and year Figure Figure 2828households, by occupation Figure Figure 2929 18 18 100% 100% 17 17 Elementary Elementary Occupations Occupations 16 16 90% 90% 15 15 80% 80% 14 14 Share of cohort that is employed (%) Share of cohort that is employed (%) 13 13 70% 70% 12 12 Employed (millions) Employed (millions) 11 11 60% 60% 10 10 9 9 50% 50% 8 8 7 7 Craft Craft & Related & Related Trades Trades Workers Workers 40% 40% 6 6 5 5 Plant Plant & Machine & Machine Operators, Operators, 30% 30% Service Service & Sales & Sales Workers Workers & Assemblers & Assemblers 4 4 Professionals Professionals 20% 20% 3 3 Skilled Skilled Agricultural, Forestry Forestry Agricultural, and Fishery and Fishery Workers Workers & & Technicians Technicians 2 2 10% 10% Associate Associate Professionals Professionals 1 1 Clerical Clerical Support Support Workers Workers Managers Managers 0 0 Armed Armed Forces Forces 0%0% 2010 2012 2010 2012 2014 2014 2016 2016 2018 2018 2020 2020 2022 2022 2010 2012 2010 2014 2016 2012 2014 2016 2018 2018 2020 2020 2022 2022 1950-59 1950-59 1960-69 1960-69 1970-79 1970-79 1980-89 1980-89 1990-99 1990-99 2000-09 2000-09 Note: Main job. Change in definition excludes subsistence farmers. Sector is not collected in the 2022 VHLSS. Source: World Bank calculations using VHLSS. Source: World Bank calculations using VHLSS. Labor trends during and post-pandemic are also and the need for continuous income. For example, among the affected by the demographic transition. Viet Nam’s younger cohort born between 2000-09, employment shares in labor force is getting smaller both from aging, but also 2022 vary from almost 40 percent in the poorer Midlands and from a younger population entering the labor force later Northern Mountains region, to only 17.4 percent in the Red due to higher education completion (Figure 29). Education River Delta where Hanoi is located. Among the older cohort is improving, but there are regional differences. In poorer born between 1950-59, 71 percent was still working in the regions, individuals enter the labor force sooner due to lower Midlands and Northern Mountains, compared to 30 percent education completion, but exit later due to lack of pensions in the Southeast, home to HCMC. Box 1.B. Household income dynamics The dependence on different income sources evolves across the distribution of households in expected ways (Figure 30 and Figure 31). Income from family businesses constitute larger shares of total income among richer households, whereas poor households have larger shares of family farm income. Household non-farm business income has been growing at similar levels compared to wage income since 2010. However, much fewer households are engaged in non-farm family businesses, and those that do are more likely to be at the upper ends of the distribution. These businesses are primarily in the retail trade and food and restaurant. Among the richest decile, average family business income is larger than total income of households at the bottom decile. Towards more inclusive cities 29 PART 1. POVERTY & EQUITY UPDATE Box 1.B. Household income dynamics (contd) Figure Figure 30Share of household income by source 3030. Figure Figure Figure 31 Figure 31 31. Annual household income by source and decile, 2022 and decile, 2022 100%100% 350 350 Annual household income (millions, VND) Annual household income (millions, VND) 90% 90% Share of average household income Share of average household income 300 300 80% 80% 70% 70% 250 250 60% 60% 200 200 50% 50% 150 150 40% 40% 30% 30% 100 100 20% 20% 50 50 10% 10% 0% 0% 0 0 1 12 23 34 45 56 67 78 89 9 10 10 1 12 23 34 45 56 67 78 89 9 10 10 Family Family FarmFarm Family Family Business Business WageWage Public transfers Public transfers Family FarmFarm Family Family Business Family BusinessWageWage Public transfers Public transfers Private transfers Private Pension transfers Pension Other Other Private Private transfers transfers Pension Pension Other Other Note: Deciles based on welfare = household consumption per capita. Decile 1 is the poorest, and decile 10 is the richest. Source: World Bank calculations using VHLSS. In 2022, average annual net household income was VND185 million, but ranged regionally from a low of VND132 million in the Central Highlands to VND230 million in the Southeast (Figure 32). Over the last 12 years, the change in real household income is lowest in the southern regions (Figure 33). Household income increased by VND24 million from 2010 to 2022 in the Mekong Delta, and by VND78.8 million in the Red River Delta. Some trends in household income categories are broadly consistent across regions, while others are region specific (see Figure 39 in Annex A). Nationally, across all regions, households are exiting farming as an income source. The share of households that report any income from farming is declining rapidly, with the largest decline seen in the Red River Delta region. In other income categories, there are smaller shifts across regions. The six regions of Viet Nam experienced varied degrees of impacts from COVID-19. The Southeast region, where HCMC is located, and the Mekong Delta show the largest dampening of household wage income. The Mekong Delta is the only region where household incomes have declined over a four-year monitoring period (2018-22). The share of households with non-diversified labor incomes is increasing. In 2010, 39.2 percent of households had non-diversified labor incomes, meaning they received labor income from wages, farm or non-farm sources. In 2022, this share rose to 47.4 percent of households (Figure 35). Households with the lowest total incomes are those who either do not have any labor income sources, or only labor income from family farming (Figure 34). The share of households receiving incomes in only farming is declining (16.2 percent in 2010 to 10.6 percent in 2022), but due to an aging population the share of households without any labor income sources is also increasing (2.8 percent in 2010 to 6.1 percent in 2022). 30 Viet Nam bi-annual poverty & equity update - June 2024 2. Sources and dynamics behind changes in poverty  Box 1.B. Household income dynamics (contd) Figure 32 32 Figure Figure 32. Trends in net annual household Figure Figure Figure 3333. 33Increase in household income from income, by region 2010 to 2022, by region 250 250 90 90 Annual household income (millions VND) Annual household income (millions VND) Change in annual household income Change in annual household income 80 80 200 200 70 70 60 60 (2017 VND) (2017 VND) 150 150 50 50 100 100 40 40 30 30 50 50 20 20 10 10 0 0 2010 2010 2012 2012 2016 2016 2014 2014 2018 2018 2022 2022 2020 2020 0 0 Central Central Highlands Highlands Mekong Mekong Delta Delta Central Central Highlands Highlands Mekong Mekong Delta Delta Midlands Midlands and Northern and Northern Mtns Mtns Northern Northern and Coastal Central and Coastal Central Midlands Midlands and Northern and Northern Northern Mtns Mtns Northern and Coastal and Coastal Central Central Red River Red River Delta Delta Southeast Southeast Red River Red River Delta Delta Southeast Southeast National National National National Note: Average annual household total net income in VND and millions. Using CPI adjusted to 2017 prices. Source: WB staff calculations using VHLSS. Figure 34. Annual household income, by labor Figure 35. Distribution of households, by labor Figure Figure 34 34 Figure Figure 35 35 income diversification income diversification 300 300 100% 100% Annual net household income (millions VND) Annual net household income (millions VND) 90% 90% 250 250 80% 80% Share of households (%) Share of households (%) 70% 70% 200 200 60% 60% 150 150 50% 50% 40% 40% 100 100 30% 30% 20% 20% 50 50 10% 10% 00 0% 0% 2010 2012 2010 2014 2016 2012 2014 2016 2018 2018 2020 2020 20222022 2010 2010 2012 2012 2014 2016 2018 2014 2016 2018 20202020 20222022 No No labor labor income income Non-farm Non-farm only only Farm Farm only only No No labor labor income income Non-farm Non-farm only only Farm Farm only only Farm Farm and and non-farm non-farm Wage Wage only only Wage Wage and and non-farm non-farm Farm Farm and and non-farm non-farm Wage Wage only only Wage Wage and and non-farm non-farm Wage Wage and and farm farm Wage, Wage, farm, farm, and and non-farm non-farm Wage Wage and and farm farm Wage, Wage, farm, farm, and and non-farm non-farm Note: Median annual household total net income in VND and millions. Source: WB staff calculations using VHLSS. Using CPI adjusted to 2017 prices. Households receiving different combinations of labor income may also receive non-labor income. Source: World Bank calculations using VHLSS. Towards more inclusive cities 31 PART 1. POVERTY & EQUITY UPDATE 3. Looking forward (Figure 36). After a short period of decline in 2020 and 2021, labor incomes are once again growing, but are at lower Examination of household conditions during the levels than if pre-COVID income growth trajectories were COVID-19 pandemic and immediate recovery period sustained. In 2021, workers in the services and manufacturing in Viet Nam have revealed dampened household sectors were earning more than VND1 million less per month incomes, which negatively affected consumption, and than if labor income growth had followed a pre-COVID stalled progress in poverty reduction. As growth returns, trajectory. In 2023, this gap had narrowed to VND580,000 poverty reduction should as well. Yet, experiences from the less per month. Labor incomes in the agricultural sector are COVID-19 period are still important to reflect on the timing, higher than projected pre-COVID, but are still much lower utilization, and mix of promotion and protection policies to than those in the manufacturing or services sectors. support households both in good times and in bad. There is potential for more shocks in the future that require strong A smaller post-COVID labor force is due to both social protection and investments. A changing profile in measurement, demographic, and economic factors. First, terms of an older population and more households in urban a survey definition change in 2021 no longer categorizes areas also requires updating strategies to tackle vulnerability. subsistence farmers as being in the labor force (Figure 37). Between 2020 and 2021, the agricultural sector shrank by With the return of economic 1.8 million workers, with the largest declines in the rural growth, income and poverty rates Midlands and Northern Mountains region, which is primarily rural and a poorer region in Viet Nam. Second, the labor are improving force in the manufacturing and services sector is once again Labor income indicators in 2023 are recovering to pre- growing, but at a much slower pace compared to pre-COVID. COVID levels, but are still lower than expected incomes During the 2015-19 pre-COVID period, the labor force in the without COVID. During the pre-COVID 2015-19 period, manufacturing sector expanded by about one million workers labor incomes were growing by nearly 10 percent per year per year annually, but contracted during 2019-21. Figure 36. Recovering labor incomes, but still lag Figure 37. National trends in labor force size behind pre-COVID trends Figure 36 Figure 37 10,000 60,000 9,000 50,000 8,000 Monthly VND (thousands) Labor force (thousands) 7,000 40,000 6,000 30,000 5,000 4,000 20,000 3,000 2,000 10,000 1,000 0 0 2015 2016 2017 2018 2019 2020 2021 2022 2023 2015 2016 2017 2018 2019 2020 2021 2022 2023 Services Manufacturing Agriculture Total Agriculture Manufacturing Services Note: Nominal labor income includes wage and non-wage incomes. Linear trends Source: LFS. based on data from 2015-19. Actual in solid lines, and trends in dashed lines. Source: WB staff calculations using LFS. 32 Viet Nam bi-annual poverty & equity update - June 2024 3. Looking forward  The labor force in the services sector experienced a Revised poverty projections up to 2026 show renewed decline later in 2021, coinciding with severe lockdowns progress. However, results should be interpreted with caution after the arrival of the more contagious COVID-19 since projections may not be as accurate in unprecedented Delta variant. Lockdowns more adversely affected sectors contexts without previous data trends for calibration. During where face-to-face client interactions are necessary including this unprecedented period, while GDP growth was still restaurants, retail, and travel. While manufacturing jobs in positive in Viet Nam, labor incomes declined in 2021, which industrial parks were somewhat more insulated and some negatively affected household welfare. Growth has resumed, remained open throughout the pandemic, lockdowns affected but there are variations in how it is transmitted to households, the services sector more acutely (World Bank, 2023). requiring more distributional analysis of growth. Post-COVID poverty projections show progress in the Despite a return to growth, policy makers should recovery period, but rates remain elevated compared continue monitoring household developments in the to past outlooks. Poverty projections made at the outset of medium- and long-terms. Global and regional economic COVID-19 predicted a stagnation in poverty reduction in outlooks for 2024 are also subdued (World Bank, 2024a). 2021, followed by declines in 2022 and onwards (World Bank, In the EAP region, there are added risks from weaker-than- 2023). Figure 38 illustrates actual poverty rates and poverty expected growth in China. As households rebound, it is projections based on pre- and post-COVID estimates. Actual important to monitor which groups have slower or faster poverty rates in 2022 were higher than predicted using recoveries. To foster a longer-term sustainable recovery, policy earlier growth forecasts. Earlier forecasts did not consider the solutions must address potential long-term economic scarring potentially long-lasting and residual effects from the economic faced by vulnerable households that cannot quickly bounce slowdown onto households. back following a prolonged crisis. Households that are already Figure 38. Poverty trends and projections Note: Actual poverty rates are bi-annual, and poverty projections are annual. Projections made using March and October 2021 growth forecasts start in 2021. Poverty projections made using February 2024 growth forecasts start in 2023. Poverty projections based on methods described in Lakner, Mahler, Negre, and Prydz (2020). Source: World Bank staff calculations using VHLSS. Towards more inclusive cities 33 PART 1. POVERTY & EQUITY UPDATE poor during a crisis face a longer economic recovery than protection policies. The East Asia region and Viet Nam better-off ones. One thing to watch out for is if urban areas stand out for achieving an extremely successful growth- rebound faster or slower than rural ones. While incomes are driven poverty reduction process through promotion, recovering, it is yet to be seen if all households will recover, where high levels of economic growth provided decades of or if impacts will be permanent. In particular, there are three prosperity through job creation and rising wages, thus lifting groups and conditions to watch closely: 1) the already poor households out of poverty. This growth was broad-based and if they become worse off, 2) the economically insecure and benefitted most of the population. In such periods of and if they fall back into poverty, and 3) the economically high growth, protection policies were not as critical since secure and if they become insecure. promotion and growth were the main drivers of poverty reduction, with few being left out of the process. However, Policies for the Next Mile COVID-19 stalled growth throughout the world, including in Viet Nam. Strong protection policies were used by some Like other successful middle-income countries, there countries during this period to offset shocks, but less so in is now less concern regarding Viet Nam’s Last Mile Viet Nam. Income trends showed that as labor incomes challenges. These include development challenges such as declined, private transfers increased. More households also malnutrition, stunting, or low access to basic utilities such as received social assistance, but the amounts received were improved water or electricity. Primary and lower secondary small in comparison to private transfers. education completion is near universal, literacy rates among the younger generations are high, most of whom are also fluent As Viet Nam continues eliminating structural factors in using digital devices. The minimum levels of subsistence related to extreme poverty, future prosperity will be have been raised for most, but pockets of deprivation still more interlinked with economic developments and the exist among certain chronically disadvantaged or vulnerable creation of good jobs. The structural transformation of groups, such as ethnic minorities. Policy makers are now the labor force out of agriculture and into manufacturing straddling the dual objectives of tackling pockets of remaining and services sectors has been the main channel of upwards Last Mile challenges, while having their sights firmly set on economic mobility for households (World Bank, 2022b). Next Mile aspirations. However, the transformation is not complete, with the vulnerability of certain jobs highlighted during COVID-19, The Next Mile is the journey to upper and high- especially in urban areas. In cases where formal jobs were lost, income country living standards. For society, this means unemployment insurance measures and other worker support creating more economic opportunities to build a strong programs were slow to take effect. In some cases, workers in middle-class, while at the same time expanding support to informal jobs could not obtain evidence they lost their jobs, low-income and economically vulnerable households. The and were initially left out of some COVID-19 support. millions who have escaped poverty over the last decade Migrant workers also faced challenges, since it was difficult to now need to continue climbing up the economic ladder, obtain support if they worked in provinces other than where be provided safety nets to prevent them from falling back registered. Social assistance is primarily distributed by local into