Global Poverty Monitoring Technical Note 36 March 2024 Update to the Poverty and Inequality Platform (PIP) What’s New R. Andres Castaneda Aguilar, Adriana Castillo, Nancy P. Devpura, Reno Dewina, Carolina Diaz-Bonilla, Ifeanyi Edochie, Maria G. Farfan Bertran, Jaime Fernandez Romero, Elizabeth Foster, Tony H. M. J. Fujs, Maria F. Gonzalez Icaza, Dean Jolliffe, Erwin W. Knippenberg, Nandini Krishnan, Christoph Lakner, Gabriel Lara Ibarra, Diego G. Lestani, Daniel G. Mahler, Veronica S. Montalva Talledo, Jose Montes, Laura Moreno Herrera, Minh C. Nguyen, Sergio Olivieri, Anna Luisa Paffhausen, Silvia Redaelli, Trinidad B. Saavedra, Diana M. Sanchez Castro, Samuel K. Tetteh-Baah, Martha C. Viveros Mendoza, Haoyu Wu, Nishant Yonzan and Nobuo Yoshida March 2024 Keywords: What’s New; March 2024; PIP Development Data Group Development Research Group Poverty and Equity Global Practice Group GLOBAL POVERTY MONITORING TECHNICAL NOTE 36 Abstract The March 2024 update to the Poverty and Inequality Platform (PIP) involves several changes to the data underlying the global poverty estimates. In particular, some welfare aggregates have been revised, and the CPI, national accounts, and population input data have been updated. This document explains these changes in detail and the reasoning behind them. Moreover, 101 new country-years have been added, bringing the total number of surveys to more than 2,300. Depending on the availability of recent survey data, global and regional poverty estimates are reported up to 2022. This is the first time PIP is reporting global poverty estimates post-2019, covering the period of the COVID- 19 pandemic. All authors are with the World Bank. Corresponding authors: Christoph Lakner (clakner@worldbank.org) and Minh C. Nguyen (mnguyen3@worldbank.org). The authors are thankful for comments and guidance received from Deon Filmer, Haishan Fu, and Luis-Felipe Lopez-Calva. We would also like to thank the countless Poverty Economists that 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. The authors gratefully acknowledge financial support from the UK government through the Data and Evidence for Tackling Extreme Poverty (DEEP) Research Programme. This note has been cleared by Umar Serajuddin. 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 2. Global and regional line-up and coverage rules ......................................................................... 5 3. Changes to welfare aggregates.................................................................................................... 8 3.1. Argentina 2003-2021 ........................................................................................................... 8 3.2. Bangladesh 2022 .................................................................................................................. 9 3.3. Brazil 2016-2022 ............................................................................................................... 11 3.4. Chile 2006, 2009, 2011, 2013, 2015, 2017, 2020 .............................................................. 11 3.5. Colombia 2021 ................................................................................................................... 12 3.6. India 2019-2021 ................................................................................................................. 13 3.7. Iran 2013-2019 ................................................................................................................... 16 3.8. Luxembourg Income Study (LIS) ...................................................................................... 17 3.9. Mexico 1989-2020 ............................................................................................................. 18 3.10. Peru 2018-2021 ................................................................................................................ 19 3.11. Saint Lucia 2015 .............................................................................................................. 20 3.12. Samoa 2013...................................................................................................................... 21 3.13. Sao Tome and Principe 2017 ........................................................................................... 22 3.14. Thailand 2017-2021 ......................................................................................................... 