Global Poverty Monitoring Technical Note 31 On the construction of the World Bank’s Subnational Poverty and Inequality Databases Documentation Minh Cong Nguyen, Judy Yang, Hai-Anh Dang, Carlos Sabatino August 2023 Keywords: Subnational poverty and inequality; August 2023; Documentation. Development Data Group Development Research Group GLOBAL POVERTY MONITORING TECHNICAL NOTE 31 Poverty and Equity Global Practice Group Abstract In many countries, large differences in poverty persist at the subnational level. In addition, global challenges such as climate change, fragility, economic crises, and food insecurity are often trans- border issues that pose significant risks for poverty reduction both across and within countries. Traditional poverty measures are generally presented at the national level, potentially obscuring local and regional variations of poverty and inequality. To overcome these challenges, this note describes the construction of two databases designed to provide a more granular perspective on poverty. The Subnational Poverty and Inequality Database (SPID) presents direct survey estimates of poverty and inequality from nationally representative household surveys over time. The Global Subnational Atlas of Poverty (GSAP) presents poverty estimates of survey-representative administrative areas projected to a common year. Both databases use the same underlying household survey data used by the World Bank to monitor global poverty. All authors are with the World Bank. Corresponding authors: Minh Cong Nguyen (mnguyen3@worldbank.org) and Judy Yang (jyang4@worldbank.org). The work could not have been completed without contributions from the Data for Goals (D4G), regional and country teams, and the Global Poverty & Inequality Data (GPID) team. Regional teams: Jose Montes, Ifeanyi Nzegwu Edochie, Elizabeth Mary Foster, Hugo Rolando Nopo Aguilar, Diana M. Sanchez Castro, Sergio Olivieri, Minh C. Nguyen, Ikuko Uochi, Veronica S. Montalva Talledo, Reno Dewina, Laura Liliana Moreno Herrera, Zurab Sajaia; D4G team: Carolina Diaz-Bonilla, Gabriel Lara Ibarra, Haoyu Wu; GPID team: Christoph Lakner, R. Andres Castaneda Aguilar, Daniel Gerszon Mahler. We would also like to thank Joao Pedro Azevedo, Paul A. Corral Rodas, Hongxi Zhao, Qiong Lu for comments on an earlier note, and countless Poverty Economists in the Poverty and Equity Practice who have provided data and documentation, and patiently answered our questions. Without them, the database of household surveys that underpins the World Bank’s global poverty measures would not exist. This note has been cleared by Benu Bidani. The Global Poverty Monitoring Technical Note Series publishes short papers that document methodological aspects of the World Bank’s global poverty estimates. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. Global Poverty Monitoring Technical Notes are available at pip.worldbank.org/. Contents 1. Introduction ............................................................................................................................. 2 1.1. Subnational Poverty and Inequality Database ................................................................ 3 1.2. Global Subnational Atlas of Poverty .............................................................................. 3 2. Methodology for constructing subnational poverty databases and shapefiles ........................ 4 3. Creation of subnational geospatial shapefile .......................................................................... 6 3.1. Selection of administrative units ..................................................................................... 6 3.2. Selection of administrative boundaries ........................................................................... 6 3.3. Labeling the geospatial feature class .............................................................................. 8 4. Survey data for poverty estimation ....................................................................................... 10 4.1. Household survey data .................................................................................................. 10 4.2. Calculating SPID subnational indicators ...................................................................... 11 4.3. Calculating GSAP line-up poverty estimates ............................................................... 14 5. Notes about the SPID and GSAP .......................................................................................... 18 6. References ............................................................................................................................. 19 7. Annex A ................................................................................................................................ 20 8. Annex B ................................................................................................................................ 26 1 1. Introduction Spatial poverty data informs evidence-based decisions on everything from infrastructure investments to crisis response during conflict to vulnerability assessments to natural disasters and climate change. A global picture of poverty at subnational levels of disaggregation helps policymakers identify lagging areas and pockets of chronic poverty that may require dedicated interventions. However, the production of this subnational poverty data poses several challenges related to availability, quality, comparability, and representativeness of the spatial disaggregation. In some cases, country boundaries and survey representativeness can change over time, complicating the production of panel data. In others, subnational data may not be readily available or might be of lower quality compared to national-level data, resulting in gaps or inaccuracies. Moreover, ensuring the comparability of subnational data across different regions and over time is a challenge due to variations in the survey instrument, data collection tools or the construction of welfare aggregates. This note describes the construction of two databases designed to address these challenges and provide granular subnational poverty, inequality, and multi-dimensional poverty indicators over time and at a global level.1 The first is the Subnational Poverty and Inequality Database (SPID), a novel panel database of subnational indicators calculated directly from household surveys over time. The second is the Global Subnational Atlas of Poverty (GSAP), a database that shows poverty estimates for all subnational areas in the world for a common reference year. Both databases are constructed using geospatial boundary features and survey data from the Global Monitoring Database (GMD), the World Bank’s repository of income and expenditure household surveys used to monitor poverty.2 The databases can be accessed through the World Bank’s 1 GSAP database can be accessed here: https://datacatalog.worldbank.org/search/dataset/0042041/global_subnational_poverty_atlas_gsap