poverty, and be equipped with the human capital and levels of government for residents. Some also faced difficulty skills to engage in more productive and sophisticated work. returning home during lockdowns. The emergence of COVID-19 added to challenges regarding skills, productivity, climate change, and an aging society. Policies needed to tackle risk-induced insecurity are Whether these constraints to Last Mile poverty reduction not the same as policies currently in place that mostly and Next Mile high-income development challenges end up address chronic insecurity. The chronically insecure usually being short-term growing pains or long-term bottlenecks to live below the economic security line, even in good years, as the welfare trajectory of Viet Nam’s households will depend they lack the human and physical assets to earn a sufficient in part on government action and prioritization. livelihood even in good conditions. These households will most benefit from better livelihoods, cash transfers and Lessons from the exceptional period from 2020 to 2022 delivery of basic services to facilitate investments in physical call for the need to strengthen and time promotion and and human capital. Viet Nam has done well targeting this 34 Viet Nam bi-annual poverty & equity update - June 2024 3. Looking forward  group with investments in rural areas, National Targeted of few reported COVID-19 cases in Viet Nam, thus the Programs, subsidies, and other social assistance support negative impacts were more a result of self-imposed mobility programs. The pandemic also spotlighted those who face restrictions. Around two years later in 2022, when infections risk-induced insecurity. Individuals in this group may live spiked to some of the highest in the ASEAN region, about 40 above the poverty line, but their consumption can be highly percent of households still reported experiencing a negative variable because of idiosyncratic or covariate shocks, meaning shock. The crisis affected households across the entire sometimes they fall below the line. While these households welfare distribution through a variety of channels. Wealthier do not necessarily need the same investments in assets, they households more likely experienced income losses from family do need protection from shocks and may require insurance businesses, while poorer households were hit by losses from programs to increase resilience. As the COVID-19 pandemic farming activities. Adverse labor impacts occurred broadly showed, there are less policies in place for those who may face across different socioeconomic groups. risk-induced security. Urban households were more likely than rural ones to Identify and address shocks face vulnerabilities or risk-induced vulnerabilities, since most reported impacts were related to reduced economic Shocks or crisis can be felt at an individual or community activity. Negative events such as job losses, wage and business level, motivating different policy actions. In 2022, these disruptions were more frequently reported in urban areas included both structural changes in farming, and economic than rural ones. For example, results from the World Bank shocks to non-farming activities due to COVID-19. Examples COVID-19 survey (round 6) showed that by the end of 2021, of two different types of shocks are: (i) idiosyncratic shocks, or 36 percent of urban respondents knew of someone looking those which only affect specific individuals or households and for a job, compared to 31 percent of rural respondents. (ii) covariate shocks, or those which affect entire communities, Disruption from economic activity was the main impact cited regions, or countries. Idiosyncratic shocks include illness by households rather than from health shocks. or accidents, which usually affect a single individual or household. Covariate shocks, such as a natural disaster or an Declining labor incomes resulted in some short- economic shock, affect many more people. The distinction can term coping mechanisms with negative long-term be important as a household’s risk profile varies by the type of consequences to households. According to the Ministry of shock: susceptibility to illness may depend upon underlying Labour, Invalids and Social Affairs (MOLISA), during 2016- health, access to preventative care, living and working 21, about the same number of people joined (4.25 million) conditions. Susceptibility to natural disasters or lockdowns as those who prematurely withdrew (4.06 million) from the can depend on a household’s location, while vulnerability to Viet Nam Social Security (VSS) fund. In 2020, the number of economic shocks may be influenced by the sectors of workers. people joining VSS declined from the previous year. Increasing Idiosyncratic shocks are generally best dealt with by effective cases of social insurance withdrawals among the poorest social insurance coverage, such as unemployment insurance, suggest a broader presence of rising income insecurity during health insurance and old age pensions. Covariate shocks often the COVID-19 crisis. However, the number of workers’ require a more coordinated government response (preemptive withdrawals even before Viet Nam was hit by COVID-19 disaster risk management and ex-post disaster responses) or suggests a degree of economic insecurity had always existed large social programs (wage subsidies or direct income support and was magnified by the crisis. Meeting families’ cash needs as many countries deployed during COVID-19). was the primary reason cited by workers for the withdrawals. The COVID-19 pandemic was the most significant shock Social protection needs to be to hit Viet Nam in the last half century, with widespread modernized to guard against shocks impacts affecting most households. At the outset, almost 70 percent of households reported experiencing a negative COVID-19 highlighted the need to make more progress shock between February and June 2020, which covers the in addressing the coverage gap. The economic impacts of duration of the first nationwide lockdowns in April 2020 the pandemic were disproportionately felt by the non-poor (World Bank, 2021a). However, this was during a period informal sector in many countries, including Viet Nam. Towards more inclusive cities 35 PART 1. POVERTY & EQUITY UPDATE This led to ad hoc attempts to expand social assistance and monitoring surveys. Nationally, 2.3 times as many households cash transfer programs to this population. Some countries applied for cash support in 2021 than 2020, with the increase quickly scaled up by leveraging a variety of administrative much higher in the Southeast and Mekong Delta regions16. databases, but most faced serious challenges in their horizontal According to a MOLISA report, by June 30, 2022, more than expansions and were unable to reach many affected households 24 million workers and 500,000 business households were (Johnson and Palacios, forthcoming). This highlighted a supported nationwide. major gap in many social protection systems that focus on a static pool of social assistance beneficiaries on one hand and Some programs, besides cash transfers, faced hurdles or a relatively small formal sector covered by social insurance, on took longer to implement. Support of pandemic-impacted the other. employees and employers through the Unemployment Insurance Fund17 took longer than expected to disburse, In the first two years of the pandemic, Viet Nam spent although the processes of registration, verification and the least on support to households compared to other enrollment were based on existing databases. More than 99 economies in EAP (World Bank, 2023). In Mongolia, percent of employees received the support via personal accounts. spending on pandemic-related support cost nearly 25 percent Implementation of rent support for workers in industrial parks, of GDP, with direct support to households exceeding 12 export processing zones and key economic zones18 as part of percent of GDP. Among six large East Asian economies in part of the recovery package19 was limited, with disbursement a regional study, Viet Nam ranked lowest in both overall only utilizing 58.06 percent of the estimated budget. This pandemic relief, as well as in direct support to households, initiative aimed to retain and encourage workers to return to measured as a share of GDP. However, Viet Nam ranked industrial parks. The average level of support for an employee highest in terms of support offered to firms. The contrast in returning to the labor market was VND1 million per month, support delivered to households and firms partly reflects the and VND500,000 per month for a maximum of three months government’s focus on economic growth, but also the lack of for an employee working in an enterprise. a delivery chain in place to support households and workers outside the government’s system and registry. Another Other forms of relief disbursement and social assistance possible reason for the low level of official support is the were available to vulnerable groups, and sometimes filled perceived lack of need to support households. In 2020, Viet gaps when formal delivery chains failed. Based on the Nam and China were the only economies in the EAP region World Bank’s COVID-19 monitoring surveys, in the first two to boast positive GDP growth. years, households were more likely to receive assistance from donations and NGOs, than formally from the government. During the second half of the pandemic when In December 2021, migrants were more likely than non- COVID-19 cases climbed and lockdowns were longer, migrants to receive cash support from the government and more households began to apply for support. The social assistance from NGOs and charities. Urban households coverage radius of the second package was much broader than facing livelihood disruptions and in need of cash support also the first and focused on impacted sectors. While the initial applied for COVID-19 cash relief and other aid from the package focused on existing beneficiaries of government government, NGOs and charities. Recent migrants were also social assistance programs, the second package introduced more likely than non-migrants and non-recent migrants to the following year in 2021 targeted workers with certain be assisted by NGOs and charities, ranging from health kits difficulties and the most impacted households in hard-hit to food. Against these comparison groups, recent migrants sectors, such as tourism. More households reported applying were also more likely to apply for government cash relief, for pandemic-related cash relief in 2021 than 2020, rising despite having limited awareness of available assistance due to from 13.3 to 30.8 percent based on World Bank COVID-19 information gaps. 16 The World Bank Monitoring COVID-19 impacts on households in Viet Nam (round 6 data snapshot). 17 Resolution No.03/2021/UBTVQH15 and Resolution No.24/2022/UBTVQH15 of the National Assembly Standing Committee, Resolution No.116/NQ-CP (September 24, 2021) of the government. 18 Decision No.08/2022/QD-TTg (March 28, 2022). 19 Resolution No.43/2022/QH15 (January 11, 2022) of the National Assembly. 36 Viet Nam bi-annual poverty & equity update - June 2024 3. Looking forward  Countercyclical fiscal policies costs, and a structural transition of the labor force based on low-skill occupations and exiting agriculture. Viet Nam could also use fiscal policy more proactively to help households manage risk, by financing a modern Creating higher-quality jobs is more important than social protection system and cushioning shocks in a ever for Viet Nam, not only to accelerate post-pandemic countercyclical fashion. Households face risks at both macro growth but also to realize 2045 high-income status and micro levels. A critical policy to help households manage aspirations. Export-oriented and labor-intensive sectors these risks is a modern social protection system, including provided millions of jobs to a young population, but youth targeted social assistance to lift the remaining poor out of are now better educated and demand higher-quality jobs. As poverty, social assistance and social insurance to protect gains the population continues to age and the structural transition already achieved, and an adaptive social protection system to out of agricultural to non-farming is slowing, a new shift to scale-up support to existing and new beneficiaries when larger higher skilled jobs is necessary. shocks occur. Viet Nam needs to bring social protection spending in line with international norms to develop such a A Viet Nam Future of Jobs study (Cunningham and system. However, fiscal policy can also manage household risk Pimhidzai, 2018) offered three broad reform areas at a macro level when used in a countercyclical fashion. to support the creation and transition to better jobs. First, the creation of more good jobs in the modern sector is Fiscal policy can play a critical role in both driving Viet essential for a young population with higher education and Nam toward high-income status and doing so in an seeking higher-skilled employment. The best jobs – defined inclusive manner to assist the movement of people into by higher labor productivity, wages and social benefits – are a prosperous middle-class. It can achieve this in two ways. largely in the modern sector. Policy changes include fostering First, it can help finance the required investments needed for the creation and growth of enterprises more conducive to the the country and its workers to become more productive and creation of high-value jobs and to position Viet Nam higher higher earners, such as those discussed in this chapter so far: on the value chain. While Viet Nam ranks with the highest modernization of agriculture, better skills and higher quality Human Capital Index (HCI) among lower middle-income education, a more robust digital backbone, and accompanying countries, labor demand tied to exports has been dominated services. Second, it can also finance policies which can address by low value-added production activities. Skills building is key Last Mile and Next Mile constraints today, such as a modern to climbing the value chain. Second, the quality of existing social protection system and the strengthening of National jobs in traditional sectors should be further improved. Low- Targeted Programs. Finally, the revenues required to finance quality work is still prevalent and more needs to be done to these investments can themselves be more or less progressive. shift agricultural workers into higher value-added crops, and to facilitate business links between household enterprises and Better and more secure jobs SMEs. Third, support for training more qualified workers and connecting them with the right kinds of jobs is necessary. Viet Nam needs a productivity-led growth path to Continued upgrading and reforms to the education and reach higher levels of prosperity. The growth strategies training system are needed to develop more skills for future that underpinned Viet Nam’s poverty reduction success in jobs, while information systems are needed to match these the previous decade are no longer ones that can sustain the higher skilled workers with appropriate roles. Significant country’s growth path to higher income levels and the build- investments are required to develop cutting-edge research up of a large and secure middle-class. Even before the onset and training facilities, including laboratories, materials, and of COVID-19, the efficacy of the developing EAP region’s technologies, at Vietnamese universities and colleges. Strategic growth model based on low-skilled outward-oriented growth location of universities near companies, including industrial was expected to diminish as the world rapidly changes (Mason parks, is also essential. and Shetty, 2018). Given these emerging challenges, Viet Nam may not be as well positioned as it was a decade ago when it reaped advantages from a demographic dividend, low labor Towards more inclusive cities 37 PART 1. POVERTY & EQUITY UPDATE Annex A. Figures – Part 1 Figure 39. Household income trends, by region Source: WB staff calculations using VHLSS. 38 Viet Nam bi-annual poverty & equity update - June 2024 Annex A. Figures – Part 1  Figure 40 Figure 40. Regional annual household income in 2010 and 2022 Central Highlands Mekong Delta 250 250 200 200 6 150 150 2 3 2 4 15 4 1 1 1 7 10 1 1 41 2 30 23 100 2 100 17 29 37 67 50 55 50 52 66 27 31 0 0 2010 2022 2010 2022 Midlands and Northern Mountains Northern and Coastal Central Region 250 250 200 200 4 150 6 150 3 1 14 1 8 10 2 44 100 1 31 100 4 1 2 7 5 5 19 15 11 23 50 50 38 48 75 65 25 31 0 0 2010 2022 2010 2022 Red River Delta Southeast 250 250 5 7 2 5 12 13 11 200 1 200 14 0 3 14 54 55 150 150 48 8 4 1 9 10 9 100 29 100 41 37 144 123 50 50 85 53 0 0 2010 2022 2010 2022 Wage Farm Non-farm Private transfers Public Transfers Pension Other Note: Annual total household income in millions VND. Adjusted using national CPI to 2017 prices. Source: World Bank staff calculations using VHLSS. Towards more inclusive cities 39 PART 1. POVERTY & EQUITY UPDATE Figure 41 Figure 41. Decomposing monthly household income from farm production Central Highlands Mekong Delta Change in household income (2017 VND, millions) Change in household income (2017 VND, millions) 25 25 20 20 15 15 10 10 5 5 0 0 -5 -5 -10 -10 -15 -15 -20 -20 -25 -25 2 4 6 8 0 2 2 4 6 8 0 2 -1 -1 -1 -1 -2 -2 -1 -1 -1 -1 -2 -2 10 12 14 16 18 20 10 12 14 16 18 20 20 20 20 20 20 20 20 20 20 20 20 20 Average earnings % HH earning income Average earnings % HH earning income Midlands and Northern Mountains Northern and Coastal Central Region Change in household income (2017 VND, millions) Change in household income (2017 VND, millions) 25 25 20 20 15 15 10 10 5 5 0 0 -5 -5 -10 -10 -15 -15 -20 -20 -25 -25 2 4 6 8 0 2 2 4 6 8 0 2 -1 -1 -1 -1 -2 -2 -1 -1 -1 -1 -2 -2 10 12 14 16 18 20 10 12 14 16 18 20 20 20 20 20 20 20 20 20 20 20 20 20 Average earnings % HH earning income Average earnings % HH earning income Red River Delta Southeast Change in household income (2017 VND, millions) Change in household income (2017 VND, millions) 25 25 20 20 15 15 10 10 5 5 0 0 -5 -5 -10 -10 -15 -15 -20 -20 -25 -25 2 4 6 8 0 2 2 4 6 8 0 2 -1 -1 -1 -1 -2 -2 -1 -1 -1 -1 -2 -2 10 12 14 16 18 20 10 12 14 16 18 20 20 20 20 20 20 20 20 20 20 20 20 20 Average earnings % HH earning income Average earnings % HH earning income Source: World Bank staff calculations using VHLSS. 