22 3.15. Uruguay 2000................................................................................................................... 23 3.16. Vietnam 2020 ................................................................................................................... 23 4. Economy-years added ............................................................................................................... 24 4.1. New surveys from the West African Economic and Monetary Union (WAEMU) ........... 24 4.2. Syria 2007, 2009, 2022 ...................................................................................................... 24 4.3. Zambia 2022 ...................................................................................................................... 26 5. Changes to CPI data .................................................................................................................. 27 6. Changes to national accounts and population data ................................................................... 27 7. Lining up methodology: revised extrapolation and interpolation rules .................................... 28 7.1. Interpolation of poverty data for Syria............................................................................... 29 8. Comparability database............................................................................................................. 30 9. References ................................................................................................................................. 31 10. Appendix ................................................................................................................................. 33 10.1. Complete list of new country-years ................................................................................. 33 10.2. CPI data sources ............................................................................................................... 36 1 1. Introduction The March 2024 global poverty update from the World Bank revises previously published poverty and inequality estimates and reports global estimates of poverty up to 2022, covering the period of the COVID-19 for the first time. Regional poverty series are also available up to 2022 for all regions, except Sub-Saharan Africa and the Middle East and North Africa where there is currently a lack of sufficient recent data (Table 1). The most recent year for which poverty is reported for Sub-Saharan Africa is 2019 and for the Middle East and North Africa 2018 (see Section 2 for more details). Extreme poverty rates in 2022 were lower than 2019 pre-pandemic estimates for all regions with recent data, including East Asia and the Pacific, Europe and Central Asia, Latin America and the Caribbean, the advanced countries, and South Asia (see Tables 1 and 2). However, for the world, the level of extreme poverty was still slightly higher in 2022 than in 2019. Taken together, this suggests the uneven nature of recovery from and resilience against global crises, including economic contractions during the pandemic and inflationary pressures following Russia’s invasion of Ukraine in 2022. More prosperous regions recovered faster from the pandemic, even in the face of food and energy price hikes. By contrast, poverty reduction became harder in Sub-Saharan Africa, where most of the extreme poor live. At $3.65 and $6.85, the poverty lines more relevant for assessing poverty in lower- and upper-middle-income countries, global poverty rates were lower in 2022 relative to 2019. This is consistent with the recovery being faster in more prosperous regions, considering that Sub-Saharan Africa accounts for a smaller share of the global poor at these higher lines compared to the extreme poverty line. Table 2 documents revisions to the regional and global poverty estimates between the September 2023 data vintage and the March 2024 data vintage for the 2019 reference year at the three global poverty lines. The global poverty headcount ratio at the International Poverty Line ($2.15 per person per day, 2017 PPP) has reduced marginally by 0.1 percentage points to 8.9 percent, resulting in a revision in the number of poor people from 701 to 689 million. The global reduction in the millions of extreme poor occurs despite an upward revision in Sub-Saharan Africa (14 million). The reduction is driven by Europe and Central Asia and the Middle East and North Africa, where new survey data have recently become available to replace extrapolations of very old 2 surveys.1 