SPID database can be accessed here: https://datacatalog.worldbank.org/search/dataset/0064796/subnational_poverty_and_inequality_database_spid 2 The Global Monitoring Database (GMD) is the World Bank’s repository of multitopic income and expenditure household surveys used to monitor global poverty and shared prosperity. The household survey data are typically collected by national statistical offices in each country, and then compiled, processed, and harmonized. The process is coordinated by the Data for Goals (D4G) team and supported by the six regional statistics teams in the Poverty and Equity Global Practice. The Global Poverty & Inequality Data Team (GPID) in the Development Economics Data Group (DECDG) also contributes historical data from before 1990 and recent survey data from the Luxemburg Income Study (LIS). Selected variables have been harmonized to the extent possible such that levels and trends in poverty and other key sociodemographic attributes can be reasonably compared across and within countries over time. The GMD’s 2 Poverty and Inequality Platform (https://pip.worldbank.org/home) or the World Bank’s Geospatial Poverty Portal (https://pipmaps.worldbank.org/). 1.1. Subnational Poverty and Inequality Database The Subnational Poverty and Inequality Database is a novel database of subnational indicators built upon countries’ subnational boundaries and official household income and consumption surveys. This first edition of the database presents subnational poverty and inequality indicators for around 6,500 subnational data points in 141 economies, representing around 75% of the world’s population. In most cases, a representative subnational unit refers to a province or state (i.e., first-level administrative boundaries – ADM1), but it can also be a group of regions determined by the specific sampling strategy and representativeness of the household survey. For the 141 countries included, the SPID offers a panel component that is representative of 1,650 unique subnational areas. On average, a country has data for 14 geographical areas over three years. More than 90 percent of the survey data ranges from 2003 to 2022.3 77 countries or economies are not included due to a lack of data, no access to microdata, with most cases being territories, small island nations, or FCV countries where data is lacking. 4 Developing economies with large populations missing from this database version include Afghanistan, Eritrea, and Somalia. See Table A.1 in Annex A for a complete listing of surveys and years. Unlike the Global Subnational Atlas of Poverty (see below), the SPID presents direct survey estimates of poverty and inequality over time, while the GSAP displays poverty estimates for all subnational areas extrapolated or interpolated to a common year. 1.2. Global Subnational Atlas of Poverty The Global Subnational Atlas of Poverty is a database and data product that shows poverty estimates for all subnational areas lined-up to a common year, in this case 2019. Projecting to a common year allows for comparisons of subnational estimates across regions and the globe, but it harmonized microdata are currently used in the Poverty and Inequality Platform (PIP), the World Bank’s Multidimensional Poverty Measure (WB MPM), the Global Database of Shared Prosperity (GDSP), and Poverty and Shared Prosperity Reports. Additional information on the latest country data can be found in Castaneda et al., 2023. 3 Total data included in the SPID is available from the 2003 to 2020 period. Users can access the World Bank Geospatial Poverty Portal for subnational line-up poverty rates that are calculated in 2019 for all countries when available. 4 The World Bank’s country classification includes a total of 218 economies. 3 also requires additional assumptions (see below), since not every country has a household survey in every year. The Spring 2023 edition of the GSAP features poverty estimates lined-up up to 2019 for 1,735 subnational areas across 168 economies. The subnational line-up poverty estimates follow the same methodology as the country-level estimates that underpin the global poverty numbers published in Poverty and Inequality Platform (PIP) in March 2023.5 2. Methodology for constructing subnational poverty databases and shapefiles The SPID and the GSAP combine various data sources, including household survey data, administrative boundaries, and other related information, in a single database to illustrate global poverty at subnational levels (Figure 1). Subnational estimates of poverty and inequality are calculated from nationally representative household surveys in the GMD. In addition, custom shapefiles are constructed to match the subnational geographic areas that are available in the GMD. The custom shapefiles use a comparable spatial hierarchy across countries and cover four administrative levels, from country to district and finer units. Figure 1. Integration of geographic boundaries, and poverty and inequality data into a single database Source: Authors 5 For more details on the March 2023 update, please see Castaneda et al. (2023). 4 The processes adopted for forming the SPID and GSAP databases is schematically represented in Figure 2. The geospatial database comprises two main components: non-spatial datasets (data in the relational tables) and spatial datasets (feature classes). The Global Monitoring Database (GMD) is the primary source of the non-spatial data used here. The spatial datasets (feature classes) in the geodatabase were established based on the geo-reference of the non-spatial survey data. Rigorous processing and quality checks were performed before compiling these datasets in a geodatabase. The methodologies are presented in further detail below. Figure 2. Methodology for construction of the SPID and GSAP Source: Authors Note: For the GSAP, there is an additional processing step, which is the addition of lined-up poverty estimates from GMD household survey data. 5 With these geospatial poverty databases, geographically referenced subnational poverty data can be used to produce maps, interactive queries, and various spatial analyses. A global picture of poverty at administrative levels of disaggregation helps identify areas of poverty, potentially informing the design of interventions. These databases and the accompanying Geospatial Poverty Portal are an effective tool to help policymakers and the public identify poverty hotspots and allocate resources to the target population, enabling the most efficient use of scarce resources. 3. Creation of subnational geospatial shapefile The geospatial boundary feature class forms the main trunk of the global subnational database. The geographical area choice for each survey data is based on the survey representativeness based on the sampling design and survey documentation. 