40 Viet Nam bi-annual poverty & equity update - June 2024 Annex A. Figures – Part 1  Figure 42 Figure 42. Decomposing monthly household income from wages Central Highlands Mekong Delta Change in household income (2017 VND, millions) Change in household income (2017 VND, millions) 25 25 20 20 15 15 10 10 5 5 0 0 -5 -5 -10 -10 -15 -15 -20 -20 -25 -25 2 4 6 8 0 2 2 4 6 8 0 2 -1 -1 -1 -1 -2 -2 -1 -1 -1 -1 -2 -2 10 12 14 16 18 20 10 12 14 16 18 20 20 20 20 20 20 20 20 20 20 20 20 20 Average earnings % HH earning income Average earnings % HH earning income Midlands and Northern Mountains Northern and Coastal Central Region Change in household income (2017 VND, millions) Change in household income (2017 VND, millions) 25 25 20 20 15 15 10 10 5 5 0 0 -5 -5 -10 -10 -15 -15 -20 -20 -25 -25 2 4 6 8 0 2 2 4 6 8 0 2 -1 -1 -1 -1 -2 -2 -1 -1 -1 -1 -2 -2 10 12 14 16 18 20 10 12 14 16 18 20 20 20 20 20 20 20 20 20 20 20 20 20 Average earnings % HH earning income Average earnings % HH earning income Red River Delta Southeast Change in household income (2017 VND, millions) Change in household income (2017 VND, millions) 25 25 20 20 15 15 10 10 5 5 0 0 -5 -5 -10 -10 -15 -15 -20 -20 -25 -25 2 4 6 8 0 2 2 4 6 8 0 2 -1 -1 -1 -1 -2 -2 -1 -1 -1 -1 -2 -2 10 12 14 16 18 20 10 12 14 16 18 20 20 20 20 20 20 20 20 20 20 20 20 20 Average earnings % HH earning income Average earnings % HH earning income Source: World Bank staff calculations using VHLSS. Towards more inclusive cities 41 PART 1. POVERTY & EQUITY UPDATE Annex B. Household There are two possibilities for these trends, lingering consumption behavioral changes from COVID-19 lockdown policies, and lower incomes. In 2020, 84 percent of households reported out-of-home meals in the regular consumption Household consumption is measured bi-annually using module, compared to 62.5 percent of households in 2022. This the Viet Nam Household Living Standards Surveys. sharp decline could be related to avoidance of public spaces during the pandemic. Second, households may have spent Average household consumption more carefully and focused more on essentials. Consumption declined declined most in luxury or processed food categories including milk, alcohol, ice cream, and cigarettes. At the same time, there The nominal consumption aggregate in 2022 is lower was a rise in produce consumption of fruits and vegetables, than 2020. Average consumption by components is more likely to be own-produced than processed items. summarized in Table 3. Across some categories, for some items, a decline in both the share of households consuming Expenditures on education and health are asked on and the amount of consumption was observed, such as food, a 12-month recall period, which would be impacted education, and healthcare. The reduction of consumption by 2021’s COVID-19 effects. On health expenditures, in these items could be related to residual impacts from the outpatient and inpatient expenditures declined, which is COVID-19 pandemic, when households avoided public consistent with healthcare avoidance during the pandemic facilities and restaurants. On the other hand, expenditures (Table 4). Education expenses include those related to required in some necessary items, such as utilities payments and fuel, schooling, and any voluntary or continuing education type increased. The most common coping mechanisms cited by expenses. Between 2020 and 2022, there was a small decline in households were reducing food and non-food consumption both the share of households spending on education, as well (World Bank COVID-19 household monitoring surveys). as the amount of education expenditure. Since the reference period for education expenditure is longer, over the last 12 Food consumption experienced the largest decline. months, it is likely that responses from some households still In 2022, in terms of average expenditure, food away from reflect reduced schooling during the pandemic, when some home (FAFH), pork, and rice were the largest regular food face-to-face learning was limited, and households spent less on expenditures (Figure 43). A decline in food consumption was extra education that required in-person participation. Some more likely in processed food items, while an increase was seen expenses increased such as the purchase of regular materials for items that could be own-produced. including paper, pens, bags, and other contributions to schools. Table 3. Household consumption, 2020 vs 2022 % HHs Average annual expenditure 2020 VHLSS 2022 VHLSS 2020 VHLSS 2022 VHLSS Food 100 100 55,134.8 51,961.1 Non-Food 100 100 44,468.5 43,375.7 Education 58.8 57.2 6,062.6 5,987.6 Health 89.8 85.5 6,706.4 4,736.5 Other non-food 100 100 26,321.3 26,301.7 Utilities 99.4 99.7 5,378.2 6,349.9 Implicit Rent 99.9 100 35,474.1 37,001.0 Durables ~100 ~100 20,980.7 19,988.7 Total 156,058.0 152,326.5 Note: Annual, in thousands VND. Nominal and not weighted. Source: World Bank. 2024b. Construction of the Viet Nam 2022 Consumption Aggregate note. 42 Viet Nam bi-annual poverty & equity update - June 2024 Annex B. Household consumption  Figure 43. Food expenditures, 2020 vs 2022 Note: Regular and holiday consumption combined. Right panel is in log-scale, and nominal VND. Source: Construction of the Viet Nam 2022 Consumption Aggregate note. Table 4. Average health consumption 2020 VHLSS 2022 VHLSS % HHs Average % HHs Average Outpatient expenditures 56.9 1,828.3 32.9 1,139.4 Inpatient expenditures 21.5 2,177.2 11.1 1,420.6 Medical tools 51.3 133.6 48.9 200.8 Aid for being sick 24.2 1,218.3 17.8 946.5 Health insurance 60.4 1,349.0 65.1 1,029.2 Total 89.8 6,706.4 85.5 4,736.5 Note: Nominal household level consumption. In annual VND in thousands. Unweighted. Average includes zeros. Source: World Bank. 2024b. Construction of the Viet Nam 2022 Consumption Aggregate note. Towards more inclusive cities 43 PART 2.  PART 2. Poverty and Inclusion in Urban Areas and Cities 44 June 2024 - Viet Nam bi-annual poverty & equity update Viet Nam is urbanizing rapidly. As the country continues to climb the development ladder, strategies to support urban livelihoods will gain importance. Less and less of Viet Nam’s population is reliant on rural livelihoods and farm income, and more are moving to urban areas. For its level of development and population density, the urbanization rate of Viet Nam is lower than expected, but the share of the population in an urbanized setting is higher when based on measurements using satellite imagery. From a social policy perspective, this means an increased focus is needed on living conditions and welfare in urban areas, especially since they are also centers of growth for the population and economic production. Importantly, there are gradients of urbanization, ranging from urban cores to suburbs that can hide heterogeneity in living conditions of “urban” areas and confound the extent, nature, and solutions to reducing urban poverty. Due to the heterogeneity of areas that are “urban” and the structure of governance, city-level analysis can be an important way to study variations in urban livelihoods, and devise practical policies related to investments, public services, and identifying risks. Granular within-city analysis is usually more challenging due to data constraints. But when feasible, it is much more informative and insightful about the distribution of socio-economic conditions. As a case study, spatial within city-level analysis of Ho Chi Minh City, the largest municipality of Viet Nam, is also presented. 1. Urbanization is important for the Next Mile 2. The urbanization of poverty 3. Case Study: Mapping spatial inclusion in Ho Chi Minh City 4. Data-informed policies in urban areas Towards more inclusive cities 45 PART 2. POVERTY AND INCLUSION IN URBAN AREAS AND CITIES 1. Urbanization is Globally, there is a strong correlation between important for the Next Mile urbanization, GDP, and poverty. Higher urbanization rates are correlated with both higher levels of economic Experiences in Viet Nam during the COVID-19 development and lower poverty rates (Figure 44). At Viet pandemic highlighted challenges and needed attention Nam’s level of urbanization, the average GDP per capita is to social conditions in urban areas and cities. Due to slightly higher than expected, and poverty rates based on the dense populations and shared public spaces, the health crisis World Bank’s UMIC poverty line are much lower. The role of spread faster, and lockdowns were more severe. Disruptions in urbanization in supporting the growth of the middle class as cities led to larger negative aggregate economic impacts, since well as the economy to higher income levels is well understood they are commercial hubs and centers of economic activity. (for the ASEAN region, see Brueckner et al., 2018). The In addition, other types of shocks, such as environmental agglomeration of individuals in cities leads to exchanges of disasters and flooding can lead to larger damage in urban ideas and technological growth (Duranton and Puga, 2014). areas due to more buildings and infrastructure. The concept Geographical proximity allows for knowledge spillovers of the “urbanization of poverty” has regained significance between people and between firms (Jacobs, 1985; Glaeser et post-pandemic. From 2010 to 2022, the urban share of al., 1992). In cities, individuals can access a wider variety of the population in Viet Nam increased from 30 to almost services and amenities, and firms experience productivity gains 40 percent. At the same time, the share of the poor (based (Combes et al., 2012). The matching of workers and jobs in on World Bank poverty lines) located in urban areas also cities result in what has been called an “urban productivity increased. As discussed in Part 1, the relatively larger rise in premium” (Andersson et al., 2007). Higher incomes and poverty within urban areas occurred during the period 2020- opportunities in cities and urban areas also support upwards 22, related to larger shocks and economic disruptions during economic mobility across generations. the pandemic. Figure 44. Relationship between urbanization, GDP, and poverty Notes: Values are circa 2020-22. Source: World Development Indicators (WDI). 46 Viet Nam bi-annual poverty & equity update - June 2024 1. Urbanization is important for the Next Mile  While Viet Nam’s official urban are heterogeneous and there is a gradient of urbanization, population is growing, the share ranging from urban cores to suburbs and non-urban areas. In Viet Nam, official urban areas include core urbanized areas, of people living in an urbanized but also suburbs. Within official rural areas, there are likely setting could be larger those that are more urbanized or better classified as suburbs or towns. While urban cores are usually well established, it is In the EAP region, Viet Nam’s official urban population the periphery and suburban areas that usually require more share is low given its level of development and population attention for development and need stronger policies for density. Compared to other countries in the developing EAP planning. Peripheral areas are more likely to be poorly served region, Viet Nam has a much higher population density, by public services, are less dense, and thus more costly to suggesting that urban population shares should be higher than expand services. These are also areas where many new migrants regional comparators (Figure 45). Among selected developing will reside, adding additional pressures to infrastructure. EAP countries, the Philippines and Viet Nam stand out as having higher population densities (panel A). On the other Despite population growth in Viet Nam, there has hand, Viet Nam’s urban population share is one of the lowest been limited reclassification of communes from rural among comparators (panel B). Only Lao PDR and Cambodia to urban. Based on the Population and Housing Census, have lower urban population shares. When gauging urban from 2009 to 2019, the national population grew from population shares by level of economic development, Viet Nam 78.6 to 91.5 million people. Rural areas had an additional again has low shares compared to other nations when at Viet five million people in 2019 compared to 2009, while urban Nam’s current level of GDP per capita (panel C). Thailand areas had about 7.8 million more people. In 2009, there were had a lower level of urbanization at Viet Nam’s current level of 1,944 urban communes, which increased to 2,203 in 2019. GDP per capita, but that was nearly 25 years ago. Very little of the population resided in communes reclassified from rural to urban between 2009 and 2019 (Figure 46). Whether or not people live in urban or rural areas is Administrative boundaries of municipalities are fixed, but meaningful for policies and strategies if these areas districts or communes can transform from rural to urban reflect distinct characteristics. But geographic areas classifications. Figure 45. Viet Nam’s urban population is low compared to regional benchmarks Source: World Development Indicators (WDI). Towards more inclusive cities 47 PART 2. POVERTY AND INCLUSION IN URBAN AREAS AND CITIES Classification of urban or rural areas is sometimes a picture of urbanization independent from administrative matter of administrative identification20. In Viet Nam, classifications. For example, using the Degree of Urbanization classifications are based on a range of parameters (see Box (DOU) methodology, the urban population share was 2.A). To understand socio-economic outcomes, it is the modeled to be as high as 73 percent, and 57 percent using the characteristics of areas not the classification that matters, D­­artboard methodology (DB) (Figure 47). Under both DOU as the former determines the quality of living standards and DB methods, the share of the urban population that reside and conditions for families and workers in these areas. in a core or center is lower than the official urbanization rate, Characteristics are more useful to distinguish between degrees indicating that peripheral areas are likely to be still considered of urban or rural areas, and to customize policies. There is rural in official classifications, but are rapidly urbanizing. Viet accepted recognition that there is a continuum of development Nam is not unique, other countries in a global study also and density from rural to urban areas (World Bank, 2009). found countries with low official urbanization rates compared With the availability of satellite data, new techniques are being to DOU or DB measures (Combes et al., 2022). developed to measure and study the gradient of urbanization. For example, areas can be classified using different techniques The rate of urbanization is an important consideration to consider population density, green space, and built-up area for policy direction in a Next Mile context. A large (see Box 2.B). There are also strong correlations between the urban population calls for more urban-focused policies. degree of urbanization and poverty rates. As more people move to cities and the country becomes denser, conditions in urban areas will have a larger role to The official proportion of the urban population in Viet play in poverty reduction and upwards economic mobility. Nam in 2022 was about 39 percent, but some research Sustaining poverty reduction will require effective has shown an urbanized population could be much management and provision of services in ever denser and higher when measured using different techniques. In crowded cities. The extent of populations living in urban areas official household and population surveys, the source of urban may be higher than already thought, meaning that stronger population and poverty statistics, urban areas are typically policy focus is needed for monitoring and supporting larger determined by commune-level administrative classifications. populations of people living in urban contexts. Classifications based on these dimensions may give a different Figure 46. Population, Figure Figureby 46 urban-rural classification 46 urban Figure 47. Share of Figurepopulation 47 Figure 47 in Viet Nam 2022, various classifications 100100 100 100 90 90 80 80 80 80 70 70 Population (millions) Population (millions) 60 60 60 60 % Population % Population 50 50 40 40 40 40 30 30 20 20 20 20 10 10 0 0 2009 2009 2019 2019 0 0 Rural commune Rural 2009, commune and 2009, urban and commune urban 2019 commune 2019 Official Official DBDB DOU DOU Urban commune Urban 2009 commune and 2009 2019 and 2019 Urban (official) Urban (official) Core (DB) Suburb (DB) Core (DB) Town (DB) Suburb Town(DB)(DB) Rural commune Rural 2009 commune and 2009 2019 and 2019 Urban center Urban (DOU) center (DOU) Urban cluster Urban (DOU) cluster (DOU) Rural (official) Rural (official) Source: 2009 and 2019 Population and Housing census Source: WB staff calculations using VHLSS 2022 and applying DB and DOU classifications from Nakamura et al.(2023) 20 See UN (2018) for a comparison of urban definitions across the world. 48 Viet Nam bi-annual poverty & equity update - June 2024 1. Urbanization is important for the Next Mile  A large share of Viet Nam’s urban In 2019, the five municipalities of Viet Nam were population and growth are found home to nearly a quarter (24.6 percent) of the country’s population, accounting for about 40 percent of the in its five municipalities urban population, while covering just 2.9 percent of Figure 48. Population trends across large cities in the country’s land mass. The two largest cities are the Figure 48 Viet Nam commercial hub of Ho Chi Minh City (HCMC) in the south, and the capital Hanoi located in the north. Haiphong 9 is a few hours’ drive east of Hanoi and is a key industrial area 8 and seaport. Danang and Can Tho municipalities have much smaller populations, but remain important economic and 7 commercial centers for their respective regions, the Northern 6 and Central Coast and the Mekong Delta. Population (millions) 5 Urban areas and cities in Viet Nam have witnessed 4 rapid population growth. Increasing urbanization has been spurred by rural-to-urban migrants seeking higher earnings 3 and better opportunities in large cities. HCMC and Hanoi 2 also stand out as having the largest growth in their populations. From 2009 to 2019, the population largely moved into or grew 1 in Hanoi and HCMC, with each city increasing by around 0 1.6 million people and higher growth seen in the latter half Can Tho Danang HCMC Haiphong Hanoi of the 2010s (Figure 48). From 2009-14, the total population 2009 2014 2019 of Viet Nam grew by 6.6 million people, with 16.7 percent Source: World Bank staff calculation using Population and Housing Census 2009, of net growth from just the five municipalities21 (Figure 49). 