At the $3.65 and $6.85 poverty lines, poverty rates have reduced by 0.6 and 0.7 percentage point (i.e., 52 and 44 million fewer poor people, respectively). These downward revisions in poverty estimates are driven by Europe and Central Asia and South Asia. Table 1 Poverty estimates for 2022, the most recent year with global and regional estimates $2.15 (2017 PPP) $3.65 (2017 PPP) $6.85 (2017 PPP) Head- Region Survey Head- Number Number Head- Number count coverage count of poor of poor count of poor ratio (%) ratio (%) (mil) (mil) ratio (%) (mil) (%) East Asia & Pacific 94.4 1.0 22 6.3 134 29.2 622 Europe & Central Asia 93.1 0.5 2 1.7 8 8.6 42 Latin America & Caribbean 85.8 3.5 23 8.9 58 25.2 165 Middle East & North Africa 28.7 Other High Income 63.2 0.3 3 0.5 6 1.0 11 South Asia 82.8 9.7 187 39.0 749 79.2 1,520 Sub-Saharan Africa 38.5 Eastern & Southern Africa 36.9 Western & Central Africa 40.8 World 74.4 9.0 712 22.7 1,804 45.5 3,619 Source: PIP Note: Regional poverty estimates are reported if survey coverage is above 50% within a three-years window of the reference year with a break in 2020. The global estimate is reported if survey coverage is above 50% and coverage for low- and lower-middle-income countries is above 50%. For 2022 the latter is 63.9%. The update in the global and regional poverty series reflects a broad set of revisions to survey and auxiliary data at the country level. Table 3 provides an overview of the survey data used in this update. Revisions have been made to 90 welfare distributions from the previous update to improve the quality of the data (see Section 3) and more than 100 country-years have been added (see Section 4), bringing the total number of distributions to 2,367.2 PIP now has survey data for 169 countries, including Grenada which is a new economy that has been added to the database. 1 The regional estimate for Middle East and North Africa cannot be shown since it does not meet the 50% population cut-off. New survey data for 2022 have been added for Syria and Uzbekistan. Before this update, the latest surveys for both countries were from 2003. 2 A distribution is defined as a unique combination of country, year, and data type (income or consumption). There are country-years with both income and consumption data. 3 Table 2 Poverty estimates for reference year 2019, changes between September 2023 and March 2024 PIP vintage by region and poverty line $2.15 (2017 PPP) $3.65 (2017 PPP) $6.85 (2017 PPP) Survey Headcount Number of Headcount Number of Headcount Number of Coverage Region ratio (%) poor (mil) ratio (%) poor (mil) ratio (%) poor (mil) 2019 (%) Mar Sep Mar Sep Mar Sep Mar Sep Mar Sep Mar Sep Mar 2024 2023 2024 2023 2024 2023 2024 2023 2024 2023 2024 2023 2024 East Asia & Pacific 97.4 1.2 1.2 25 25 7.6 7.8 161 164 32.1 32.3 676 680 Europe & Central Asia 87.3 2.2 0.5 11 2 6.1 2.0 30 10 15.0 10.8 74 53 Latin America & Caribbean 87.2 4.3 4.2 28 27 10.6 10.2 68 66 28.0 27.2 179 175 Middle East & North Africa 48.3 n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a Other High Income 82.3 0.6 0.6 7 7 0.8 0.8 9 9 1.3 1.3 15 15 South Asia 96.4 10.5 10.6 196 198 43.7 42.3 814 788 81.8 80.6 1523 1501 Sub-Saharan Africa 54.1 35.4 36.7 397 411 62.9 63.9 706 717 86.9 87.3 975 979 Eastern & Southern Africa 29.6 n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a Western & Central Africa 90.0 27.3 27.3 124 124 57.4 57.6 261 262 85.5 85.8 388 390 World 84.6 9.0 8.9 701 689 24.1 23.4 1865 1812 46.9 46.4 3633 3589 Source: PIP Note: Poverty estimates in 2019 are not reported for Eastern and Southern Africa, and Middle East and North Africa due to a limited survey data coverage of less than 50% of the regional population; however, the available data are incorporated into the poverty estimates for Sub-Saharan Africa and the world, respectively. Survey coverage for low- and lower-middle-income countries for 2019 is 78.7%. The 2011 PPP-based estimates are also available in PIP. Table 1 Overview of survey data Description Sep 2023 Mar 2024 Difference Distributions 2259 2367 108 Country-years 2182 2283 101 Countries 168 169 1 Country-years with income and consumption 77 84 7 Surveys revised 90 Note: A distribution is defined as a unique combination of country, year, and data type. There are country-years with both income and consumption data. This update also incorporates the latest versions of consumer price index (CPI), population, and national accounts data from our standard sources, including