3.1. Selection of administrative units Many countries have distinct administrative hierarchy structures; however, these administrative boundary levels might not be available from household survey data. Therefore, it is challenging to perfectly match administrative levels across countries. For this database, we use a relatively comparable spatial hierarchy across countries. Four administrative levels exist: national, region, province, district, or finer level. Geographic boundaries must match the subnational regions in these surveys. In many cases, there is a one-to-one association between the geographical regions in a household survey and the areas defined at an administrative level in the country. In cases where there is no one-to-one association, geographic boundaries are altered to fit the representativeness of the surveys. In some cases, the geographic representation is at the level of regions and “urban” and/or “rural”. In these cases, subnational areas in the household survey are aggregated to levels that boundaries can appropriately represent. For most of the database, surveys are representative at the first administrative level (ADM1) or statistical regions (areas). 3.2. Selection of administrative boundaries After determining the administrative levels, the next step is to build up a feature class to map these administrative units. The primary sources of the features are the publicly available worldwide boundaries, including the Food and Agriculture Organization of the United Nations’ (FAO) Global 6 Administrative Unit Layers (GAUL), the Global Administrative Boundaries (GADM), and the Nomenclature of Territorial Units for Statistics (NUTS). Among these, the spatial file with the most disaggregation and geographic alignment with the household survey is chosen. For example, NUTS spatial files are used prominently for the European countries in the subnational poverty database, since these files are developed and regulated by the European Union (EU). • GADM: The Database of Global Administrative Areas (GADM), is a high-resolution database of country administrative areas, with a goal of “all countries, at all levels, at any time period”.6 The database is available in a few formats, including shapefiles used in most common GIS applications. The GADM project created data for many countries from spatial databases provided by national governments, NGOs, and/or from maps and lists of names available on the Internet. • GAUL: The Global Administrative Unit Layers (GAUL) compiles and disseminates the best available information on administrative units for all the countries in the world, contributing to the standardization of the spatial dataset representing administrative units.7 The GAUL always maintains global layers with a unified coding system at the country, first (e.g., departments) and second administrative levels (e.g., districts). Where data is available, it provides layers on a country-by-country basis down to third, fourth, and lower levels. The overall methodology consists in a) collecting the best available data from the most reliable sources, b) establishing validation periods of the geographic features (when possible), c) adding selected data to the global layer based on the last country boundaries map provided by the UN Cartographic Unit (UNCS), d) generating codes using the GAUL Coding System, and e) distribute data to the users.8 • NUTS: Nomenclature of Territorial Units for Statistics or NUTS (in French: Nomenclature des unités territoriales statistiques) is a geocode standard for referencing the subdivisions of countries for statistical purposes. The standard, adopted in 2003, is developed and regulated 6 Database of Global Administrative Areas - https://gadm.org/index.html 7 The Global Administrative Unit Layers (GAUL) is an initiative implemented by FAO within the Bill & Melinda Gates Foundation, Agricultural Market Information System (AMIS), and AfricaFertilizer.org projects. http://www.fao.org/geonetwork/srv/en/metadata.show?id=12691 8 See more details in Technical Aspects of the GAUL Distribution Set. 7 by the European Union, and thus only covers the EU member states in detail. NUTS is instrumental in the European Union’s Structural Funds and Cohesion Fund delivery mechanisms and in locating the area where goods and services subject to European public procurement legislation are to be delivered. For each EU member country, a hierarchy of three NUTS levels is established by Eurostat in agreement with each member state; the subdivisions in some levels do not necessarily correspond to administrative divisions within the country. Below the three NUTS levels are local administrative units (LAUs). A similar statistical system is defined for the candidate countries and European Free Trade Association members. The current NUTS classification, dated 21 November 2016 and effective from 1 January 2018 (now updated to existing members as of 2020), lists 92 regions at NUTS 1, 244 regions at NUTS 2, 1215 regions at NUTS 3 level, and 99,387 local administrative units (LAUs).9 • Customized boundaries: In some cases, geospatial files had to be customized based on the survey information. The customization is country-specific involving either (i) dissolving different boundaries from GAUL and GADM or (ii) altering area specifics. Those areas are often classified as ADMx or GADMx in table A.1 below. 3.3. Labeling the geospatial feature class When building up the feature class of the geodatabase, it is important to ensure consistency between spatial boundaries. It is imperative to use the exact matched administrative boundaries so that there are no gaps between boundaries thus spatial boundaries and corresponding data can be analyzed together. The GAUL 2015 database has been chosen as the destination database and map. Figure 3 illustrates the key processes for the compilation of the feature class. The survey data were summarized by seven geographically contiguous regions: Sub-Saharan Africa (SSA), East Asia Pacific (EAP), Europe and Central Asia (ECA), Latin American and Caribbean (LAC), Middle East and North Africa (MENA), South Asia (SAR), and North America. The feature class was processed region by region. The first step is to identify how many survey data can be mapped using the different geospatial boundary datasets by matching representative area names of survey data and attribute tables from those geospatial boundary sources, with the following priority order: GAUL, NUTS, GADM, and other country-specific boundaries. For many European Union 9 This vintage of the NUTS classification covers the UK, which has since exited the EU. 8 member countries, the administrative units of the survey data do not match that in the GAUL dataset, as they are generally based on larger scales (‘Admin 0.5 units’). In that case, the survey data for the EU countries were matched using the administrative units of the NUTS classification. Figure 3. The compilation of the geospatial feature class Source: Authors In the end, using the process described in Figure 3, the extracted and created features were appended. The attributes of the features were organized in a consistent format, and the feature class 9 of the geodatabase was created. The match between the administrative units of the survey data and the feature class was further validated. The consistency of the boundaries was also checked to ensure that all the features were interconnected.10 Figure 4 shows different types of boundaries used to construct the global subnational geospatial class, with most survey representative areas using GAUL administrative definitions. Figure 4. The administrative units of the global subnational database Source: Authors Note: Categories with “x” indicate areas have been altered to match the areas of the household survey based on the original source of administrative boundaries (GAUL, GADM or NUTS). DKx indicates country-specific or custom shapefiles. 