2014, and 2019. Figure 49 Figure 49. Distribution of population growth, by location (%) 2009-14 2014-19 Can Tho Danang HCMC Can Tho Danang HCMC Haiphong Hanoi Rest of the country Haiphong Hanoi Rest of the country Source: World Bank staff calculation using 2009, 2014, and 2019 Viet Nam Population and Housing Census. 21 There are 63 provinces in Viet Nam, of which five are considered municipalities. Towards more inclusive cities 49 PART 2. POVERTY AND INCLUSION IN URBAN AREAS AND CITIES In the latter half of the decade from 2014-19, the country’s migration less out-migration), and (iii) reclassification of population grew by 6.2 million people, with 40 percent existing populations between areas with the same urban of population growth witnessed in the five municipalities. and rural classifications over time. Each has implications Notably, 20 percent of total population growth from 2014- that require different policy discussions. 19 was in HCMC alone. Natural population growth has accelerated in Viet Population changes in the five municipalities can Nam’s two largest municipalities: Hanoi and HCMC. be attributed to various factors with different Representative of the net births and deaths occurring implications (Figure 50). Changes in population can in a location of interest, natural population growth was be decomposed into three factors: (i) natural growth of more robust in 2014-19 than 2009-14. While birth rates existing urban or rural populations, (ii) net migration (in- are declining nationally, because cities are also attracting new residents their children are also being born into these Figure 50 municipalities and boosting population growth. Hanoi Figure 50. Source of population change, 2009-19 and HCMC have the largest populations of children aged five years or under. In 2019, there were up to nine million 2.5 children in Viet Nam in this age group, with 9 percent living 2 in the capital and 7.2 percent in HCMC. 1.5 Net migration is slowing. Net migration to the Population (millions) 1 Southeast from 2004-09 was 117 per 1,000 persons, and 0.5 73 per 1,000 persons from 2014-19 (GSO, 2020). Possible reasons for the slowdown in migration could be that first- 0 tier cities are becoming too crowded (World Bank, 2020b). -0.5 Another explanation could be that larger proportions of -1 the population migrated earlier in the decade, suggesting that the bulk of those who had the opportunity to move C o g ng oi an Th M an ho an C H an aip to urban areas, did so early in Viet Nam’s structural D H C H transformation. These early movers may have then Out-migration In-migration Natural population growth Natural population in reclassified rural/urban areas permanently settled in their new locations. Yet, another possibility is that improved transportation has allowed for Notes: Reclassification of rural to urban communes. Source: World Bank staff calculation using Viet Nam Population and more transient migration or a floating population who Housing Census. travel regularly from their home to different provinces to temporarily engage in economic opportunities. Box 2.A. Urban and rural classification in Viet Nam In Viet Nam, administrative units are broadly divided into three tiers: provincial (level 1), district (level 2), and commune (level 3). At level 1, there are 63 provinces in Viet Nam, five of which are Central Cities (Municipality), and the remaining 58 are provinces. For municipalities, administrative divisions can include city, urban district (Quan), provincial town or district (Huyen). However, a municipality does not necessarily include all four types at the second administrative level, it can include just some. Whether an area is urban or rural is determined at the third level. There is a distinction between urban districts that contain only urban wards, and districts that contain district towns (urban) and communes (rural). 50 Viet Nam bi-annual poverty & equity update - June 2024 1. Urbanization is important for the Next Mile  Box 2.A. Urban and rural classification in Viet Nam (contd) Figure 51. Illustration of levels 1-3 administrative areas Figure 51 A. Municipality (Central City) - Structure of administrative levels Central City Urban Provincial District City district town Wards Wards Wards Communes District town Communes (Urban) (Urban) (Urban) (Rural) (Urban) (Rural) B. Province – Structure of administrative levels Province Provincial Provincial District town city Wards Communes Wards Communes District town Communes (Urban) (Rural) (Urban) (Rural) (Urban) (Rural) How to define urban (ward/district town) and rural (commune) at the commune level? Whether an area is urban or rural is determined at the third administrative level. As shown in Figure 51, urban areas are either wards or district towns, while communes are rural. The definition and criteria for defining these areas are shown in Table 5. Table 5. Criteria for urban wards, and district towns, or rural communes Level 3 administrative area 200922 201923 Mountainous areas Delta area Mountainous areas Others Commune (rural) Population can be Population can be below Population above Population above 8,000 below 1,000 people, 2,000 people, or above 5,000 people people or above Areas can be below 500ha Areas above 50km2 Areas above 30km2 Areas can be below or above 1,000ha or above 22 Population Census 2009: Based on Decree No. 159/2005/NĐ-CP to define wards/district towns and communes. 23 Population Census 2019: Based on Decree No. 42/2009/NĐ-CP to define city/town. Towards more inclusive cities 51 PART 2. POVERTY AND INCLUSION IN URBAN AREAS AND CITIES Box 2.A. Urban and rural classification in Viet Nam (contd) Table 5. Criteria for urban wards, and district towns, or rural communes Level 3 administrative area 200922 201923 Mountainous areas Delta area Mountainous areas Others Ward Under the district Population can be below 3,000 people, or above Population above 15,000 people (urban) Areas can be below 500ha or above Non-agricultural labor ratio above 85% Areas above 5.5km2 Under the provincial city Population can be below 3,000 people, or above Population above 7,000 people Under the city Areas can be below 500ha or above Non-agricultural labor ratio above 80% Under the central city Areas above 5.5km2 Under the provincial Population can be below 3,000 people, or above Population above 5,000 people town Areas can be below 500ha or above Non-agricultural labor ratio above 70% Areas above 5.5km2 District town (urban) City class 5 City class 4 and 5 Source: 2009 classifications based on Decree No. 159/2005/NĐ-CP came into effect in 2005. 2019 classifications based on Resolution No. 1211/2016/ UBTVQH13 came into effect in 2016, Law on Local Government Organization revised 2019. Box 2.B. DB and DOU classifications Due to the increased availability of satellite imagery, there is growing literature about the consistency of classifications of urban areas (Duranton and Rosenthal, 2021). This chapter discusses two approaches, the Degree of Urbanization (DOU) and the Dartboard Methodology (DB). These two methods are not the only ones available, but were used as part of a global urban poverty analytical activity that included Viet Nam. More information about the World Bank’s global exercise can be found in Nakamura et al. (2023). Table 6 summarizes criteria for the different gradients of urbanization as measured by the DOU and DB methods. The DOU method differentiates urban areas into urban centers and urban clusters. The DB methodology differentiates urban areas into three groups: core, suburb, and town. The DOU approach is rooted in uniform population and population density thresholds using gridded cells, and is also endorsed by the United Nations for international statistical comparisons (European Commission et al., 2020). The DB methodology compares the actual distribution of population density to the counterfactual distributions based on randomly shuffled grids. The areas identified as urban have much higher population densities than when randomly distributed. 52 Viet Nam bi-annual poverty & equity update - June 2024 1. Urbanization is important for the Next Mile  Box 2.B. DB and DOU classifications (contd) Table 6. Differentiation of urban areas Degree of Urbanization (DOU) Dartboard Methodology (DB) Urban Spatially contiguous sets of 1km grid cells 2 Urban Sets of contiguous grids that have high density (> 95th percentile) Urban Urban + Core Identified as contiguous second-order urban center pixels based on re-shuffling (2,000 random • Population density of each cell ≥ 1,500 people per km 2 reshuffles) within urban areas • Aggregate population ≥ 50,000 Urban Urban + Suburb Non-core parts of cities cluster • Population density of each cell ≥ 300 people per km 2 • Aggregate population ≥ 5,000 • Includes suburbs of urban centers and towns Town Urban areas without a core Note: See Dijkstra & Poelman (2014) & Dijkstra et al. (2021); endorsed by the United Nations in March 2020 for DOU approach (EC, 2020). See Bellefon et al. (2021) for Dartboard Methodology approach. In the case of Viet Nam, the urban share of the population is the highest when using the DOU method. The DOU method tends to result in the lowest rural population shares and largest populations in urban centers. DOU tends to be high in Viet Nam due to its overall high population density. Figure 52 illustrates the distribution of the urban and rural populations across regions and classifications. As expected, the Red River Delta and Southeast regions have the highest share of urbanization rates based on both the DB and DOU methods, since these regions contain Hanoi and HCMC, respectively. Interestingly, the DOU and DB urbanization rates of the Mekong Delta also rises. This could be due to large portions of land being designated for agricultural land use, and thus residential areas are more clustered and have higher population densities. In the poorest regions of Viet Nam, the Central Highlands and the Midlands and Northern Mountains regions, the DOU and DB methods result in similar shares of urban and rural populations compared to the official rates. Towards more inclusive cities 53 PART 2. POVERTY AND INCLUSION IN URBAN AREAS AND CITIES Box 2.B. DB and DOU classifications (contd) Figure 52 Figure 52. Distribution of urban and rural populations, by different classifications 100 80 60 % Population 40 20 0 Official DB DOU Official DB DOU Official DB DOU Official DB DOU Official DB DOU Official DB DOU Official DB DOU Central Highlands Mekong Delta Midland and Northern and Red River Delta Southeast Viet Nam Northern Coastal Center Mountains Urban Core Suburb Town Urban center Urban cluster Rural Source: VHLSS 2022 using DB and DOU classifications from Nakamura et al.(2023). 2. The urbanization of poverty “pancake spread”24, where development is low lying. Cities in developing countries tend to expand flat and horizontally, Viet Nam has a high population density, with the quality but vertical layering is what is needed for agglomeration of urbanization now more important than its extent and to make modern cities more productive (Lall et al, for continued development to high-income status. 2021). In Viet Nam, rapid urbanization has led to sprawling Benefits from urbanization can be smaller if urban areas are development, declining returns to agglomeration, lower not well planned, public services are inadequate for growing labor productivity, and increasing congestion in major cities populations, or there is inequity in access to services and (World Bank, 2020b). Higher urban poverty rates can occur utilities. While Viet Nam’s ambition is to reach high-income when crowding or congestion outweigh benefits from density status by 2045, some development challenges are more urgent and agglomeration (Lucci et al. 2018, Marx et al. 2013). The in urban than rural areas. For example, crowded informal Government of Viet Nam recognizes the importance of urban housing conditions, higher living costs, congested traffic, development for the country’s overall growth25. However, lack of mobility and access to jobs, or air pollution are more urban planning is still a challenge, and lacks an institutional likely to be risks and challenges for urban populations. Much structure to support strategic urban development (World of urban growth in developing countries is referred to as a Bank, 2011 and 2020b). 24 Pancake development is low and outward, while pyramid development expands outwards but also fills in pockets in urban centers and notably with tall buildings. 25 https://en.vietnamplus.vn/urbanisation-a-chance-for-breakthrough-development-pm/244740.vnp Government’s Resolution No. 148/NQ-CP, issued on November 11, to carry out the Politburo’s Resolution No. 06-NQ/TW (dated January 24), on the planning, construction, management, and sustainable development of Vietnamese cities by 2030, with a vision to 2045. 54 Viet Nam bi-annual poverty & equity update - June 2024 2. The urbanization of poverty  Under certain conditions, urban urban poverty rates rising more than rural poverty ones. In poverty can rise more than rural the case of Viet Nam, this occurred during the pandemic, as discussed in Part 1 of this report. Second, it can be due poverty to the dynamic nature of (or less successful) rural-to-urban Certain risks and crisis can have larger negative impacts migration that is spurred by higher incomes and better wages on cities and urban areas, and result in higher damage in urban areas. However, when new urban residents do not or costs. The COVID-19 pandemic was a recent example of a initially achieve incomes in a similar range as established crisis that impacted urban areas more than rural ones. Due to urban residents, urban poverty rates can also be pushed dense populations and shared public spaces, health crisis can higher. This has been referred to as an incomplete Kuznets spread faster in cities. Disruptions in cities led to larger negative curve (Ravallion et al., 2007). There are a range of factors aggregate economic impacts, since they are commercial hubs and related to less successful assimilation into urban areas, to be centers for economic activity. Environmental events can also lead discussed in later sections of this report. Whether longer- to greater damage in cities with more infrastructure. Natural term migrants achieve better outcomes than recent migrants hazards, such as flooding, are a common risk in large Southeast cannot be assessed from data in Viet Nam26, but is the case in Asian cities in delta areas. These challenges are greater when there other countries, such as Indonesia (World Bank 2018b). is a prevalence of lower quality housing or crowded quarters without adequate flood management infrastructure. A recent In Viet Nam, the share of the population and poor in study published in 2022 found that 46 percent of Viet Nam’s urban areas has increased in recent years. From 2010 population resided in flood zones, the third highest in the world to 2022, the share of the population in Viet Nam located after the Netherlands and Bangladesh (Rentschler, Salhab, and in administrative urban areas increased from 30 percent Jafino, 2022). Viet Nam’s largest city, Ho Chi Minh City, is one to almost 40 percent (Figure 53). Urban poverty rates also of the top 10 cities in the world with the largest population most declined throughout the decade, driven by high population likely to be affected by climate change. and wage growth. The concept of the “urbanization of poverty” has Figure 53. Share of population in urban areas is regained significance post-pandemic. Ravallion et al. Figure 53 increasing, as is the urban poor (2007) initially reviewed empirical claims related to trends in the urbanization of poverty, namely if: 1) the urban sector’s 45% share of poor is rising over time, 2) the poor are urbanizing 40% faster than the population as a whole and 3) population 35% urbanization is a positive factor in overall poverty reduction. In this global study, the poor were found to be urbanizing 30% Urban share (%) faster than the population, and that urban poverty rates 25% were not declining as quickly as it was among the entire 20% population. There is evidence that urban poverty might 15% increase with urbanization, and that this type of poverty has particular characteristics that might make it more challenging 10% to tackle and identify than rural poverty (Mathur, 2013; 5% Sridhar, 2015). While in rural areas poverty rates seem to be 0% more directly dependent on agriculture and specific types of 2010 2012 2014 2016 2018 2020 2022 jobs, it is usually more difficult to determine the root causes Share of LMIC poor Share of UMIC poor of urban poverty (Alcantara et al., 2023; Benfica et al., 2021). Share of total population Relative lower reductions in urban poverty can be due Note: Official urban and rural classifications. to different processes. First, unique conditions can lead to Source: World Bank staff calculations. 26 For migration-related information, only residence in the last five years is available. Residence in the last 10 years or location of birth is not available. Towards more inclusive cities 55 PART 2. POVERTY AND INCLUSION IN URBAN AREAS AND CITIES Figure 54 Figure 54. Stylized illustration of poverty, by urban or rural classification Panel A Panel B Urban 2% Urban core 2% Urban core 2% Urban 3% Peri-urban 4% Peri-urban 4% Rural 7% Rural 10% Rural 10% Rural 10% Source: Author’s illustration. Recently, the share of poor in urban areas increased at 55. 