the World Development Indicators (WDI), World Economic Outlook (WEO), and Maddison Project Database (MPD). See Sections 5-6 for more details on the changes in the auxiliary data. 4 In this update, revisions have been introduced into the PIP methodology that is used for filling in the gaps for years without any survey data, since not all countries have surveys in every year. This is necessary for producing regional and global poverty aggregates since the aggregates require a balanced sample of countries over time. The revisions to this methodology barely make any difference to the global poverty series and improve the regional and country-level poverty series from 1981 to 2022 substantially. See Section 7 for more details. 2. Global and regional line-up and coverage rules With this update, we report for the first time estimates of global poverty until 2022, spanning the COVID-19 pandemic. In the previous PIP update, poverty estimates were reported for Latin America and the Caribbean and South Asia until 2021, and for East Asia and the Pacific, Europe and Central Asia and the group of advanced countries (“Other High Income”) until 2020. Poverty estimates could not be reported for Sub-Saharan Africa, the Middle East and North Africa, and for the world due to insufficient survey data coverage. As explained in Castaneda et al. (2020) and the PIP Methodology Handbook, a region is considered to have coverage if at least 50% of its population has coverage in the line-up year. There is global coverage if at least 50% of the global population has coverage and at least 50% of the population in low- and lower-middle-income countries (LIC/LMIC) has coverage. Regional and global data coverage in turn depend on coverage at the country level. Conventionally, a country is considered to have coverage if it has survey data representative of its population within three years either side of the line-up year. Consider the survey conducted in Sierra Leone in 2018. By the conventional rule, Sierra Leone has data coverage for the years ranging from 2015 to 2021. The population of Sierra Leone will be included in the share of the total population of Sub-Saharan Africa having data coverage for 2015 through 2021. In other words, the 2018 survey is recent enough to contribute to poverty estimates reported for Sub-Saharan Africa for 2015 through 2021. The 2018 distribution will be extrapolated forward using growth rates in national accounts data to report poverty estimates for years up to 2021. For any year in which at least 50% of the population in Sub-Saharan Africa has data coverage, a regional poverty estimate is reported. 5 The conventional coverage rules were applied until COVID-19 hit. While we routinely extrapolate surveys forward, doing so over the period of COVID-19 based on pre-COVID surveys is problematic. During the COVID-19 pandemic, countries experienced economic shocks and volatility at an unprecedented global scale. To avoid making statements about poverty in the pandemic with pre-pandemic data, new coverage rules were introduced for deciding the reporting of regional and global poverty estimates in the pandemic years. We shifted from the three-year coverage rule to an annual coverage rule. Starting from 2020, a country is considered to have data coverage only if a survey was conducted in the year in question. Based on the country coverage data, poverty estimates are reported for regions and for the world in a year with at least 50% data coverage. In addition, at least 50% of the population in low- and lower-middle-income countries should have data coverage for a poverty number to be reported for the world. Given these new rules, the most recent PIP updates reported a global series that ended in 2019. For example, in the September 2023 PIP update, we did not report global poverty estimates for 2020, even though the three-year coverage rules were met. Four years after COVID-19 hit, it makes sense to revert to the conventional, three-year coverage rules. However, it is important to introduce a “break” in 2020, which was an extraordinary year for the global economy. Pre-COVID-19 survey data will still not count when assessing coverage for post-2019 years. Equivalently, post-2019 survey data will not count when assessing coverage for pre-COVID-19 years. As a result, the three-year coverage rule is no longer symmetric for years close to 2020. That means, coverage for 2019 will be based on survey data from only three years before, which