4. Survey data for poverty estimation 4.1. Household survey data 10 Country borders and disputed areas are further denoted based on World Bank’s guidelines. 10 To create the subnational data, we use the most comprehensive data on poverty taken from the World Bank’s Global Monitoring Database (GMD) produced by the Poverty and Equity Global Practice and the Global Poverty & Inequality Data Team (GPID) in the Development Economics Data Group (DECDG). The Global Monitoring Database (GMD) is the World Bank’s repository of multitopic income and expenditure household surveys used to monitor global poverty and shared prosperity. The nationally representative household survey data are typically collected by national statistical offices in each country, and then compiled, processed, and harmonized. From these surveys, household welfare aggregates are constructed and harmonized by Regional Statistics teams coordinated by the Data for Goals (D4G) team in the Poverty and Equity GP to ensure a minimum level of cross-country comparability. For several high-income economies, which are not covered by a World Bank staff, the GPID Team obtains harmonized data from the Luxembourg Income Study (LIS). The harmonized household survey data is collated in the GMD, where variables are standardized, and data is stored with vintages. The harmonized surveys in the GMD are combined with data on inflation, purchasing power parity conversion factors, national accounts, and population to calculate the final poverty and inequality estimates. These national or country-level estimates are reported in the Poverty and Inequality Platform (PIP).11 The subnational estimates presented here use the same household survey data. 4.2. Calculating SPID subnational indicators Subnational poverty rates are calculated at different international poverty lines for the purpose of global poverty monitoring. Poverty rates are provided at the subnational level that is representative for the associated consumption or income survey used. Overall, the database represents 141 economies in all geographical regions, accounting for 75% of the world population (Figures 6 and 7). To complement monetary poverty rates, we also report the multidimensional poverty measure together with its six component indicators at the subnational level. Since the data requirement for this multidimensional indicator is more demanding, multidimensional poverty statistics are shown 11 Poverty and Inequality Platform, https://pip.worldbank.org/home 11 for about 3,200 subnational areas from 2010 to 2022 (compared with 6,500 areas for monetary poverty).12 We also use the most widely accepted inequality measures, such as the Gini index, Theil (or GE(1)) and mean log deviation (or GE(0)). These indices are computed at the subnational level based on the same welfare aggregate (income and consumption) used for poverty measurement. Table 1. Available indicators at the subnational level Indicator group Indicator description Poverty rates at different international poverty lines ($2.15, $3.65, $6.85 at 2017 Poverty PPP) Inequality Mean and Median (2017 PPP), Gini index, Theil index, Mean Log Deviation Multidimensional Multidimensional poverty Headcount ratio, and its indicators (Monetary, poverty Education Attainment, School Enrollment, Electricity, Sanitation, Water) Source: Authors. As both country boundaries and survey representativeness can change overtime, constructing a panel dataset of poverty at subnational areas is not a simple task. This issue could be solved by (1) regrouping areas to a new area that matches the previous definition of areas, or (2) maintaining a higher level of geographical disaggregation over time. Overall, in this first edition of the panel database, subnational poverty and inequality statistics are reported for about 1,650 unique panel subnational areas from 2003 to 2022 (resulting in 6,500 cross-sectional subnational data points) (Figure 5).13 12 For more details on the multidimensional poverty measure, see Poverty and Shared Prosperity Report, World Bank, 2018. 13 Suggested citation: World Bank. (2023). Subnational Poverty and Inequality Database. (version June 2023) [Data set]. World Bank Group. https://pipmaps.worldbank.org/. 12 Figure 5. Illustration of global subnational map with subnational data across years from the SPID Notes: Poverty is measured by the headcount ratio at $2.15/day 2017PPP, averaged for the period 2010 – 2020. Source: Authors Figure 6. Number of economies across World Bank regions in SPID Source: Authors 13 Figure 7. Number of subnational areas over time in SPID Source: Authors 4.3. Calculating GSAP line-up poverty estimates As mentioned, the Global Subnational Atlas of Poverty (GSAP) differs from the SPID as it presents the poverty estimates for all subnational areas from household survey data lined-up to a common year. Because household surveys are conducted in different years and at varying frequencies across economies, producing global and regional poverty estimates entails bringing each of the national poverty estimates to a common reference, or “line-up” year. For economies with surveys available in the reference year, the direct estimates of poverty from the surveys are used. For other cases, the poverty estimates are extrapolated or interpolated to the reference year using the country’s recent household survey data and real growth rates from national accounts data. The procedures for this exercise depend on the survey years available for the country. The lined-up poverty estimates at the subnational level are consistent in methodology with the national lined-up estimates and the global poverty numbers, such that the subnational 14 numbers aggregate up to the national line-up estimates in each country. The national line-up estimates for this subnational database are based on the line-up estimates from PIP in March 2023. When a survey is available only before the reference year, the consumption (or income) vector from the latest survey is extrapolated forward to the reference year using real growth rates of per capita GDP (or household final consumption expenditure) obtained from national accounts. Each observation in