55 Figure Figure Share of the UMIC poor by urban both Lower-Middle income (LMIC) and Upper-Middle classifications, Viet Nam 2016 income (UMIC) poverty lines defined by the World 100 Bank (see Box 1.A in Part 1 for definitions). As discussed in Part 1, the relatively larger change in poverty in urban 90 areas occurred during 2020-22, related to pandemic-triggered 80 shocks and economic disruptions in urban areas. 70 Share of the UMIC poor (%) While urban poverty rates are 60 much lower than rural ones, 50 they can vary by different 40 measurement or classifications 30 20 The classification of areas as either urban or rural 10 affects their poverty rates. When reclassifying peri-urban 0 areas from rural to urban, both urban and rural poverty Official DOU DB rates increase. It may seem counterintuitive that subgroup Urban (official) Core (DB) Suburb (DB) Town (DB) Urban center (DOU) Urban cluster (DOU) Rural (official) poverty rates for both areas can increase, so an example is used to illustrate the point. As a stylized example, consider Note: World Bank Upper Middle income poverty line is $6.85/day 2017 PPP. the simple chart in Figure 54, where poverty rates across Source: Nakamura et al. 2023. three classifications are the highest in rural areas (10 percent), slightly lower in peri-urban areas (4 percent), and the lowest in An important finding from analysis using an the urban core (2 percent). Assume that peri-urban areas are international database of urban poverty, is that there is newly urbanized and have historically been classified as rural a gradient of poverty rates across levels of urbanization areas. For simplicity, if the population in each geographic (Nakamura et al, 2023). The share of poor in urban areas group is the same, the rural poverty rate including peri-urban is higher when considering urban clusters or suburbs to be areas would be 7 percent, compared to an urban poverty rate urban areas instead of rural areas (Nakamura et al., 2023). of 2 percent (panel A). If peri-urban areas are re-classified as This is also the case in Viet Nam. After reclassification, a urban, then the urban poverty rate including peri-urban areas larger share of poor reside in urban clusters (DOU method) would be higher at 3 percent (panel B). Thus, both the rural or suburbs (DB method) and not in rural areas (Figure 55). and urban poverty rates are higher when peri-urban areas are re-classified from rural to urban. 56 Viet Nam bi-annual poverty & equity update - June 2024 2. The urbanization of poverty  In Viet Nam, there is a gradient of poverty across Non-rural populations (by DB or DOU methods) outside of various urbanization classifications. In the case of Viet the urban core are a mix of urban and rural populations from Nam, across various measures of urban areas, urban cores official classifications. Their poverty rates are between official consistently have the lowest poverty rates as shown in Figure rural and urban poverty rates. Based on alternative DOU and 56. However, poverty rates outside the core are much higher. DB classifications, rural poverty is also higher, since more For example, following the DB method, 2022 UMIC poverty urbanized areas in official rural areas are most likely to be re- rates in towns were 24 percent, in contrast to less than 5 classified into urban areas. The DOU method recategorizes percent in urban cores. Based on the DB method, most of the largest share of rural population into urban categories, the urban core population is in the Red River Delta and leaving a smaller share of remaining rural populations. Across Southeast regions where Hanoi and HCMC are located, rural classifications, DOU rural poverty rates are the highest respectively. Recent dynamics in poverty rates in the urban since the remaining rural populations are likely to be living in cores/centers (panels B and C) are also smaller than the change the most sparsely populated remote regions, which tend to be in urban poverty based on official classifications (panel A). associated with higher poverty rates. Figure 56 Figure 56. Gradient of poverty across urbanization categories A. Official B. Dartboard C. Degree of urbanization 60 60 60 50 50 50 Poverty rate (%) Poverty rate (%) Poverty rate (%) 40 40 40 30 30 30 20 20 20 10 10 10 0 0 0 2018 2020 2022 2018 2020 2022 2018 2020 2022 Rural Urban Rural Suburb Town Core Rural Urban cluster Urban center Note: World Bank Upper Middle income poverty line is $6.85/day 2017 PPP, using 2018 classifications. Source: WB staff calculations using classifications from Nakamura et al. 2023. Figure 57 Figure 57. Profiling by different classifications of urbanization A. Ethnicity – Ethnic minority B. Age - Household head greater than 50 100 100 90 90 80 80 Share of the population (%) Share of the population (%) 70 70 60 60 50 50 40 40 30 30 20 20 10 10 0 0 Official DOU DB Official DOU DB Urban (official) Core (DB) Suburb (DB) Town (DB) Urban (official) Core (DB) Suburb (DB) Town (DB) Urban center (DOU) Urban cluster (DOU) Rural (official) Urban center (DOU) Urban cluster (DOU) Rural (official) Note: In panel B, households are split by the age of the household head. Source: WB staff calculations. Towards more inclusive cities 57 PART 2. POVERTY AND INCLUSION IN URBAN AREAS AND CITIES The profile of the population also varies across look at urban areas in a more disaggregated way. There is urbanization categories. Importantly, some profiles are a gradient of urbanization, ranging from urban cores to more affected by reclassification than others. For example, suburbs that can hide heterogeneity in living conditions of among ethnic minorities, their populations are primarily “urban” areas and confound the extent, nature, and solutions concentrated in rural areas, across all three classification for reducing urban poverty. From a policy perspective, it can groups (official, DB, and DOU) (Figure 57, panel A). The be useful to examine living standards by cities (municipality) low share of ethnic minorities in non-rural areas across rather than by urban areas in general. In Viet Nam, five city- classifications is important for illustrating that location is level municipalities (Can Tho, Danang, Haiphong, Hanoi, a strong structural determinant of higher poverty among and Ho Chi Minh City) accounted for about 40 percent of ethnic minority households. While for other characteristics, the official urban population in 2019. Policies are made and reclassification has stronger re-distribution implications. For resources are managed at the municipality level, in some example in panel B, the share of households with heads aged cases a city-level analysis is more useful than an urban-level 50 years or older is more distributed across peri-urban areas analysis for planning and decision-making. When exploring when using the DOU and DB classifications. the benefits from agglomeration, challenges of service delivery or financing, these discussions also make sense at a city level. Survey-level indicators mask 3. Case Study: Mapping variation across a highly spatial inclusion in Ho Chi populated city Minh City Within-city analysis requires larger samples of data than usually available in traditional surveys. Moreover, City-level analysis is important some survey samples in Viet Nam are small. For example, to understand the gradient of Lao PDR had a population of 7.4 million in 2021, and its 2018 household survey interviewed 23,000 households. urbanization and poverty27 In Indonesia, the SUSENAS has a sample of more than Municipalities containing urban cores, suburbs, and 1,000 households in each of the five districts within Jakarta outer peripheries have a gradient of poverty and living (population of more than 10 million) and then additional conditions, requiring more granular analytics to inform similar-sized samples for districts which form the urban policy making. Some of Viet Nam’s biggest cities have periphery. Comparatively, HCMC had an official population larger populations than neighboring smaller countries. In of about nine million in 2019, yet in the annual Viet Nam dense urban areas, having a clear understanding of the spatial Household Living Standards Surveys just 1,400 HCMC distribution of population characteristics is useful to design households make up the 45,000 sample. policies efficiently. For example, knowing the percentage of people living in poverty in a city or urban area, in general, is This section examines HCMC as a case study not as useful as understanding how the poor are distributed highlighting examples of insights possible from throughout the city. granular spatial city-level analysis. It is a good example of a diverse metropolis with a gradient of rural to urban Due to the heterogeneity of areas that are “urban”, and areas. By official counts, almost 80 percent of the population the structure of governance, city-level analysis can be an reside in its urban districts, with the remainder in rural important way to study variations in urban livelihoods. districts. However, even in its urban districts, there is Based on the reasons described earlier in Part 2 regarding the variation in poverty, housing characteristics, education classification of geographic areas and the gradient of urban and demographics. This is especially true in extremely areas with different degrees of urbanization, it is useful to large districts. In the case of HCMC, several districts are 27 Cities can comprise of rural and urban districts and communes, from an administrative perspective. Cities and municipalities are used interchangeably. 58 Viet Nam bi-annual poverty & equity update - June 2024 3. Case Study: Mapping spatial inclusion in Ho Chi Minh City  home to more than half a million people each. This section Poverty rates in HCMC are some of the lowest in Viet describes household conditions in HCMC areas with added Nam, but there can be variations in living quality granularity, examining the distribution of the population, and standards across the municipality. HCMC is a poverty, and household characteristics. heterogeneous city featuring rural and urban districts, and a wide variation of living conditions. The distribution of For spatial granularity and comparison, trends are shown the population across the city varies by average education across grid cells (1km x1km), as well as at the district-level completion, concentration of migrants, housing features, (see Box 2.C. for more information about the HCMC assets, and wealth, amongst others. Some characteristics have population). In the annual Labor Force Survey and the Viet greater dispersion than others. Higher-quality amenities and Nam Household Living Standards Survey, HCMC statistics assets are less common in the city than lower-quality ones. are available at the municipality level. Small area estimations For example, while virtually all households own a refrigerator, can be used to compute estimates down to the district level, of fewer own a computer at home and they tend to live in the which there are 24 districts in HCMC. However, given such city center (Figure 58). HCMC rural districts have the lowest a large municipality, survey-level indicators mask variations ownership of computers. and the grid perspective reveal important insights into the distribution of population characteristics. Grid cell values are created using non-survey data. Figure 58. Household assets Note: Averages and sums are at grid level. Grids with less than population of 100 are not shown. Source: WB staff calculations. Towards more inclusive cities 59 PART 2. POVERTY AND INCLUSION IN URBAN AREAS AND CITIES As another example, there are variations in education completion, there are noticeable differences in the levels of completion across and within districts. The east-center dissimilarity. As Figure 59 (panel B) shows, northern rural area of HCMC features the highest share of households areas show high levels of dissimilarity, as do rural districts to whose heads have completed at least upper-secondary the west, while central areas exhibit lower levels of within- education (Figure 59, panel A). When looking at within- district dissimilarity. district segregation of households without high school Figure 59. Education spatial distribution A. Share of household heads with education completion above upper-secondary B. Upper-secondary completion dissimilarity index, by district Note: Averages and sums are at grid level. Grids with less than population of 100 are not shown. Source: WB staff calculations. Ho Chi Minh City is the largest city in Viet Nam, with a gradient of urbanization Box 2.C. and characteristics Ho Chi Minh City is home to nearly nine million people (2019 census) with a land area covering only 0.6 percent of the entire country. Its 24 districts have varied population densities ranging from the hundreds to the tens of thousands. The highest population density is in the urban core at almost 40,000 persons per km2 (Figure 63). Eight out of 24 districts have population densities above 30,000 persons per km2. Quan 4, a small central district, has the highest population density at more than 41,000 people per km2. While the population density is highest in the center as expected, most of the population lives outside the city center (Figure 64). From 2009 to 2019, the peripheral areas featured the most population growth, with central districts even experiencing small declines in population (Figure 60). This is consistent with the shift of the new built-up area in HCMC moving outwards as well. 60 Viet Nam bi-annual poverty & equity update - June 2024 3. Case Study: Mapping spatial inclusion in Ho Chi Minh City  Ho Chi Minh City is the largest city in Viet Nam, with a gradient of urbanization Box 2.C. and characteristics (contd) Figure 60 Figure 60. Net population change from 2009 to 2019, by district 300 250 Population change (thousands) 200 150 100 50 0 -50 -100 Quan 1 Quan 11 Quan 6 Quan 5 Quan Phu Nhuan Quan 4 Quan 3 Quan 10 Huyen Can Gio Quan 8 Quan Binh Thanh Quan 2 Quan Tan Binh Quan Tan Phu Quan 7 Huyen Nha Be Huyen Cu Chi Quan Thu Duc Quan 9 Quan Go Vap Quan Binh Tan Huyen Hoc Mon Quan 12 Huyen Binh Chanh Source: WB staff calculations. In HCMC, there is a gradient of urbanization and even rural districts. The city districts can be labeled into different groups: core districts, districts adjacent to core, outer urban districts, and official rural districts (Figure 62). The most populous districts in HCMC also include some lower density ones much larger in area, such as Huyen Binh Chanh, Quan Binh Tan and Quan Go Vap, located outside the city core (Figure 65). Officially, six of HCMC’s districts are still considered rural. These rural districts have a much lower population density. In particular, the majority of Huyen Can Gio in the south is traversed with waterways, sparsely or completely unpopulated, and has a population density of only 96 people per km2, yet it accounts for 32 percent of HCMC’s total land area. There are also smaller pockets in the north and peripheries of the city that are unpopulated residential areas. Demographics vary across HCMC. The eastern districts in HCMC have closer proximity to Binh Duong province and industrial parks, and also a younger population, consistent with having large populations of workers in industrial parks. Nearly a quarter of households in HCMC have a dependency ratio higher than 0.5, the threshold considered a deprivation in the national multidimensional poverty index for 2021-25. The west, south, and north peripheries of the city have higher dependency ratios, indicating a presence of family compositions with more members younger than 15 or older than 64 years. Households in these peripheries have fewer members in prime working age, and may be indicative of areas with fewer employment opportunities. Towards more inclusive cities 61 PART 2. POVERTY AND INCLUSION IN URBAN AREAS AND CITIES Ho Chi Minh City is the largest city in Viet Nam, with a gradient of urbanization Box 2.C. and characteristics (contd) Figure 61. The 24 administrative districts of HCMC Figure 62. HCMC districts, grouped and mapped Figure 63. Distribution of HCMC’s population Figure is Figure 64. Population 64lowest in the core, 2019 using the grid 4,000 35 3,500 30 Population density (thousand/km2) 3,000 25 Population (thousands, bar) 2,500 20 2,000 15 1,500 10 1,000 500 5 0 0 Core Districts adjacent Outer Rural districts to the core urban districts Districts Population (thousands) Population Density 62 Viet Nam bi-annual poverty & equity update - June 2024 3. Case Study: Mapping spatial inclusion in Ho Chi Minh City  Ho Chi Minh City is the largest city in Viet Nam, with a gradient of urbanization Box 2.C. and characteristics (contd) Figure 65 Figure 65. HCMC districts vary by population and density, 2019 800 45 700 40 600 35 Population density (dot) Population (bar) 30 500 25 400 20 300 15 200 10 100 5 0 0 Huyen Can Gio Huyen Cu Chi Huyen Nha Be Huyen Binh Chanh Quan 9 Quan 2 Huyen Hoc Mon Quan 7 Quan 12 Quan Thu Duc Quan Binh Tan Quan 1 Quan Tan Binh Quan 8 Quan Binh Thanh Quan Tan Phu Quan 6 Quan Go Vap Quan Phu Nhuan Quan 5 Quan 3 Quan 10 Quan 11 Quan 4 Population (thousand) Population density (thousand/km2) Figure 66 A simple framework for within- Figure 66. Framework city analysis Livelihoods/ income The next sections examine spatial trends of HCMC • Wealth Index using additional indicators and following a simple framework. The framework utilizes selected indicators as illustrated in Figure 66 and also described in Box 2.D. As the Human Capital framework shows, channels affecting household welfare can Household • Access to school Environment Welfare • Flooding be direct or indirect. Access to amenities and housing features • Access to hospital impact the quality of daily life. Investment in human capital and levels in younger generations are strongly associated with Living Standards • Access to bus economic outcomes in later years. The wealth index is a proxy • Overcrowding (housing) for household purchasing power and income. Environmental • Housing quality • Access to parks risks28, such as exposure to floods, can also erode household welfare through damage to household assets or limiting Source: Author’s illustration. transportation and access to jobs. Identification of smaller pockets of wealth or poverty index is District 2, where 45.5 percent of the population is in is feasible with more granular data. The left panel of HCMC’s top 20 percent based on a constructed wealth index. Figure 67 illustrates district-level averages of the wealth score Conversely, the poorest district is Huyen Can Gio, where in HCMC. The district with the highest average wealth only 3.1 percent of the population is in HCMC’s top 20. 