is more restrictive. For this reason, 2019 poverty estimates will still not be reported for the Middle East and North Africa in this March 2024 PIP update, even though new 2022 data from Syria are available that could otherwise increase data coverage to more than 50% for the region. Consider again the survey conducted in Sierra Leone in 2018. The population of Sierra Leone will be included in the share of the total population of Sub-Saharan Africa having data coverage for 2015 through 2019, but not beyond due to the break introduced in 2020. Figure 1 shows the coverage data for all the regions for the period 2019-2022. With these results, poverty estimates are reported in this March 2024 PIP update for 2020-2022 for all regions, except Sub-Saharan Africa and the Middle East and North Africa. 6 Figure 1 Data coverage by region using three-year window rule with a break in 2020 Note: AFE = Eastern and Southern Africa, AFW = Western and Central Africa, EAP = East Asia & Pacific, ECA = Europe & Central Asia, LAC = Latin America & Caribbean, MNA = Middle East and North Africa, OHI = Other High Income (a group of advanced countries), SAS = South Asia, SSA = Sub-Saharan Africa, WLD = World By this March 2024 PIP update, enough time has passed to allow for the processing of survey data collected in 2020, as well as the many surveys that were postponed to the subsequent years. In the next PIP update, more recent surveys are expected to be added to the PIP database so that we hope to report poverty estimates to 2022 for all regions. 7 3. Changes to welfare aggregates 3.1. Argentina 2003-2021 • 2003-2012: There is a slight revision to imputed rent: The rent imputation model was imputing the coefficients for four regions incorrectly.3 This impacts the imputed rent value and, therefore, the overall household income. The effect on the poverty headcount rates is up to 0.2 pp. Table 4 Changes to poverty and inequality estimates, Argentina 2003-2012 Poverty headcount Poverty headcount Poverty headcount Gini Index $2.15 $3.65 $6.85 Sep Mar Sep Mar Sep Mar Sep Mar Country Year 2023 2024 2023 2024 2023 2024 2023 2024 Argentina 2003 5.28 5.36 12.54 12.72 29.56 29.72 50.90 51.00 Argentina 2004 3.81 3.88 9.60 9.76 24.73 24.87 48.38 48.49 Argentina 2005 2.63 2.68 7.87 8.01 20.78 20.96 47.71 47.81 Argentina 2006 2.12 2.17 5.91 6.03 16.79 16.90 46.35 46.45 Argentina 2007 1.72 1.80 4.89 4.99 15.40 15.57 46.17 46.29 Argentina 2008 1.59 1.64 4.68 4.80 14.07 14.23 44.88 44.99 Argentina 2009 1.52 1.55 4.00 4.06 12.77 12.88 43.65 43.76 Argentina 2010 0.75 0.75 3.09 3.15 11.25 11.35 43.58 43.68 Argentina 2011 0.71 0.71 2.10 2.11 9.01 9.11 42.65 42.74 Argentina 2012 0.64 0.65 2.29 2.33 8.50 8.57 41.33 41.44 • 2013-2017: The household survey used for Argentina does not gather data on how much rent tenants pay or how much owners would pay if they were renting. To address this data constraint, the Encuesta Nacional de Gastos de los Hogares (ENGHO) is used as an alternative source. By applying a hedonic pricing model to the ENGHO data, the team estimates the value of imputed rent. Subsequently, using the derived coefficients along with the same observable variables, the imputed income for non-tenants is predicted within the survey that is used to report poverty and inequality. In this update, the implicit rent model is refined by integrating data from the latest ENGHO conducted in 2018. The model now uses the coefficients derived from both the 2012 and 2018 ENGHO. In the old version of the harmonization, the imputed rent exclusively relied on the 3 A model is used to impute the rental value of owner-occupied housing, dwellings received as a gift, usufruct, or ceded dwellings. 8 2012 ENGHO coefficients. This improvement impacts the imputed rent value and, therefore, overall household income. The effect on poverty and inequality is reported in the table below. Table 5 Changes to poverty and inequality estimates, Argentina 2013-2027 Poverty headcount Poverty headcount Poverty headcount Gini Index $2.15 $3.65 $6.85 Sep Mar Sep Mar Sep Mar Sep Mar Country Year 2023 2024 2023 2024 2023 2024 2023 2024 Argentina 2013 0.62 0.67 1.99 2.07 8.46 8.56 40.95 41.09 Argentina 2014 0.61 0.60 2.10 2.13 9.14 9.47 41.63 41.82 Argentina 2016 0.67 0.70 2.14 2.31 8.78 9.21 42.03 42.35 Argentina 2017 0.42 0.60 2.13 2.36 7.87 8.25 41.13 41.44 • 2018-2021: Changes have also been made to imputed rent. The imputed rent model now uses the coefficients derived from the 2018 ENGHO. In