the welfare distribution is multiplied by the growth rate in per capita GDP (or household final consumption expenditure) between the reference year and the time of the survey. Poverty measures can then be estimated for the reference year. This procedure assumes distribution-neutral growth—that is, no change in inequality—and that the growth in national accounts is fully transmitted to growth in household consumption or income. If the only available surveys are after the reference year, a similar approach is applied to extrapolate backward. Surveys are interpolated when a survey is available at either side of the reference year. More details on the line-up method can be found in Prydz et al. (2019) and PIP’s methodological handbook. In this edition of the GSAP, the majority of developing and high-income economies in the world are represented (Figure 8). Another 50 economies are not included due to a lack of data or restricted access. Most of the survey input data is recent, data for 125 economies are from 2015 or later. See Table B.1 for a full variable list, and Table B.2 in Annex B for a full listing of surveys. Figure 8. Line-up poverty in the Spring 2023 edition of the GSAP utilizes survey data from 168 economies Source: GSAP, Spring 2023 edition 15 Subnational poverty can be computed for most cases; but for 48 economies, poverty indicators can only be shown at the national level (Figure 9). Sometimes this is because of restricted access to the underlying survey, or because the survey itself is too old, or the country is too small for further subnational disaggregation. In this edition 1,735 subnational areas are available, based on survey representativeness and the availability of spatial boundary files. In some cases, subnational areas in surveys are denoted only as urban or rural, which cannot be mapped, and are thus aggregated.14 See Figures B.1, B.2, and B.3 in Annex B for global maps of poverty rates. Figure 9. The frequency of the number of subnational areas Source: GSAP, Spring 2023 edition 14 Suggested citation: World Bank. (2023). Global Subnational Atlas of Poverty (version June 2023) [Data set]. World Bank Group. https://pipmaps.worldbank.org/. 16 Figure 10. The frequency of the number of subnational areas, by region and country Source: GSAP, Spring 2023 edition 17 5. Notes about the SPID and GSAP In both databases, there are several limitations that are worth noting. First, SPID and GSAP do not provide subnational data for countries which have no income/consumption household surveys, or which have only national data (and no data at a lower level), or which provides only limited data access (e.g., China provides highly aggregated data consisting of only 20-bin urban/rural distributions rather than micro data files). In those cases, in the GSAP only national line-up estimates are used. Second, the SPID offers estimates that are directly estimated from actual survey data (rather than prediction values), so it is constrained by data availability. Put differently, since not all surveys are available in all years, the SPID does not offer subnational data for all years. An implication of this is that the SPID might not provide updated population dynamics across areas (e.g., internal migration), which might not be captured in infrequent household surveys. On the other hand, GSAP offers subnational line-up poverty estimates for all countries in a common year, but requires additional assumptions (the line-up methodology is described above). Third, as is standard practice with most global databases on welfare, income/consumption data are both used for welfare measures because some countries implement income surveys, while others use consumption surveys. Fourth, not all countries employ the same approach to spatial deflation. In some countries in the GMD data collection, welfare aggregates from household survey data might not be adjusted for spatial price differences due to unavailability of spatial deflators. 18 6. References Azevedo, Joao Pedro, Minh Cong Nguyen, Paul Andres Corral Rodas, Hongxi Zhao, Q. Lu, J. J. Lee, Raul Andres Castaneda Aguilar, et al. 2018. “Global Subnational Poverty: An Illustration of the Global Geodatabase of Household Surveys.” World Bank, Washington , DC. Castaneda Aguilar R. Andres; Diaz-Bonilla, Carolina; Fujs, Tony H.M.J.; Jolliffe, Dean; Kotikula, Aphichoke; Lakner, Christoph; Lara Ibarra, Gabriel; Mahler, Daniel G.; Montalva Talledo, Veronica S.; Nguyen, Minh C.; Sanchez Castro, Diana M.; Suharnoko Sjahrir, Bambang; Tetteh-Baah, Samuel K.; Uochi, Ikuko; Viveros Mendoza, Martha C.; Wu, Haoyu; Yonzan, Nishant. March 2023 Update to the Poverty and Inequality Platform (PIP): What’s New (English). Global Poverty Monitoring Technical Note; no. 27 Washington, D.C.: World Bank. http://documents.worldbank.org/curated/en/099923403272329672/IDU089370bcb048b9 044fd0ab49037249b87aef6 Judy Yang, Minh Cong Nguyen, Natalie Kreitzer, Miyoko Asai. (2021). “Global Subnational Poverty: An update.” World Bank, Washington, DC. https://blogs.worldbank.org/opendata/introducing-second-edition-world-banks-global- subnational-atlas-poverty Prydz, Espen Beer, Dean Jolliffe, Christoph Lakner, Daniel Gerszon Mahler, and Prem Sangraula. 2019. “National Accounts Data Used in Global Poverty Measurement.” Global Poverty Monitoring Technical Note; no. 8; World Bank, Washington, DC. https://documents.worldbank.org/en/publication/documents- reports/documentdetail/664751553100573765/national-accounts-data-used-in-global- poverty-measurement Eurostat. NUTS - Nomenclature of territorial units for statistics, 2015, http://ec.europa.eu/eurostat/web/nuts. FAO. GAUL 2015 Boundaries, 2015, http://www.fao.org/geonetwork/srv/en/metadata.show?id=12691. GADM. GADM database of Global Administrative Areas, 2015, http://gadm.org/. World Bank. 2018. Poverty and Shared Prosperity 2018: Piecing Together the Poverty Puzzle 19 7. Annex A Table A.1. Economies-year used in the SPID Economy Data years Survey Subnational Number of acronyms layer subnational units Angola 2008, 2018 IBEP-MICS, Province 18 IDREA Albania 2012, 2015, 2016, 2017, 2018, 2019, HBS, LSMS County 12 2020 Armenia 2015, 2016, 2017, 2018, 2019, 2020, ILCS Province 11 2021 Australia 2010, 2014, 2016, 2018 SIH-HES- State 7 LIS, SIH- LIS Austria 2014, 2015, 2016, 2017, 2018, 2019, EU-SILC NUTS1 3 2020 Azerbaijan 2005 HBS Economic 10 region Burundi 2013 ECVMB Province 17 Belgium 2014, 2015, 2016, 2017, 2018, 2019, EU-SILC NUTS1 3 2020 Benin 2015, 2018 EHCVM, Department 12 EMICOV Burkina Faso 2014, 2018 EHCVM, Region 13 EMC Bangladesh 2010, 2016 HIES Division (6) 7 Bulgaria 2014, 2015, 2016, 2017, 2018, 2019, EU-SILC NUTS1 2 2020 Bosnia and 2007, 2011 HBS Entity 3 Herzegovina Belarus 2015, 2016, 2017, 2018, 2019, 2020 HHS Region 7 Bolivia 2015, 2016, 2017, 2018, 2019, 2020, EH Department 8 2021 Brazil 2015, 2016, 2017, 2018, 2019, 2020, PNADC-E1, State 27 2021 PNADC-E5 Bhutan 2012, 2017 BLSS District 20 Botswana 2015 BMTHS Statistical 7 