28 Air pollution based on PM2.5 was considered, but not included since HCMC’s average PM2.5 levels are below the national targets of 30 micrograms per cubic meter. Towards more inclusive cities 63 PART 2. POVERTY AND INCLUSION IN URBAN AREAS AND CITIES Results from the asset-based wealth index reveal more the examination of spatial segregation based on education levels variation at grid level than at district level. In the right panel, holds significance. The spatial distribution of various education there is much more nuance to the variation in the wealth strata within a city carries far-reaching implications for societal score, but it is also clear there are sparse or uninhabited areas. cohesion, economic dynamics, and the equitable availability of The grid-level view in the right panel reveals pockets of poor opportunities. Access to amenities is a measure of the equitable in the western periphery as well as in the lower peninsula of distribution of public services and infrastructure. Evidence HCMC not visible from district-level information. Similar shows that people are willing to pay more to live in closer to findings in Nakamura et al. (2023), there is a gradient of proximity to certain amenities (McLeod, 1984; Letdin & Shim, wealth from urban cores having the lowest rates of poverty to 2019). This can lead to differences in accessibility between rural areas having the highest poverty rates. those who can or cannot afford to be closer to better amenities. Access to public transport29 is important for the poor and A city is more inclusive if residents can participate linked to having better job access. Compared to other countries in markets, services, and spaces to the same degree. in a 2016 global study, populations in the municipalities in Viet Inequality in urban areas and cities is generally higher than in Nam have lower usage rates of public transport (UN Habitat, rural areas. Access to jobs, transportation, health, education, 2016). While motorcycle ownership is almost universal across and recreation play an important role in determining where households, there can still be mobility challenges for the elderly people want to live within an urban area. In urban contexts, or disabled groups in the population. Figure 67. Map of the median asset-based wealth scores, by district and grid Notes: Dark orange represents lower wealth scores and blue represents higher wealth scores. Grids with less than a population of 100 are not shown. Source: WB staff calculations. 29 Access to urban transport is SDG indicator 11.2.1. 64 Viet Nam bi-annual poverty & equity update - June 2024 3. Case Study: Mapping spatial inclusion in Ho Chi Minh City  High-quality housing plays a vital role in ensuring the Pluvial flooding usually occurs during the summer monsoon well-being and stability of communities. Additionally, season, when the easterlies discharge moisture from the East it often boosts property values, contributes to a sense of Sea, implying even higher water levels in extended areas, and community, and facilitates provision of essential services such stalling the drainage valves system of the city. This causes as healthcare, education, and emergency responses. Housing economic damage and disruptions to daily life and business, quality in HCMC is examined using quality of roofing and and intangible losses from traffic congestion and delays. wall materials, as well as the source of drinking water and Annually, the city receives about 75 inches of precipitation, toilet infrastructure. Access to electricity is not considered as which is about average nationally. While areas close to the river it is universal. Another consideration for adequate housing in the east are at-risk of fluvial flooding, the majority of the is overcrowding. The National Housing Development city is at risk from pluvial flooding to some extent. More than Strategy30 sets a target for 2025 to reach average housing areas half of the city’s area is less than 2 meters above sea level, with of 28m2 per person in urban areas and 26m2 in rural areas. little above 4 meters. HCMC is one of the top 10 cities in the world with its population most likely to be affected by climate Environmental risks can impact household income change, with the frequency and strength of storm surges likely directly and indirectly. Flooding is a major environmental to increase. This is compounded by an increasing population risk for the population of HCMC and occurs annually, density that is adding pressure on urban infrastructure. often due to a combination of heavy rainfall, discharge from upstream reservoirs, and storm surges coinciding with high The next sections describe the spatial distribution of tides. Analysis finds that flooding from heavy rain (pluvial) each indicator individually (from Table 7), followed by a is more severe than from unmitigated river flooding (fluvial). discussion of a composite index mapped across HCMC. Box 2.D. Data While indicators selected for the HCMC case study are based on data availability, they are meant to serve as an example to illustrate the landscape of living conditions across a diverse city. Indeed, indicators yield meaningful patterns of living quality across HCMC, which are discussed in later sections. Some indicators tend to be more spatially based than a reflection of household demographics. For example, four of the indicators are distance to amenities (bus stations, hospitals, schools, and parks). Flooding is a regular event in HCMC, and a measure of flood depth for 1-in-100-year floods is used to reflect flooding exposure in the grid cell. Finally, some indicators vary by household. Three indicators are measured at the household level related to housing features, and an asset-based wealth index. Table 7. Selected indicators for HCMC Group Indicator Measure Unit Livelihoods and income Wealth index Asset-based wealth index Household level Access to public transport (bus Distance to bus stops (km) Grid level stops) Human capital Access to schools Distance to schools (km) Grid level Access to hospitals Distance to hospitals (km) Grid level 30 Decision No. 2161/QD-TTg, 2021-2030 National Housing Development Strategy with a vision towards 2045. Towards more inclusive cities 65 PART 2. POVERTY AND INCLUSION IN URBAN AREAS AND CITIES Box 2.D. Data (contd) Table 7. Selected indicators for HCMC Group Indicator Measure Unit Living standards Housing quality Water, toilet, wall, and roof Household level materials Housing overcrowding Dwelling area per capita (m2) Household level Access to parks Distance to parks (km) Grid level Environment Flood exposure Average flood depth (cm), for a Grid level 1-in-100-year flood Source: Authors Distance to amenities The locations of hospitals, schools, parks, and bus stations are extracted from Open Street Maps (OSM). To achieve this, a bounding box for HCMC is defined, and requests are passed onto the OSM API for the different categories of points of interest. This resulted in 1,747 polygons corresponding to hospitals, 10,045 schools, 12,376 parks, and 410 bus stops. With this information, the Euclidean distance between each point of interest and each square of the 1km-by-1km grid corresponding to the area of HCMC is calculated. Environment Flood data for Ho Chi Minh City is computed from Fathom flood data. Fathom is a private company that synthesizes terrain data and hydrologic modeling to compute detailed flood datasets at 30m resolution. Floods risks consider two sources: 1) Fluvial – flooding from river water; and 2) Pluvial – flooding from rainfall. Analysis features a comparison of 1-in-10-year (10 percent likelihood) and 1-in-100-year (1 percent likelihood) extents. The spatial statistics derived from these unique inputs distinguish susceptibility to a more probable occurrence (1-in- 10-year) at a comparably lower magnitude, and less probable occurrence (1-in-100-year flood) asserting greater risk. To differentiate hazard vulnerability and the effect on risk, four flood depth categories were created for each return period. Any inundation depth above ‘0’ fundamentally informs where any amount of rainfall or riparian waters may reach. This spatial result suggests the greatest magnitude considering the extent of inundation. The unmapped area is equally informative to identify perennial dry locations. Depths greater than 10cm, but less than 20cm, defines an inconvenient depth that may not cause significant damage, but could impede social movement and activity around uncovered outdoor markets. Depths greater than 20cm, but less than 50cm categorizes a flood depth that exceeds the average height of a sidewalk. An unmitigated flood hazard at this scale will proliferate vulnerability to pedestrian and motor traffic and may increase risk to undefended stores. Depths greater than 50cm present the greatest magnitude, considering flood depth, evaluated within this report. This threshold may present a physical risk to agriculture and inundated vegetation in addition to a heightened vulnerability of infrastructure and population. 66 Viet Nam bi-annual poverty & equity update - June 2024 3. Case Study: Mapping spatial inclusion in Ho Chi Minh City  Across HCMC, distances live closer to schools and parks, but this decreases significantly to amenities are shorter in when considering distance to bus stops, with less than one million people living within a 1km range. In addition, Figure peripheral areas of the city 69 also shows that average proximity to the four types of Distribution of services varies. Unsurprisingly, the central amenities is higher among the poorest quintiles in HCMC, area of HCMC exhibits better access to all four amenities when based on a constructed wealth index. measured in this case study: hospitals, schools, parks, and bus stops (Figure 68). This can be attributed to the concentrated There are pockets of presence of these facilities within the city center, where overcrowded and low-quality population density and infrastructure development are housing in HCMC typically higher. However, in the peripheral areas, access to these amenities deteriorates. Accessibility levels of hospitals Lower-quality housing characteristics are more common and bus stations display relatively poorer overall access in some outer areas of HCMC. Around a quarter (24.5 compared to schools and parks. Addressing these disparities percent) of households in HCMC reside in housing with will be important to ensure that all HCMC residents have an area less than or equal to 20m2. Binh Tan District has the reasonable access to amenities and services, regardless of their most cramped housing, with 69 percent of households reside geographical location. These calculations do not take into in housing 20m2 or less. This district is also highly populous account the quality of these services, which may also vary, with one of the highest net increases in population from 2014- particularly for schools and hospitals. 19, which may relate to overcrowding and smaller average housing. The GSO and UNDP define living quarters under Across the different amenities measured, average 8m2 per capita to be a multi-dimensional poverty deprivation, distance to bus stations is highest from households. and about 10 percent of households in HCMC live in such Figure 68 shows the number of people who live within specific small quarters. More aspirational targets from the National distance ranges to certain amenities: hospitals, schools, bus Housing Development Strategy aim for 28m2 per capita in stations, and parks. The majority of people in HCMC live urban areas by 2025. Grid cell areas with lower average housing within 2.5km to a hospital, however more than 1.5 million areas per capita also have higher proportions of households people live outside such a radius. A larger number of people that rent their dwellings, rather than own (Figure 70). Figure 68. Population distribution by proximity Figure 69. Average distance by quintile of the Figure Figure68 68 Figure Figure69 69 to amenities wealth index 88 1010 77 99 88 66 Population (millions) Population (millions) 77 55 66 44 55 km km 33 44 33 22 22 11 11 00 00 Hospitals Hospitals Schools Schools Parks Parks Bus stops Bus stops Q1 (poorest Q1 20%) (poorest 20%) Q2 Q2 Q3 Q3 Q4 Q4 Q5 Q5(richest 20%) (richest 20%) 0-1 km 0-1 km 1-2.5 km 1-2.5 km 2.5-5 km 2.5-5 km 5-10 km 5-10 km 5-15 km 5-15 km 15+ km 15+ km Hospital Hospital School School Park Bus Park station Bus station Source: WB staff calculations, OpenStreetMap data. Note: Quintiles are based on the wealth index. Source: WB staff calculations, OpenStreetMap data. Towards more inclusive cities 67 PART 2. POVERTY AND INCLUSION IN URBAN AREAS AND CITIES Figure 70. Housing features Note: Averages are at the grid-level. Grids with less than population of 100 are not shown. In grid cells highlighted in orange, average dwelling area per capita is less than 28m2 per person based on the National Housing Development Strategy. Source: WB staff calculations. Figure 71. Housing quality Note: Averages are at the grid-level. Grids with less than population of 100 are not shown. Source: WB staff calculations. 68 Viet Nam bi-annual poverty & equity update - June 2024 3. Case Study: Mapping spatial inclusion in Ho Chi Minh City  Durable housing material is common, but there are area exposed to flood risk at various depths is generally higher pockets of lower-quality water and sanitation amenities than the share of the population that would be exposed. Under (Figure 71). Roofs and walls are considered durable if the 10-year flood scenario31, 8 percent of HCMC’s area, and materials are concrete, tiles or slabs, with virtually all dwellings less than 1 percent of the population are exposed (Figure 72). in HCMC constructed with such materials. However, when However, exposure is higher in eastern districts located close to it comes to sources of drinking and cooking water, households major rivers. Some 17 percent of Quan 2 is exposed to flood in the northern areas of HCMC such as Huyen Cu Chi risk, affecting nearly 10 percent of the population (Figure 79 have the most limited access to indoor tap water. About 30 in Annex C for comparison across districts). The exposed percent of households in Huyen Cu Chi access water through population of HCMC under 10-year flooding scenarios is drilled wells, while those in the northern border of Huyen about 19,000 households. This is lower than the number Can Gio District are most likely to have outdoor rather than of poor households identified by HCMC’s Department of indoor toilets. Labour, Invalids and Social Affairs. In 2021, the city had 37,772 poor households (1.49 percent of total households) Flooding is a risk across many and an additional 20,247 near-poor households (0.8 percent of areas of HCMC total households). While 10-year flood risk impacts less than 1 percent of the population, due to the potential size of migrant How much of HCMC’s population is exposed to flood and floating populations, these could be lower bound estimates. risk? As previous maps have shown, not all land is residentially inhabited, and there are different densities of the population By definition, 100-year floods have a 1 percent annual across the city. Thus, the proportion of the population exposed risk of occurrence. Under scenarios of rarer, but more to flood risk could be very different than the proportion of devastating 100-year floods and those with depths exceeding land exposed. Since not all land is inhabited, the share of land 20cm, 31.8 percent of HCMC’s area is exposed, and 19.7 Figure 72. Distribution of the population in areas with exposure to flood risk Note: Averages are at the grid-level. Grids with less than population of 100 are not shown. Source: WB staff calculations. 31 Ten percent annual risk of flood occurrence. Towards more inclusive cities 69 PART 2. POVERTY AND INCLUSION IN URBAN AREAS AND CITIES percent of the population is exposed. There are differences Under multiple cases of modeled flooding scenarios, 1-20 between which districts would flood at low-depths or high- percent of the population of HCMC are exposed to flooding. depths. These considerations are also important with respect In some cases, nearly 100 percent of the population of a to insurance, disaster response safety nets, or infrastructure district is exposed (Quan 2, 100-year flood extent at 50cm and improvements. For example, in vulnerable districts such as greater depths). The analysis also found that in some cases, Quan 2, almost 74 percent of the district would be exposed those exposed in a district are much poorer than the rest of to floods at depths of 50cm or greater, which would affect 91 the district’s population, but this is not the case in all districts. percent of Quan 2’s population. Still in other cases, when flood extents are large, wealthier groups in a district could be more exposed to flood risk. Thus, Climate change is expected to exacerbate existing risks, it is important to understand the nature of flood risks on a making larger floods more likely. Under a worst-case population under various scenarios and who is most likely climate change scenario, sea levels are set to rise by 30cm and impacted in different cases and likelihoods. 