the old version of the harmonization, the rent imputation used the 2012 ENGHO coefficients. The effect on poverty and inequality is reported in the table below. Table 6 Changes to poverty and inequality estimates, Argentina 2018-2021 Poverty headcount Poverty headcount Poverty headcount Gini Index $2.15 $3.65 $6.85 Sep Mar Sep Mar Sep Mar Sep Mar Country Year 2023 2024 2023 2024 2023 2024 2023 2024 Argentina 2018 0.77 1.04 2.61 2.87 9.67 10.35 41.34 41.72 Argentina 2019 0.78 1.11 2.94 3.38 11.18 12.01 42.91 43.32 Argentina 2020 1.07 1.17 3.46 3.99 14.08 15.38 42.34 42.75 Argentina 2021 0.96 0.89 2.49 2.80 10.62 11.43 42.01 42.45 3.2. Bangladesh 2022 The Bangladesh Bureau of Statistics (BBS) conducts the Household Income and Expenditure Survey (HIES) approximately every five years. From 2000 onwards, BBS followed similar sampling designs and covered almost the same items, especially for food and non-food consumption modules. However, in HIES-2022, various changes were made to enhance data quality. These are: (i) the introduction of COICOP (Classification of individual consumption according to purpose), (ii) the addition of new items in the food (increasing from 149 to 263) and non-food (increasing from 261 to 441) consumption modules, (iii) switching from CAFE (Computer Assisted Field Entry) to CAPI (Computer Assisted Personal Interview) for data 9 collection/entry and effective monitoring of the field activities. These changes helped improve data quality for the HIES-2022. But at the same time, they posed challenges in comparing consumption data with the previous surveys. The numbers shown in the September 2023 PIP update were a simple attempt to achieve comparability by constructing a 2022 consumption aggregate solely based on the common items with previous rounds. In the current PIP update, poverty numbers and inequality are calculated using the full consumption aggregate, as reported officially by BBS. It implies that consumption, and consequently the welfare indicators, are no longer comparable with the estimates for earlier years. For this PIP update, the poverty headcount is 5.0% using the $ 2.15 international line and 30.0% and 74.1% using the $ 3.65 and $ 6.85 lines, respectively. Table 7 Changes to poverty and inequality estimates, Bangladesh 2022 Poverty headcount Poverty Poverty Gini Index $2.15 headcount $3.65 headcount $6.85 Sep Mar Sep Mar Sep Mar Sep Mar Country Year 2023 2024 2023 2024 2023 2024 2023 2024 Bangladesh 2022 9.58 5.01 42.32 30.03 83.14 74.10 31.77 33.37 To address the comparability issue, Fernandez et al. (2023, forthcoming) proposed a two-step imputation process that uses the full consumption aggregate from 2022 to reconstruct a comparable consumption trend backwards in time. The approach allows for a consistent assessment of poverty and inequality measures over the last 12 years (using the 2010, 2016 and 2022 surveys). Their findings suggest that applying this correction to previous survey rounds would have revealed a notable reduction in poverty rates, as measured against the national poverty lines. Specifically, the poverty rate at the official national poverty line would have decreased by approximately 10.6 percentage points between 2010 and 2016, and by a further 7.8 points between 2016 and 2022. Similarly, poverty rates at the national extreme poverty line would have seen a decline of around three percentage points in the earlier period and a more significant drop of 3.6 points in the later period. This work helps interpret the recent trend reported in PIP, which has a break in comparability between 2016 and 2022, as described above. 10 3.3. Brazil 2016-2022 The Continuous National Household Sample Survey (PNAD-C) questionnaire in Brazil does not directly include a question for imputed rent. However, it gathers data on the actual rent paid by tenants. Using this information, a hedonic model is estimated to predict the implicit rent for households that do not pay rent. In the latest data revision, we have redefined the variable that captures the type of energy used for cooking, one of the variables used in the hedonic model. In previous versions, the variable grouped in the same category i) bottled or piped gas and ii) firewood or coal. In this update, the variable groups bottled or piped gas together with electric energy, with firewood or coal being a separate category. This modification reflects the closer alignment in rental market value between homes that use electricity or