region Central 2008 ECASEB Statistical 7 African region Republic Canada 2010, 2011, 2012, 2013, 2014, 2015, CIS-LIS, Province 10 2016, 2017, 2018 SLID-LIS Switzerland 2014, 2015, 2016, 2017, 2018 EU-SILC NUTS1 1 Chile 2015, 2017, 2020 CASEN Region 16 Côte d’Ivoire 2015, 2018 EHCVM, District 14 ENV Cameroon 2007, 2014 ECAM-III, Region 10 ECAM-IV 20 Congo, Dem. 2012 E123 Province (11) 11 Rep. Congo, Rep. 2011 ECOM Department 12 Colombia 2015, 2016, 2017, 2018, 2019, 2020, GEIH Department 24 2021 Comoros 2013 EESIC Region 3 Cabo Verde 2015 IDRF Island 9 Costa Rica 2015, 2016, 2017, 2018, 2019, 2020, ENAHO Province 6 2021 Cyprus 2014, 2015, 2016, 2017, 2018, 2019, EU-SILC NUTS1 1 2020 Czech 2014, 2015, 2016, 2017, 2018, 2019, EU-SILC NUTS2 8 Republic 2020 Germany 2010, 2011, 2012, 2013, 2014, 2015, GSOEP-LIS State 16 2016, 2017, 2018, 2019 Djibouti 2012, 2017 EDAM Region 6 Denmark 2014, 2015, 2016, 2017, 2018, 2019, EU-SILC NUTS1 1 2020 Dominican 2015, 2016, 2017, 2018, 2019, 2020, ECNFT- National 1, 2 Republic 2021 Q03, ENFT Ecuador 2015, 2016, 2017 ENEMDU Province 24 Egypt, Arab 2010, 2012, 2015, 2017, 2019 HIECS Statistical 4 Rep. region Spain 2014, 2015, 2016, 2017, 2018, 2019, EU-SILC NUTS2 19 2020 Estonia 2014, 2015, 2016, 2017, 2018, 2019, EU-SILC NUTS1 1 2020 Ethiopia 2010, 2015 HICES Region 11 Finland 2014, 2015, 2016, 2017, 2018, 2019, EU-SILC NUTS2 4 2020 Fiji 2013, 2019 HIES Division 4 France 2014, 2015, 2016, 2017, 2018, 2019, EU-SILC NUTS2 22 2020 United 2014, 2015, 2016, 2017 EU-SILC NUTS1 12 Kingdom Georgia 2015, 2016, 2017, 2018, 2019, 2020, HIS Region 10 2021 Ghana 2012, 2016 GLSS-VI, Region (10) 10 GLSS-VII Guinea 2012, 2018 EHCVM, Region 8 ELEP Gambia, The 2010, 2015, 2020 IHS Local 8 government areas Guinea- 2010, 2018 EHCVM, Region 9 Bissau ILAP-II Greece 2014, 2015, 2016, 2017, 2018, 2019, EU-SILC NUTS1 4 2020 Honduras 2015, 2016, 2017, 2018, 2019 EPHPM National 1 Croatia 2014, 2015, 2016, 2017, 2018, 2019, EU-SILC NUTS1 1 2020 Haiti 2012 ECVMAS Department 10 21 Hungary 2014, 2015, 2016, 2017, 2018, 2019, EU-SILC NUTS1 3 2020 Indonesia 2010, 2011, 2012, 2013, 2014, 2015, SUSENAS Province 33 2016, 2017, 2018, 2019, 2020, 2021, 2022 India 2009, 2011 NSS-SCH1 State 35 Ireland 2014, 2015, 2016, 2017, 2018, 2019, EU-SILC National 1 2020 Iran, Islamic 2015, 2016, 2017, 2018, 2019 HEIS Province 31 Rep. Iraq 2006, 2012 IHSES Governorate 18 Iceland 2014, 2015, 2016, 2017 EU-SILC National 1 Israel 2010, 2011, 2012, 2013, 2014, 2015, HES-LIS District 6 2016, 2017, 2018 Italy 2014, 2015, 2016, 2017, 2018, 2019, EU-SILC NUTS1 5 2020 Jordan 2008, 2010 HEIS Statistical 4 region Japan 2010, 2013 JHPS-LIS Region 8 Kazakhstan 2015, 2016, 2017, 2018 HBS Oblast 16 Kenya 2015 IHBS County 47 Kyrgyz 2015, 2017, 2018, 2019, 2020 KIHS Oblast 9 Republic Lao PDR 2012, 2018 LECS Region 4 Lebanon 2011 HBS Governorate 6 Liberia 2014, 2016 HIES County 16 Sri Lanka 2012, 2016, 2019 HIES District 25 Lesotho 2017 CMSHBS District 10 Lithuania 2014, 2015, 2016, 2017, 2018, 2019, EU-SILC National 1 2020 Luxembourg 2014, 2015, 2016, 2017, 2018, 2019, EU-SILC National 1 2020 Latvia 2014, 2015, 2016, 2017, 2018, 2019, EU-SILC National 1 2020 Morocco 2013 ENCDM Region 12 Moldova 2016, 2017, 2018, 2019, 2021 HBS Region 4 Madagascar 2012 ENSOMD Region 22 Maldives 2016, 2019 HIES Atoll 21 Mexico 2016, 2018, 2020 ENIGHNS State 32 North 2017, 2018, 2019 SILC-C Region 8 Macedonia Mali 2009, 2018 EHCVM, Region 9 ELIM Malta 2014, 2015, 2016, 2017, 2018, 2019, EU-SILC National 1 2020 Myanmar 2015, 2017 MLCS, Division 5 MPLCS Montenegro 2014, 2015, 2016, 2017, 2018 SILC-C Statistical 4 region Mongolia 2016, 2018 HSES Province 22 Mozambique 2008, 2014 IOF Province 11 Mauritania 2008, 2014 EPCV Region 13 22 Mauritius 2012, 2017 HBS District 10 Malawi 2016, 2019 IHS-IV, District 28 IHS-V Malaysia 2016, 2019 HIESBA, State 14 HIS Namibia 2009, 2015 NHIES Region 13 Niger 2014, 2018 ECVMA, Region 8 EHCVM Nigeria 2018 LSS State 36 Netherlands 2014, 2015, 2016, 2017, 2018, 2019, EU-SILC National 1 2020 Norway 2014, 2015, 2016, 2017, 2018, 2019 EU-SILC National 1 Nepal 2010 LSS-III Region 5 Pakistan 2004, 2005, 2007, 2010, 2011, 2013, HIES Division 4 2015, 2018 Panama 2015, 2016, 2017, 2018, 2019, 2021 EH Province 13 Peru 2015, 2016, 2017, 2018, 2019, 2020, ENAHO Region 25 2021 Philippines 2012, 2015, 2018, 2021 FIES Region 17 Papua New 2009 HIES Region 5 Guinea Poland 2017, 2018, 2019 EU-SILC NUTS1 7 Portugal 2014, 2015, 2016, 2017, 2018, 2019, EU-SILC National 1 2020 Paraguay 2018, 2019, 2020, 2021 EPH Statistical 8 region West Bank 2011, 2016 PECS State 2 and Gaza Romania 2014, 2015, 2016, 2017, 2018, 2019, EU-SILC NUTS1 4 2020 Russian 2019, 2020 HBS Oblast 81 Federation Rwanda 2013, 2016 EICV-IV, District 30 EICV-V Sudan 2009, 2014 NBHS State 15 Senegal 2011, 2018 EHCVM, Region 14 ESPS-II Solomon 2012 HIES Province 10 Islands Sierra Leone 2011, 2018 SLIHS District 13 El Salvador 2010, 2011, 2012, 2013, 2014, 2015, EHPM Department 14 2016, 2017, 2018, 2019, 2021 São Tomé 2017 IOF Province 2 and Príncipe Slovak 2014, 2015, 2016, 2017, 2018, 2019 EU-SILC National 1 Republic Slovenia 2014, 2015, 2016, 2017, 2018, 2019, EU-SILC National 1 2020 Sweden 2014, 2015, 2016, 2017, 2018, 2019, EU-SILC NUTS1 3 2020 Eswatini 2009, 2016 HIES Region 4 Seychelles 2013, 2018 HBS Region 6 23 Chad 2011, 2018 ECOSIT-III, Region 20 EHCVM Togo 2015, 2018 EHCVM, Region 6 QUIBB Thailand 2017, 2018, 2019, 2020, 2021 SES Province 77 Tajikistan 2009, 2015 HSITAFIEN Region 5 , TLSS Timor-Leste 2007, 2014 TLSLS Statistical 5 region Tonga 2015 HIES National 1 Tunisia 2010, 2015 NSHBCSL Region 7 Taiwan, 2010, 2013, 2016 FIDES-LIS County 20 China Tanzania 2011, 2018 HBS Region 26 Uganda 2016, 2019 UNHS Region 4 Ukraine 2014, 2015, 2016, 2019, 2020 HLCS Oblast 25 United States 2010, 2011, 2012, 2013, 2014, 2015, CPS-ASEC- State 51 2016, 2017, 2018, 2019, 2020 LIS Uzbekistan 2003 HBS Region 14 Vietnam 2010, 2012, 2014, 2016, 2018, 2020 VHLSS Region 6 Vanuatu 2010, 2019 HIES, NSDP Province 6 Kosovo 2012, 2013, 2014, 2015, 2016, 2017 HBS District 7 Yemen, Rep. 2014 HBS Governorate 22 South Africa 2010, 2014 IES, LCS Province 9 Zambia 2010, 2015 LCMS-VI, Province 10 LCMS-VII Zimbabwe 2017, 2019 PICES Province 10 24 Table A.2. Subnational Poverty and Inequality Database (SPID) Data Dictionary Indicator group Variable in data Indicator name Indicator description code Country code year Year of survey data surveyid Survey ID Metadata survname Survey name welfaretype Welfare type: (income or consumption) byvar Data level level Level of disaggregation geo_code2 GEO ID to link with the global shapefile Poverty rate at Poverty headcount ratio at $2.15 (2017 PPP) (% of poor215 $2.15 (2017 PPP) population) Poverty Poverty rate at Poverty headcount ratio at $3.65 (2017 PPP) (% of poor365 $3.65 (2017 PPP) population) Poverty rate at Poverty headcount ratio at $6.85 (2017 PPP) (% of poor685 $6.85 (2017 PPP) population) Survey mean/average consumption or income per mean