70cm by the years 2050 and 2100, respectively. This would result in the flooding of land currently inhabited by 20 million A composite view illustrates people in Viet Nam, equivalent to a quarter of the population a gradient of poverty and (Rentschler et al., 2020). A study of HCMC indicates that deprivations flooding risks in the city may be 10 times more severe by 2050 (McKinsey, 2020). The expected increase in flooding severity A composite view summarizes the joint deprivation is also related to the sinking of the city, thus certain areas are and risks from different indicators examined in earlier becoming lower and more susceptible to flooding. This is a sections. Previously, this report described a simple framework result of soil subsidence caused by excessive groundwater of examining spatial elements of inclusion and poverty in usage to support growing populations. Some parts of HCMC HCMC, and then discussed the spatial characteristics of have seen subsidence exceeding 70mm per year over the same each indicator separately. These indicators are selected partly period (Minh et al., 2015). Thus, even the potential impacts opportunistically based on data availability or because they from a 100-year flood as shown in Figure 79 (in Annex C) reflected conditions from a spatial perspective. For instance, should be considered seriously. four out of eight indicators are measures of distances to amenities. This section spotlights the distribution of these Are the poor more at risk of flooding? The answer indicators in a simplified composite index to help identify the depends on flood depth as well as the district and location. most vulnerable areas of HCMC. Table 8. Deprivation thresholds for selected indicators Group Indicator Deprivation threshold Livelihoods and income Wealth index Bottom 10 percent Access to public transport (bus stops) Farther than 2km Human capital Access to schools Farther than 2km Access to hospitals Farther than 2km Living standards Housing quality Has two or less out of four preferred features (*) Housing overcrowding Less than 28m2 in urban district and less than 26m2 in rural district Access to parks Farther than 2km Shocks Flood exposure, 1-in-100-year Highest flood depth (top 10% of districts with highest depths) Note: (*) this threshold is not based on the bottom 10 percent relative threshold. See Box 2.D. for more data information. Source: WB staff calculations. 70 Viet Nam bi-annual poverty & equity update - June 2024 3. Case Study: Mapping spatial inclusion in Ho Chi Minh City  To measure the degree to which households are living Results across the composite of selected indicators show in worse conditions or areas, relative deprivation better access to amenities and higher wealth in the urban thresholds are created. For some indicators larger values center. This is similar to global findings where urban centers reflect worse conditions, such as distance to amenities, while have much lower poverty than urban clusters, suburbs, or for other indicators smaller values point to worse conditions. towns. Figure 73 shows the average number of deprivations For the Wealth Index, a household is worse off when the score based on all eight indicators and when excluding the four is a lower value. Alternatively, for flood exposure depths, indicators on distance to amenities, at the grid-level. In both higher values are considered worse outcomes. For access cases, areas on the edges of HCMC show higher deprivations. to amenities, households are deprived if they are farther Two-thirds of the HCMC population have only two or fewer from amenities. There can be different views on what is an privations out of eight, mainly in central areas of HCMC32. appropriate distance that is considered “too far”. In this case study, a distance of 2km is used. The concept of livable cities By population, the outer districts of Huyen Cu Chi and with amenities available within a reasonable walking distance Huyen Binh Chanh have the largest populations with are part of Master Plans in some cities, many ranging from five or more deprivations (Figure 74). The figure below 15-20 minutes of walking (Moreno et al, 2021). For housing also notes that core districts are better off, while outer districts quality, deprivation is determined by the number of improved tend to be much larger in population and have more diversity or durable housing features. Households with two or less in outcomes. For large districts with diverse populations, preferred housing features are considered deprived, reflecting granular data is even more important to distinguish between 3 percent of households. For the indicator on overcrowding, pockets of poor and better off areas. deprivations are based on National Housing Development Strategy targets for 2025. Figure 73. Average number of deprivations, by grid cell Note: Grids with less than 100 people are excluded. Source: WB staff calculations. 32 There is a minimum bound on the proportion of households with at least one deprivation since the wealth and flood exposure indicators have thresholds based on a share of the population. Towards more inclusive cities 71 PART 2. POVERTY AND INCLUSION IN URBAN AREAS AND CITIES Figure 74. Number of deprivations, by district 800 700 Population (thousands) 600 500 400 300 200 100 0 Quan 1 Quan Phu Nhuan Quan 3 Quan 10 Quan 11 Quan 4 Quan 5 Quan 6 Quan Go Vap Quan Binh Thanh Quan Tan Binh Quan Tan Phu Quan 8 Quan Binh Tan Quan 7 Quan 2 Quan 12 Quan Thu Duc Quan 9 Huyen Cu Chi Huyen Hoc Mon Huyen Binh Chanh Huyen Nha Be Core districts Districts adjacent Outer urban Rural Figure74 to the core districts districts 0 1 2 3 4 5 6 7 8 Notes: There are eight total indicators as noted in Table 8. Source: WB staff calculations Box 2.E. Migrants in cities An individual or family’s place of origin matters in Figure 75. % population that are recent of the 75 Figure an urban setting. The complex residence certification migrants system called ho khau was created in the 1950s to restrict 16% large population movements by necessitating a household registration booklet to access basic public and social 14% welfare services, including housing, education, health care, % population that are recent migrants 12% and employment (World Bank, 2020b). In January 2023, household registration books were replaced by electronic 10% citizen identification cards issued for individuals. 8% The biggest cities in Viet Nam, Hanoi and HCMC, 6% have the largest inflows of migrants. A steady inflow of 4% migrant workers results in a large working age population. In 2022, the dependency ratio in the Southeast region 2% was 48.9, whereas in other regions of Viet Nam it ranged 0% from 63 to 74. Recent migrants in this study are defined Can Tho Danang HCMC Haiphong Hanoi as those who lived in a different province five years ago33. 2009 2014 2019 In the 2019 Population and Housing Census, about 3 Note: Recent migrants are those located in a different province five years ago. percent of the total population was classified as recent Source: World Bank staff calculation. migrants based on this definition, much lower than the share across the larger five municipalities (Figure 75). 33 This definition is constrained by the Census questionnaire and migration of a shorter term cannot be identified. 72 Viet Nam bi-annual poverty & equity update - June 2024 3. Case Study: Mapping spatial inclusion in Ho Chi Minh City  Box 2.E. Migrants in cities (contd) Yet, these results are likely to be an underestimate since floating or transient movements are not well captured in surveys. A 2015 World Bank migration survey showed that more than a third of HCMC residents did not have permanent residency in HCMC (World Bank and VASS, 2016). The report also estimated that 22 percent of HCMC’s population was long-term temporary, with 14 percent short-term temporary. 76. Migrant FigureFigure 76 population in HCMC HCMC population Figure 77. Share of Figure 77 that are (thousands) migrants (%) 600 25% Share of population that are migrants (%) 500 Migrant population (thousands) 20% 400 15% 300 10% 200 100 5% 0 0% Core Districts adjacent Outer Rural Core Districts adjacent Outer Rural districts to the core urban districts Districts districts to the core urban districts Districts 2009 2014 2019 2009 2014 2019 Source: WB staff calculations. Among recent migrants in HCMC, most tend to reside in the outer districts. Recent migrants to HCMC arrive from all across the country, but most are from adjacent Mekong Delta and Northern Coastal Central regions. Historically, migrants were most likely to live in outer urban districts (which the previous section has shown to have more deprivations on average), but over time this number has reduced, potentially because of a change in status to permanent settlement, but also lower arrival numbers of recent migrants (Figure 76 and Figure 77). Migrants are more vulnerable than non-migrants. Comparing poor outcomes in the eight social indicators, recent migrant households tend to have more deprivations than non-migrant households (Figure 78). Based on the 2019 Census, nationally, 19 percent of migrants lived in dwellings less than 8m2 per capita compared to 6.3 percent of non- migrants. Data also shows there are less children, and more single households in the northern districts of Thu Duc and District 9 adjacent to Binh Duong province, suggesting pockets of working migrants without families present. Ethnic minority migrants face additional challenges of social discrimination, fewer support systems, and sometimes insufficient Vietnamese language proficiency. Qualitative interviews were conducted in Hanoi and HCMC34 in 2022 to identify key challenges encountered by ethnic minority migrants (iSEE, 2023). Ethnic minorities arriving in major cities to live, study and work face additional barriers when integrating into the social and economic fabric of a city. Survey participants highlighted multiple challenges regarding inclusion into urban social and cultural life. 34 In HCMC, interviews excluded the Hoa ethnic group which is better established. The share of non-Hoa ethnic minorities in HCMC is small, less than 1 percent of the population in 2019. Towards more inclusive cities 73 PART 2. POVERTY AND INCLUSION IN URBAN AREAS AND CITIES Box 2.E. Migrants in cities (contd) Figure 78 Figure 78. Deprivations by migrant or non-migrant 35 30 25 Share of the population (%) 20 15 10 5 0 0 1 2 3 4 5 6 7 8 Non-Migrant Recent Migrant Source: WB staff calculations. These challenges can be divided into two categories: discrimination and a sense of not belonging. Jobs, income, and working conditions were top concerns for all ethnic minority migrant respondents. The majority of survey participants reported they faced numerous challenges finding jobs due to a lack of vocational training, education, qualifications, or limited Vietnamese language proficiency. Suitable housing and accessing essential services, medical care and education for children constituted the second most common concern of ethnic minority migrants interviewed. 4. Data-informed policies be location-specific to understand precisely where flood in urban areas risks are, distribution of residents, location of schools, or transportation needs. The highlighted recommended As Viet Nam continues to climb the development policies in Table 9 focus on aspects related to enabling and ladder, strategies to support urban livelihoods will supporting social development in urban areas. gain in importance. As more people move to cities and the country becomes denser, conditions in urban areas will have a larger role to play in poverty reduction and upward Better data for within-city economic mobility. Sustaining poverty reduction will analyses require effective management and provision of services in ever denser and crowded cities. A large urban population Data is an important input to inform government calls for more urban-based policies, particularly as the share and foster inclusive and equitable development. of the population living in urban contexts may be higher While numerous reports have noted urbanization than currently measured. There is a gradient of urbanization trends and aggregate movements of people from rural and poverty that requires more specialized and granular data to urban areas, quantifying the distribution of living to adequality measure and analyze household conditions. standards within a city is more insightful, but can This is essential as policies for dense urban areas need to be challenging to study since it requires more data. 74 Viet Nam bi-annual poverty & equity update - June 2024 4. Data-informed policies in urban areas  Table 9. Recommendations to mitigate urban poverty Recommendations Additional references Better data to Cities and urban areas have more inequality than rural areas, and are much more World Development Report 2021: monitor and denser. Thus, more data is needed to adequately study urban populations. The sample Data for Better Lives measure large and size of urban populations in official surveys can be increased. dense populations European Commission, 2020. The availability of satellite imagery offers an opportunity for more consistent measurements of urbanization, which may be useful to: 1) consistently compare Nakamura et al. 2023 urbanization trends across countries, as well as to: 2) map the gradient of urbanization within Viet Nam. Data collected on migration can be improved. Support inclusive Stronger support can help migrants adapt and integrate into urban settings more iSEE, 2023 policies for new successfully. residents Help narrow skills or financing gaps, support social networks, and promote more inclusive cultural norms. Ensure all residency and labor regulations are inclusive of permanent and temporary residents. Resilience under Many densely populated urban areas in Viet Nam are exposed to flood risk. Resilient Mukim and Roberts, 2023 climate change city policies require an increased focus on risks to help mitigation and adaptation, provision of insurance, and investments into resilient infrastructure. Urban planning The World Bank (2020b) report discusses policies to better leverage urbanization, and World Bank, 2020b and investment to minimize inefficiencies from lack of linkages, bottlenecks, and limited benefits from agglomeration. Mukim and Roberts, 2023 The importance of within-city improvements such as integration, adequate Lall et al., 2021. investments, upward expansion, and reducing urban sprawl are discussed in Mukim and Roberts (2023) and Lall et al. (2021). For example, understanding residential sorting is important samples large enough to compute district-level indicators. For for analyzing patterns of segregation and identifying potential example, in Indonesia, the SUSENAS has a sample of more disparities in access to resources and opportunities within a than 1,000 households in each of five districts within Jakarta city. When groups of people with similar characteristics (population of more than 10 million) and then additional tend to cluster and locate in the same parts of a city, this is similar-sized samples for districts which form the urban referred to as residential sorting. Residential sorting refers to periphery. Comparatively, HCMC had an official population the spatial concentration or segregation of certain population of about nine million in 2019. In annual Viet Nam Household characteristics within neighborhoods. It reflects the degree to Living Standards Surveys (VHLSS), about 1,400 households which individuals with similar attributes – such as education, are surveyed from HCMC, out of the 45,000 sample. In the income or ethnicity – tend to reside in the same geographical Philippines, a country with a population about 20 million areas. Although segregation is not inherently problematic, larger than Viet Nam’s, the Family Income and Expenditure residential sorting can result from various factors, including Survey (FIES) interviews 180,000 households. There are socio-economic disparities, historical patterns, and housing tradeoffs between size and frequency. While the Viet Nam market dynamics (Vaughan & Arbaci, 2011). VHLSS survey is conducted annually, the Philippines FIES is conducted every three years. Within-city analysis has been shown to be useful for city- level policy making, and requires more granular data. There is limited information on migrants in the Survey data is adequate for monitoring, but to inform specific census and official surveys. In the 2009, 2014, and 2019 decisions, more granular data is needed. Currently, official Population and Housing Censuses, there were questions surveys in Viet Nam allow for the calculation of municipality about the location of the individual five years ago. In the 2014 and province-level indicators. In some neighboring ASEAN Inter-census, there was an added question about a person’s countries, official household or labor force surveys have location one year ago. None of the censes has information Towards more inclusive cities 75 PART 2. POVERTY AND INCLUSION IN URBAN AREAS AND CITIES on location of birth. In other countries, such as Indonesia, (đăng ký tạm trú), and medical insurance payments. With the location of birth and across different time periods are available introduction of the National Database on Population (Cơ to distinguish between recent and long-term migrants. There sở dữ liệu quốc gia về dân cư) and the electronic ID card (thẻ is also limited information about migrants and migration căn cước công dân điện tử), migrants should be able to access in the household surveys. The last migration module in the services simply by presenting ID cards. This improvement VHLSS was in 2014. Regionally, some local statistical offices will reduce burdens, particularly on time and expenses from collect their own data, for example, HCMC conducts its own returning to towns of origin to obtain verifications. Achieving surveys of migrants in the municipality. this goal requires comprehensive implementation of new regulations regarding replacement of paper-based and written The rate of urbanization is an important consideration verification and identification, with online and electronic for policy direction in a Next Mile context. A large verification methods as stipulated in the Law on Residence urban population calls for more urban-focused policies, 2020 (Luật cư trú 2020) and Decree No. 104/2022/NĐ-CP particularly as the extent of population living in urban on the revision of specific articles in decrees that regulate contexts may be higher than currently measured. Availability presentation and submission of the household registration of satellite imagery provides an opportunity to consistently book and temporary registration book when applying for identify urban areas outside of administrative boundaries. public administration services36. Moreover, provision of Previous studies have commented that the “combination essential services by municipal governments should not be of administrative structure and the urban classification tied to permanent residency status. For example, children of system produces ill-defined spaces” (World Bank, 2020b). migrants should be eligible for public schooling, provided In a recent urban delineation comparison report, most they are registered as temporary residents. Having granular countries had much larger shares of population in urban information on the location and volume of new residents settings compared to official estimates of urban population would be especially instructive for managing expansion of shares (Nakamura et al., 2023). Methodologies to classify public services. the gradient of urbanization based on satellite imagery have also been endorsed by the United Nations (see European Some migrants do not have adequate skills or capacity in Commission, 2020). languages, qualifications or financing to take advantage of urban opportunities. Since migrants tend to be already Inclusive policies for vulnerable financially constrained and are looking for work, support groups in urban areas and cities packages such as transportation, social rented housing, government subsidized rent or low collateral microfinancing While the registration system has been reformed, could help ease their financial burdens. The lack of human monitoring is needed to ensure that migrants and capital—particularly vocational training, qualifications, and ethnic minorities in cities are not at a disadvantage. In skills required by the urban labor market—presents challenges January 2023, household registration books were replaced by for migrants looking to find decent jobs and their ability to electronic citizen identification cards issued for individuals. negotiate the terms of their employment. The state-financed On paper, the ho khau system has ended35, and monitoring is vocational training system for ethnic minorities, currently needed to ensure a smooth transition to equal access of public managed by MOLISA and the Committee for Ethnic services between temporary and permanent residents. Written Minority Affairs (CEMA), can incorporate more courses and stamped verification of personal identity information, specifically designed for the urban labor market, focusing particularly verification of place of residence (xác nhận nơi cư on skills required to handle jobs that currently attract a large trú) should not be treated as the prerequisite for provision of proportion of migrants. Interviews of migrants in large cities essential public services (dịch vụ công thiết yếu) for migrants, also suggest that “modern” skills such as English language such as allowances (trợ cấp), temporary residency registration competency or online trading have become increasingly 35 Law on Residence, Clause 3, Article 38. 36 In Vietnamese: Nghị định sửa đổi, bổ sung một số điều của các nghị định liên quan đến việc nộp, xuất trình sổ hộ khẩu, sổ tạm trú giấy khi thực hiện thủ tục hành chính, cung cấp dịch vụ công. 76 Viet Nam bi-annual poverty & equity update - June 2024 4. Data-informed policies in urban areas  important in urban contexts. A good example to consider and be important as a household’s risk profile varies by the type replicate is the HCMC local government policy to subsidize of shock: susceptibility to natural disasters may depend on tuition fees of ethnic minority students37. a household’s location and housing quality. Idiosyncratic shocks are generally best dealt with by effective social Resilience amid climate change insurance coverage, while covariate shocks often require a more coordinated government response. Viet Nam is rich in policy provisions related to disaster risk management, climate change adaptation, and social Urban planning protection, but an integrated policy to address these issues in a more holistic way can be improved. Up-to- The focus of this report is household conditions in urban date disaster risk and risk management strategies require not areas, but a core aspect of creating and managing livable only environmental data, but a good understanding of the and equitable cities is urban planning and investments. location, densities, and profiles of the population. Better data These considerations include equitable distribution and on the distribution of population can inform placement of access to public services and amenities, provision of public early warning information and monitoring systems. These transportation options, regulating housing quality standards, systems will also need to be flexible, especially in the face of upgrading low-quality infrastructure, providing low-cost climate change and potential future shocks. Beyond being well housing options, and other planning elements to improve managed and coordinated, they must be capable of adapting the quality of city life in an equitable and inclusive manner. to new forms of shocks that will require flexible policies and All these planning decisions require granular data to make delivery systems to scale up and down quickly in the wake of accurate assessments and impactful investments. There are covariate shocks. Leveraging regular program systems would several excellent existing reports that delve into the urban allow provinces and districts to use existing policy, budgetary planning recommendations for Viet Nam specifically, as well and delivery platforms to rapidly scale-up support, instead as discussing international best principles. These reports are of ad hoc and temporary policies. The example of mapping summarized below. Readers are encouraged to turn to these HCMC flood risks and measurements of the population for expanded recommendations. residing in flood plains can serve as inputs into estimating resources needed to support populations when disaster events • The World Bank (2020b) report discusses policies for Viet do occur, but more tailored modelling is necessary using Nam to better leverage urbanization, and to minimize country-level parameters for accuracy in predictions. inefficiencies from lack of linkages, bottlenecks, and limited benefits from agglomeration. The report recommends Intensifying climate change risks call for stronger three main areas of reform to leverage urban areas to sustain in-place safety nets and insurance. In 2018, about 10 growth: easing constraints on labor mobility, strengthening percent of communes reported experiencing an emergency, planning and land use regulations, and improving fiscal including natural disasters, fires, and epidemics. About 60 allocations to match the needs of fast-growing urban areas. percent of these communes received relief or aid, primarily • The importance of within-city improvements such as in the form of direct cash relief from the government budget. integration, adequate investments, upward expansion, and Depending on the extent of natural disasters, environmental reducing urban sprawl are discussed in Mukim and Roberts shocks could be either: (i) idiosyncratic, which only affect (2023) and Lall et al. (2021). These recommendations are specific individuals or households and (ii) covariate, which provided at a global level. affect entire communities or regions. The distinction can 37 Decision No.4453/QĐ-UBND (December 31, 2021) on the renewal of the exemption of tuition fees for Cham and Khmer students in Ho Chi Minh City in school years 2020-21 and 2021-22. Towards more inclusive cities 77 PART 2. POVERTY AND INCLUSION IN URBAN AREAS AND CITIES Annex C. Figures – Part 2 Figure 79 Figure 79. Population at risk of 10-year and 100-year floods, by district 100-year flood (1 percent annual risk of occurrence) No flood risk or less than 20cm 20 - 50 cm Greater than 50 cm 800,000 700,000 600,000 500,000 Population 400,000 300,000 200,000 100,000 0 Binh Chanh Binh Tan Binh Thanh Can Gio Cu Chi Go Vap Hoc Mon Nha Be Phu Nhuan Quan 1 Quan 10 Quan 11 Quan 12 Quan 2 Quan 3 Quan 4 Quan 5 Quan 6 Quan 7 Quan 8 Quan 9 Tan Binh Tan Phu Thu Duc 10-year flood (10 percent annual risk of occurrence) No flood risk or less than 20cm 20 - 50 cm Greater than 50 cm 800,000 700,000 600,000 500,000 Population 400,000 300,000 200,000 100,000 0 Binh Chanh Binh Tan Binh Thanh Can Gio Cu Chi Go Vap Hoc Mon Nha Be Phu Nhuan Quan 1 Quan 10 Quan 11 Quan 12 Quan 2 Quan 3 Quan 4 Quan 5 Quan 6 Quan 7 Quan 8 Quan 9 Tan Binh Tan Phu Thu Duc Note: Fluvial + pluvial flooding. See Figure 72 for geospatial visualization. Source: World Bank staff calculations. 78 Viet Nam bi-annual poverty & equity update - June 2024 References  References GSO. 2020. Results of the Vietnam Population and Housing Census of April 1, 2019. Central Population and Housing Census Steering Committee. March 2020. Alcantara, A., Brewer, S. M., & Jozefowicz, J. J. 2023. Rural-urban GSO. 2023. Press Release. On the labor and employment situation in differences in poverty: An analysis of Pennsylvania counties. the second quarter of 2023. International Journal of Economics and Financial Issues, 13(5), 1. Jacobs, J. 1985. Cities and the wealth of nations: Principles of economic Andersson, F., Burgess, S., & Lane, J. I. 2007. Cities, matching and the life. Vintage. productivity gains of agglomeration. Journal of Urban Economics, Johnson, Doug and Robert Palacios (forthcoming). Scaling up Cash 61(1), 112-128. Transfers during the COVID-19 Pandemic: What might explain Bellefon et al 2021 differences in coverage? Benfica, R., & Henderson, H. 2021. The effect of the sectoral Joliffe, Dean, and Espen Beer Prydz. 2016. Estimating International composition of economic growth on rural and urban poverty. Poverty Lines from Comparable National Thresholds. Policy Review of Income and Wealth, 67(1), 248-284. Research Working Paper No. 7606. Brueckner, M., Dabla-Norris, E., Gradstein, M., & Lederman, D. 2018. Kraay, Aart C.; Lakner, Christoph; Ozler, Berk; Decerf, Benoit Marie A; The rise of the middle class and economic growth in ASEAN. Jolliffe, Dean Mitchell; Sterck, Olivier Christian Brigitte; Yonzan, Journal of Asian Economics, 56, 48-58. Nishant. 2023. A New Distribution Sensitive Index for Measuring Chau, Lam Minh, 2023. “Ethnic Minority Migrants in Major Cities of Welfare, Poverty, and Inequality. Policy Research working paper; no. Vietnam: Challenges and Exclusion Dynamics, 2022-23”. WPS 10470. Washington, D.C.: World Bank Group. Combes, P. P., Duranton, G., Gobillon, L., Puga, D., & Roux, S. Lakner, Christoph, Daniel Gerszon Mahler, Mario Negre, and Espen 2012. The productivity advantages of large cities: Distinguishing Beer Prydz. 2020. “How Much Does Reducing Inequality Matter agglomeration from firm selection. Econometrica, 80(6), 2543-2594. for Global Poverty?” Global Poverty Monitoring Technical Note 13, Combes, Pierre-Philippe, Shohei Nakamura, Mark Roberts and World Bank, Washington, DC. Benjamin Stewart. 2022. Estimating urban poverty consistently Lall, Somik; Lebrand, Mathilde; Park, Hogeun; Sturm, Daniel; across countries. World Bank Poverty & Equity notes. Venables, Anthony; Lebrand, Mathilda. 2021. Pancakes to Cunningham, Wendy, and Obert Pimhidzai. 2018. Vietnam’s Future Pyramids: City Form to Promote Sustainable Growth. World Bank, Jobs: Leveraging mega-trends for greater prosperity. Washington, Washington, DC. D.C.: World Bank Group. iSEE (Institute for Studies of Society, Economy and Environment). 2023. Deaton, Angus, and Salman Zaidi. 2002. “Guidelines for Constructing “Ethnic Minority Migrants in Major Cities of Vietnam: Challenges Consumption Aggregates for Welfare Analysis.” LSMS Working and Exclusion Dynamics 2022-2023.” World Bank working paper. Paper; No. 135. Letdin, Mariya, and Hyoung S. Shim. 2019. Location choice, life cycle Dijkstra, L., and H. Poelman. 2014. “A Harmonised Definition of Cities and amenities. Journal of Regional Science. 59(3), 567-585. and Rural Areas: The New Degree of Urbanisation.” Regional Lucci, P., Bhatkal, T., & Khan, A. 2018. Are we underestimating urban Working Paper 01/2014. Directorate General for Regional and poverty? World Development, 103, 297–310. Urban Policy, European Commission. Mancini, Giulia, and Giovanni Vecchi. 2022. “On the Construction of a Dijkstra, Lewis, and Aneta J. Florczyk and Sergio Freire and Thomas Kemper Consumption Aggregate for Inequality and Poverty Analysis”. and Michele Melchiorri and Martino Pesaresi and Marcello Schiavina. Marx, B., Stoker, T., & Suri, T. 2013. The economics of slums in 2021. “Applying the Degree of Urbanisation to the globe: A new the developing world. Journal of Economic Perspectives, 27 harmonised definition reveals a different picture of global urbanisation”. (4), 187–210. Journal of Urban Economics. Volume 125, September 2021. Mason, Andrew D., and Shetty, Sudhir. 2018. A Resurgent East Asia: Duranton, G. and Puga, D. 2014. The growth of cities. Handbook of Navigating a Changing World. World Bank East Asia and Pacific economic growth, 2, 781-853. Regional Report Washington, D.C.: World Bank Group. Duranton, Gilles, and Stuart S. Rosenthal. 2021. Delineation of Urban Mathur, O. P. 2013. Urban poverty in Asia. Asian Development Bank, Areas, Journal of Urban Economist special issue. Volume 125, Metro Manila, Philippines, 1-122. September 2021. McLeod, P.B., 1984. “The demand for local amenity: a hedonic price European Commission. 2020. A recommendation on the method to analysis.” Environment Planning A. 16, 389–400. delineate cities, urban and rural areas for international statistical McKinsey Global Institute. 2020. Can coastal cities turn the tide on comparisons. rising flood risk? Case Study. April 2020. Ferreira, Francisco H. G., Julian Messina, Jamele Rigolini, Luis-Felipe Minh, D. H. T., Van Trung, L., & Toan, T. L. (2015). Mapping ground López-Calva, Maria Ana Lugo, and Renos Vakis. 2013. Economic subsidence phenomena in Ho Chi Minh City through the radar Mobility and the Rise of the Latin American Middle Class. interferometry technique using ALOS PALSAR data. Remote Washington, DC: World Bank. Sensing, 7(7), 8543-8562. Glaeser, E. L., Kallal, H. D., Scheinkman, J. A., & Shleifer, A. (1992). Growth in cities. Journal of political economy, 100(6), 1126-1152. Towards more inclusive cities 79 PART 2. POVERTY AND INCLUSION IN URBAN AREAS AND CITIES Mookherjee, Dilip, and Anthony Shorrocks. 1982. A Decomposition World Bank. 2009. World Development Report – Reshaping Economic Analysis of the Trend in UK Income Inequality. The Economic Geography. Journal, 92 (368), 886–902. World Bank. 2011. Vietnam urbanization review: technical assistance Moreno, C., Z. Allam, D. Chabaud, C. Gall, F. Pratlong. 2021. report (English). Washington, D.C.: World Bank Group. “Introducing the “15-minute city”: Sustainability, resilience and World Bank. 2017. Riding the wave: an East Asian miracle for the place identity in future post-pandemic cities”. Smart Cities, 4 21st century. World Bank East Asia and Pacific Regional Report. (1), 93-111. Washington, D.C.: World Bank Group. Mukim, Megha, and Mark Roberts, editors. 2023. Thriving: Making World Bank. 2018a. Poverty and Shared Prosperity 2018: Piecing Cities Green, Resilient, and Inclusive in a Changing Climate. Together the Poverty Puzzle. Washington, D.C.: World Bank Group. Washington, DC: World Bank. World Bank. 2018b. “Better Living in the City: Trends, patterns and Nakamura, Shohei and Combes, Pierre-Philippe and Moellerherm, outcomes of migration in Indonesia” Robin and Robert, Charlotte and Roberts, Mark and Stewart, World Bank. 2020a. Poverty and Shared Prosperity 2020 – Reversals of Benjamin and Yakubenko, Slava. 2023. Where is Poverty Fortune. Washington, D.C.: World Bank Group. Concentrated? New Evidence Based on Internationally Consistent World Bank. 2020b. Vietnam’s Urbanization at a Crossroads: Embarking Urban and Poverty Measurements. on an Efficient, Inclusive, and Resilient Pathway. Washington, D.C.: OECD. 2023. New PISA results: strengthening education systems in World Bank Group. the wake of the pandemic. World Bank. 2021a. A Year Deferred - Early Experiences and Lessons Ravallion, Martin, Shaohua Chen, and Prem Sangraula. 2007. New from Covid-19 in Viet Nam. Washington, D.C.: World Bank Group. Evidence on the Urbanization of Global Poverty. Population and World Bank. 2021b. World Development Report 2021: Data for Better Development Review, 33(4), 667-701. Lives. Washington, D.C.: World Bank Group. Ravallion, Martin. 2015. The Economics of Poverty: History, World Bank. 2022a. Poverty and Shared Prosperity 2020 – Correcting Measurement, and Policy. Oxford University Press. Course. Washington, D.C.: World Bank Group. Rentschler, J., Salhab, M., & Jafino, B. A. 2022. Flood exposure World Bank. 2022b. 2022 Viet Nam Poverty Assessment - From the Last and poverty in 188 countries. Nature Communications, 13(1), Mile to the Next Mile. Washington, D.C.: World Bank Group. Article 3527. World Bank. 2023. Crisis and Recovery - Learning from COVID-19’s Sridhar, K. S. 2015. Is urban poverty more challenging than rural poverty? Economic Impacts and Policy Responses in East Asia. Washington, A review. Environment and Urbanization ASIA, 6(2), 95-108. D.C.: World Bank Group. United Nations, Department of Economic and Social Affairs, World Bank. 2024a. Global Economics Prospects. January 2024. Population Division (2018). World Urbanization Prospects: The Washington, D.C.: World Bank Group. 2018 Revision, Online Edition. World Bank. 2024b. Construction of the Viet Nam 2022 Consumption UN Habitat. 2016. 2016 City Prosperity Index database. Aggregate note. Internal only. Vaughan, L., Arbaci, S., 2011. The Challenges of Understanding Urban World Bank and VASS. 2016. Vietnam’s household registration system. Segregation. Built Environment 37 (2), 128-138. Washington, D.C.: World Bank Group. World Bank. 2006. World Development Report 2006: equity and development. World development report. Washington, D.C.: World Bank Group. 80 Viet Nam bi-annual poverty & equity update - June 2024 With support from: 8 Dao Tan Street, Ba Dinh District, Hanoi, Viet Nam Telephone: (84-24) 3774 0100 Fax: (84-24) 3774 0111 Website: www.dfat.gov.au 63 Ly Thai To Street, Hoan Kiem, Hanoi, Viet Nam Telephone: (84-24) 3934 6600 Fax: (84-24) 3935 0752 Website: www.worldbank.org.vn @WorldBankVietnam @WB_AsiaPacific