Mean capita, total population (2017 PPP $ per day) Survey median consumption or income per capita, Inequality median Median total population (2017 PPP $ per day) gini Gini index Gini index (World Bank estimate) theil Theil Index Theil index (World Bank estimate) Mean Log mld Deviation MLD index (World Bank estimate) Multidimensional poverty, Monetary poverty (% of dep_poor1 Monetary population deprived) Education Multidimensional poverty, Educational attainment dep_educ_com Attainment (% of population deprived) Multidimensional Education Multidimensional poverty, Educational enrollment poverty dep_educ_enr Enrollment (% of population deprived) Multidimensional poverty, Electricity (% of dep_infra_elec Electricity population deprived) Multidimensional poverty, Sanitation (% of dep_infra_imps Sanitation population deprived) dep_infra_imp Multidimensional poverty, Drinking water (% of w2 Water population deprived) Multidimensional poverty, Headcount ratio (% of mdpoor_i1 MPM population) Source: Authors Note: geo_code2 is the geographic ID that can link with the SPID shapefile. 25 8. Annex B Figure B.1 Global Subnational Poverty Headcount Ratio at US$ 2.15/day 2017PPP (2019 line-up in GSAP). 26 Figure B.2. Global Subnational Poverty Headcount Ratio at US$3.65/day 2017PPP (2019 line-up in GSAP). 27 Figure B.3 Global Subnational Poverty Headcount Ratio at US$6.85/day 2017PPP (2019 line-up in GSAP). 28 Table B.1 Spring 2023 GSAP edition, Data Dictionary, and variable list Variable Description Note region Geographical region Region is based on geography except for advanced countries, which are grouped together. code Country code baseyear Year of survey data First year of survey fieldwork welfaretype Welfare type Monetary poverty rates are estimated either from expenditure or income, depending on the country case lineupyear Lineup year Lineup from survey years to a common year, in this case 2019 survname Survey name level Level of disaggregation geo_code2 GEO ID to link with the global shapefile poor215_ln Poverty rate at $2.15 (2017 PPP, lineup Poverty is lined-up to 2019 for all countries estimates of 2019) poor365_ln Poverty rate at $3.65 (2017 PPP, lineup Poverty is lined-up to 2019 for all countries estimates of 2019) poor685_ln Poverty rate at $6.85 (2017 PPP, lineup Poverty is lined-up to 2019 for all countries estimates of 2019) Notes: (1) Green= survey level identification information; Blue= subnational level identification information; Yellow = subnational level indicators (2) For China and India, poverty estimates shown are the 2019 national estimates reported in Poverty and Inequality Platform, PIP. (3) Country (national) line-up poverty number is as of April 2023 from Poverty and Inequality Platform, PIP. 29 Table B.2. List of economies-year survey data and geospatial level in the GSAP Region Region name Economy Economy name Survey Survey name Welfare type Geospatial level Number of code code year subnational areas EAP East Asia & Pacific CHN China 2019 CNIHS CONS ADM0 1 EAP East Asia & Pacific FJI Fiji 2019 HIES CONS ADM1 4 EAP East Asia & Pacific IDN Indonesia 2019 SUSENAS CONS ADM1 33 EAP East Asia & Pacific KIR Kiribati 2019 HIES CONS ADM0 1 EAP East Asia & Pacific LAO Lao PDR 2018 LECS CONS ADMx 18 EAP East Asia & Pacific MYS Malaysia 2019 HIS INC ADM1 14 EAP East Asia & Pacific FSM Micronesia, Fed. Sts. 2013 HIES CONS ADM0 1 EAP East Asia & Pacific MNG Mongolia 2018 HSES CONS ADM1/ADMx 22 EAP East Asia & Pacific MMR Myanmar 2017 MLCS CONS ADM1/ADMx 15 EAP East Asia & Pacific NRU Nauru 2012 HIES CONS ADM0 1 EAP East Asia & Pacific PNG Papua New Guinea 2009 HIES CONS ADM1/ADMx 5 EAP East Asia & Pacific PHL Philippines 2018 FIES INC ADM1 87 EAP East Asia & Pacific WSM Samoa 2013 HIES CONS ADM0 1 EAP East Asia & Pacific SLB Solomon Islands 2012 HIES CONS ADM0 10 EAP East Asia & Pacific THA Thailand 2019 SES CONS ADM1/ADMx 77 EAP East Asia & Pacific TLS Timor-Leste 2014 TLSLS CONS ADM1/ADMx 5 EAP East Asia & Pacific TON Tonga 2015 HIES CONS ADM0 1 EAP East Asia & Pacific TUV Tuvalu 2010 HIES CONS ADM0 1 EAP East Asia & Pacific VUT Vanuatu 2019 NSDP CONS ADM1 6 EAP East Asia & Pacific VNM Vietnam 2018 VHLSS CONS ADMx 6 ECA Europe & Central Asia ALB Albania 2019 HBS CONS ADMx 12 ECA Europe & Central Asia ARM Armenia 2019 ILCS CONS ADM1 11 ECA Europe & Central Asia AZE Azerbaijan 2005 HBS EXP ADM1/ADM2 10 ECA Europe & Central Asia BLR Belarus 2019 HHS CONS ADM1 7 ECA Europe & Central Asia BIH Bosnia and Herzegovina 2011 HBS CONS ADM1/ADM2 3 ECA Europe & Central Asia BGR Bulgaria 2020 EU-SILC INC ADMx/NUTS 2 ECA Europe & Central Asia HRV Croatia 2020 EU-SILC INC ADM0 1 ECA Europe & Central Asia CZE Czech Republic 2020 EU-SILC INC ADMx/NUTS 8 ECA Europe & Central Asia EST Estonia 2020 EU-SILC INC ADM0 1 ECA Europe & Central Asia GEO Georgia 2019 HIS CONS ADM1/ADMx 10 ECA Europe & Central Asia HUN Hungary 2020 EU-SILC INC ADMx/NUTS 3 ECA Europe & Central Asia KAZ Kazakhstan 2018 HBS CONS ADM1/DK 16 ECA Europe & Central Asia XKX Kosovo 2017 HBS CONS GADM1 7 ECA Europe & Central Asia KGZ Kyrgyz Republic 2019 KIHS EXP ADM1/GADM1 9 ECA Europe & Central Asia LVA Latvia 2020 EU-SILC INC ADM0 1 ECA Europe & Central Asia LTU Lithuania 2020 EU-SILC INC ADM0 1 ECA Europe & Central Asia MDA Moldova 2019 HBS INC GADMx 4 ECA Europe & Central Asia MNE Montenegro 2019 SILC-C INC ADM1/ADMx 4 ECA Europe & Central Asia MKD North Macedonia 2020 SILC-C INC ADM1/ADMx 8 ECA Europe & Central Asia POL Poland 2019 EU-SILC INC ADMx/NUTS 7 ECA Europe & Central Asia ROU Romania 2020 EU-SILC INC ADMx/NUTS 4 30 Region Region name Economy Economy name Survey Survey name Welfare type Geospatial level Number of code code year subnational areas ECA Europe & Central Asia RUS Russian Federation 2019 HBS EXP ADM1/ADMx/GADM1 82 ECA Europe & Central Asia SRB Serbia 2019 HBS EXP ADMx 4 ECA Europe & Central Asia SVK Slovak Republic 2020 EU-SILC INC ADM0 1 ECA Europe & Central Asia SVN Slovenia 2020 EU-SILC INC ADM0 1 ECA Europe & Central Asia TJK Tajikistan 2015 HSITAFIEN CONS ADM1/GADM1 5 ECA Europe & Central Asia TUR Türkiye 2019 HICES EXP ADM0 1 ECA Europe & Central Asia TKM Turkmenistan 1998 LSMS CONS ADM0 1 ECA Europe & Central Asia UKR Ukraine 2019 HLCS CONS ADM1/GADM1 25 ECA Europe & Central Asia UZB Uzbekistan 2003 HBS CONS ADM0 1 LAC Latin America & Caribbean ARG Argentina 2019 EPHC-S2 INC ADM0 1 LAC Latin America & Caribbean BLZ Belize 1999 LFS INC ADM0 1 LAC Latin America & Caribbean BOL Bolivia 2019 EH INC ADM1 8 LAC Latin America & Caribbean BRA Brazil 2019 PNADC-E1 INC ADM1 27 LAC Latin America & Caribbean CHL Chile 2020 CASEN INC ADM1/ADM2/ADMx 16 LAC Latin America & Caribbean COL Colombia 2019 GEIH INC ADM1/GADM2 24 LAC Latin America & Caribbean CRI Costa Rica 2019 ENAHO INC DKx 6 LAC Latin America & Caribbean DOM Dominican Republic 2019 ECNFT-Q03 INC ADMx 4 LAC Latin America & Caribbean ECU Ecuador 2019 ENEMDU INC ADM1 25 LAC Latin America & Caribbean SLV El Salvador 2019 EHPM INC ADM1 14 LAC Latin America & Caribbean GTM Guatemala 2014 ENCOVI INC ADM0 1 LAC Latin America & Caribbean GUY Guyana 1998 GLSMS INC ADM0 1 LAC Latin America & Caribbean HTI Haiti 2012 ECVMAS CONS ADM1 10 LAC Latin America & Caribbean HND Honduras 2019 EPHPM INC ADM0 1 LAC Latin America & Caribbean JAM Jamaica 2004 SLC CONS ADM0 1 LAC Latin America & Caribbean MEX Mexico 2019 ENIGHNS INC ADM1 32 LAC Latin America & Caribbean NIC Nicaragua 2014 EMNV INC ADMx 4 LAC Latin America & Caribbean PAN Panama 2019 EH INC ADM1/GADM1 13 LAC Latin America & Caribbean PRY Paraguay 2019 EPH INC ADM1/ADM2/ADMx 8 LAC Latin America & Caribbean PER Peru 2019 ENAHO INC ADM1 25 LAC Latin America & Caribbean LCA St. Lucia 2016 SLC-HBS INC ADM0 1 LAC Latin America & Caribbean SUR Suriname 1999 EHS INC ADM0 1 LAC Latin America & Caribbean TTO Trinidad and Tobago 1992 PHC INC ADM0 1 LAC Latin America & Caribbean URY Uruguay 2019 ECH INC ADM0 1 LAC Latin America & Caribbean VEN Venezuela, RB 2006 EHM INC ADM0 1 MNA Middle East & North America DZA Algeria 2011 ENCNVM CONS ADM0 1 MNA Middle East & North America DJI Djibouti 2017 EDAM CONS DKx 6 MNA Middle East & North America EGY Egypt, Arab Rep. 2017 HIECS CONS ADMx 23 MNA Middle East & North America IRN Iran, Islamic Rep. 2019 HEIS CONS ADM1/ADMx 31 MNA Middle East & North America IRQ Iraq 2012 IHSES CONS ADM1 18 MNA Middle East & North America JOR Jordan 2010 HEIS CONS ADMx 4 MNA Middle East & North America LBN Lebanon 2011 HBS CONS ADM1 6 MNA Middle East & North America MAR Morocco 2013 ENCDM CONS DKx 12 MNA Middle East & North America SYR Syrian Arab Republic 2003 HIES CONS ADM0 1 MNA Middle East & North America TUN Tunisia 2015 NSHBCSL CONS ADMx 7 MNA Middle East & North America PSE West Bank and Gaza 2016 PECS CONS ADMx 2 31 Region Region name Economy Economy name Survey Survey name Welfare type Geospatial level Number of code code year subnational areas MNA Middle East & North America YEM Yemen, Rep. 2014 HBS CONS ADM1/ADM2/ADMx 22 OHI Other High-Income AUS Australia 2018 SIH-LIS INC ADM1 7 OHI Other High-Income JPN Japan 2013 JHPS-LIS INC ADM1/ADMx 8 OHI Other High-Income KOR Korea, Rep. 2016 HIES-FHES-LIS INC ADM0 1 OHI Other High-Income TWN Taiwan, China 2016 FIDES-LIS INC GADM2 20 OHI Other High-Income AUT Austria 2020 EU-SILC INC ADMx/NUTS 3 OHI Other High-Income BEL Belgium 2020 EU-SILC INC ADM1 3 OHI Other High-Income CYP Cyprus 2020 EU-SILC INC ADM0 1 OHI Other High-Income DNK Denmark 2020 EU-SILC INC ADM0 1 OHI Other High-Income FIN Finland 2020 EU-SILC INC ADMx/NUTS 4 OHI Other High-Income FRA France 2019 EU-SILC INC ADM1 22 OHI Other High-Income DEU Germany 2019 GSOEP-LIS INC ADM1 16 OHI Other High-Income GRC Greece 2020 EU-SILC INC ADMx/NUTS 4 OHI Other High-Income ISL Iceland 2017 EU-SILC INC ADM0 1 OHI Other High-Income IRL Ireland 2019 EU-SILC INC ADM0 1 OHI Other High-Income ITA Italy 2019 EU-SILC INC ADMx/NUTS 5 OHI Other High-Income LUX Luxembourg 2020 EU-SILC INC ADM0 1 OHI Other High-Income NLD Netherlands 2020 EU-SILC INC ADM0 1 OHI Other High-Income NOR Norway 2020 EU-SILC INC ADM0 1 OHI Other High-Income PRT Portugal 2020 EU-SILC INC ADM0 7 OHI Other High-Income ESP Spain 2020 EU-SILC INC ADM1 19 OHI Other High-Income SWE Sweden 2020 EU-SILC INC ADMx/NUTS 3 OHI Other High-Income CHE Switzerland 2019 EU-SILC INC ADM0 1 OHI Other High-Income GBR United Kingdom 2017 EU-SILC INC ADMx/NUTS 12 OHI Other High-Income ISR Israel 2018 HES-LIS INC ADM1 7 OHI Other High-Income MLT Malta 2020 EU-SILC INC ADM0 1 OHI Other High-Income ARE United Arab Emirates 2019 HIES INC ADM0 1 OHI Other High-Income CAN Canada 2018 CIS-LIS INC ADM1 10 OHI Other High-Income USA United States 2019 CPS-ASEC-LIS INC ADM1 51 SAS South Asia BGD Bangladesh 2016 HIES EXP ADM1 7 SAS South Asia BTN Bhutan 2017 BLSS CONS ADM1 20 SAS South Asia IND India 2019 CPHS EXP ADM0 1 SAS South Asia MDV Maldives 2019 HIES EXP ADM1 18 SAS South Asia NPL Nepal 2010 LSS-III EXP ADM1 5 SAS South Asia PAK Pakistan 2018 HIES EXP ADM1 4 SAS South Asia LKA Sri Lanka 2019 HIES EXP ADM2 25 SSA Sub-Saharan Africa AGO Angola 2018 IDREA CONS ADM1 18 SSA Sub-Saharan Africa BEN Benin 2018 EHCVM CONS ADM1 12 SSA Sub-Saharan Africa BWA Botswana 2015 BMTHS CONS GADM2/GADMx 7 SSA Sub-Saharan Africa BFA Burkina Faso 2018 EHCVM CONS ADM1 13 SSA Sub-Saharan Africa BDI Burundi 2013 ECVMB CONS ADM1 17 SSA Sub-Saharan Africa CPV Cabo Verde 2015 IDRF CONS ADM1 9 SSA Sub-Saharan Africa CMR Cameroon 2014 ECAM-IV CONS ADM1 10 SSA Sub-Saharan Africa CAF Central African Republic 2008 ECASEB EXP ADMx 7 SSA Sub-Saharan Africa TCD Chad 2018 EHCVM CONS GADM1/GADMx 20 32 Region Region name Economy Economy name Survey Survey name Welfare type Geospatial level Number of code code year subnational areas SSA Sub-Saharan Africa COM Comoros 2013 EESIC CONS-EXP ADM1 3 SSA Sub-Saharan Africa COD Congo, Dem. Rep. 2012 E123 CONS ADM1 11 SSA Sub-Saharan Africa COG Congo, Rep. 2011 ECOM CONS ADM1 12 SSA Sub-Saharan Africa CIV Côte d’Ivoire 2018 EHCVM CONS ADM1 14 SSA Sub-Saharan Africa SWZ Eswatini 2016 HIES CONS ADM1 4 SSA Sub-Saharan Africa ETH Ethiopia 2015 HICES CONS ADM1 11 SSA Sub-Saharan Africa GAB Gabon 2017 EGEP CONS ADM0 1 SSA Sub-Saharan Africa GMB Gambia, The 2020 IHS CONS GADM1/GADM2/GADMx 8 SSA Sub-Saharan Africa GHA Ghana 2016 GLSS-VII CONS ADM1 10 SSA Sub-Saharan Africa GIN Guinea 2018 EHCVM CONS ADM1 8 SSA Sub-Saharan Africa GNB Guinea-Bissau 2018 EHCVM CONS ADM1 9 SSA Sub-Saharan Africa KEN Kenya 2015 IHBS CONS-EXP ADM1/GADM1 47 SSA Sub-Saharan Africa LSO Lesotho 2017 CMSHBS CONS ADM1 10 SSA Sub-Saharan Africa LBR Liberia 2016 HIES CONS ADM1/ADM2 16 SSA Sub-Saharan Africa MDG Madagascar 2012 ENSOMD CONS ADM1 22 SSA Sub-Saharan Africa MWI Malawi 2019 IHS-V CONS GADM1/ADM2 28 SSA Sub-Saharan Africa MLI Mali 2018 EHCVM CONS ADM1 11 SSA Sub-Saharan Africa MRT Mauritania 2014 EPCV CONS ADM1 13 SSA Sub-Saharan Africa MUS Mauritius 2017 HBS CONS ADM1 10 SSA Sub-Saharan Africa MOZ Mozambique 2014 IOF CONS ADM1/ADM2 11 SSA Sub-Saharan Africa NAM Namibia 2015 NHIES CONS ADM1 13 SSA Sub-Saharan Africa NER Niger 2018 EHCVM CONS ADM1 8 SSA Sub-Saharan Africa NGA Nigeria 2018 LSS CONS ADM1 36 SSA Sub-Saharan Africa RWA Rwanda 2016 EICV-V CONS ADM2 30 SSA Sub-Saharan Africa STP São Tomé and Príncipe 2017 IOF CONS ADM1 2 SSA Sub-Saharan Africa SEN Senegal 2018 EHCVM CONS ADM1 14 SSA Sub-Saharan Africa SYC Seychelles 2018 HBS INC GADMx 6 SSA Sub-Saharan Africa SLE Sierra Leone 2018 SLIHS CONS ADM2/ADMx 13 SSA Sub-Saharan Africa SOM Somalia 2017 SHFS-W2 CONS ADM0 1 SSA Sub-Saharan Africa ZAF South Africa 2014 LCS CONS ADM1 9 SSA Sub-Saharan Africa SSD South Sudan 2016 HFS-W3 CONS ADM0 1 SSA Sub-Saharan Africa SDN Sudan 2014 NBHS CONS ADM1 18 SSA Sub-Saharan Africa TZA Tanzania 2018 HBS CONS ADM1/GADMx 26 SSA Sub-Saharan Africa TGO Togo 2018 EHCVM CONS ADM1/ADM2/ADMx 6 SSA Sub-Saharan Africa UGA Uganda 2019 UNHS CONS ADMx 4 SSA Sub-Saharan Africa ZMB Zambia 2015 LCMS-VII CONS ADM1/ADMx 10 SSA Sub-Saharan Africa ZWE Zimbabwe 2019 PICES CONS ADM1 10 33