gas versus homes that use firewood or charcoal. This adjustment leads to minor changes in poverty and inequality. Table 8 Changes to poverty and inequality estimates, Brazil 2016-2021 Poverty Poverty Poverty headcount $2.15 headcount $3.65 headcount $6.85 Gini Index Sep Mar Sep Mar Sep Mar Sep Mar Country Year 2023 2024 2023 2024 2023 2024 2023 2024 Brazil 2016 4.71 4.73 10.62 10.62 27.02 27.04 53.34 53.37 Brazil 2017 5.27 5.30 11.00 11.01 27.05 27.10 53.33 53.33 Brazil 2018 5.30 5.31 10.88 10.88 26.66 26.71 53.87 53.89 Brazil 2019 5.39 5.41 10.75 10.76 26.17 26.25 53.49 53.51 Brazil 2020 1.95 1.95 5.32 5.33 18.73 18.73 48.88 48.90 Brazil 2021 5.82 5.83 11.26 11.26 28.36 28.38 52.92 52.94 3.4. Chile 2006, 2009, 2011, 2013, 2015, 2017, 2020 • 2006-2017: The Chilean National Statistics Institute (INE) has released a new set of weights for the CASEN that take into account: i) New population projections based on the 2017 Census and ii) Changes in the calibration methodology. This update is being adopted in the harmonized data and will be used from now on. Further information can be found here and here. 11 • 2020: In addition to the change mentioned for 2006-2017, in 2020, there was a correction to the variable that captures the labor income of people who were employed and did not receive a salary because of the pandemic (they reported having received labor income in the previous month of the survey collection). These observations were mistakenly treated as missing values in labor income instead of 0, which has now been corrected. This change impacts the indicator which flags “coherent” income observations (SEDLAC variable cohh=1).4 Only coherent observations are included in the sample. Table 9 Changes to poverty and inequality estimates, Chile 2006-2020 Poverty headcount Poverty Poverty $2.15 headcount $3.65 headcount $6.85 Gini Index Sep Mar Sep Mar Sep Mar Sep Mar Country Year 2023 2024 2023 2024 2023 2024 2023 2024 Chile 2006 2.09 2.05 7.81 7.60 29.89 29.07 47.29 47.71 Chile 2009 1.72 1.67 5.77 5.54 25.40 24.41 46.99 47.41 Chile 2011 0.93 0.90 4.02 3.86 21.23 20.91 46.05 46.87 Chile 2013 0.53 0.52 2.04 1.96 12.86 12.49 45.83 46.83 Chile 2015 0.43 0.43 1.62 1.55 10.19 9.73 44.37 45.34 Chile 2017 0.34 0.36 1.06 1.03 7.52 7.47 44.44 45.27 Chile 2020 0.75 1.34 1.68 2.70 8.01 9.88 44.92 47.01 3.5. Colombia 2021 The Colombian household survey (GEIH) used a 2005 census-based geostatistical framework until 2021. From 2021 onwards, it adopted a new framework based on the 2018 census, introducing an updated sample design for municipalities. To maintain comparability, both the old and new frameworks were used in parallel in 2021. The Colombian National Statistics Institute (DANE) released comparable monetary and extreme poverty data for 2021 and 2022. The 2021 data was adjusted with bridging factors to match the 2018 geostatistical framework, while the 2022 data was directly based on this updated framework. Therefore, the underlying harmonized microdata for this year (2021) was replaced with the latest version. 4 In the SEDLAC harmonization, some observations are identified as incoherent. For example, this applies to individual observations that are identified as employed but record no income in the main occupation. 12 Changes in the 2021 and the new 2022 data limit comparability between GEIH 2021-2022 and previous surveys (2008-2020). This situation might be addressed once DANE publishes historical data comparable to the most recent years. Further information can be found here. Table 10 Changes to poverty and inequality estimates, Colombia 2021 Poverty headcount Poverty headcount Poverty headcount $2.15 $3.65 $6.85 Gini Index Sep Mar Sep Mar Sep Mar Sep Mar Country Year 2023 2024 2023 2024 2023 2024 2023 2024 Colombia 2021 6.61 7.33 15.98 16.44 39.20 38.78 51.50 55.13 3.6. India 2019-2021 This update summarizes two changes for 2019-2021 poverty and inequality estimates – namely (1) transition from rural/urban population shares in the World Development Indicators (WDI) to population shares of the Ministry of Health (MoH), and (2) transition from stochastic to multiple imputations (MI). Shift to population projections of the Ministry of Health (MOH) The latest population census in India is from 2011. To bridge the data gap, the MoH has provided population projections until 2036. The latest series adopts these projections for India. These estimates, available both at the national and state levels, make it feasible to estimate state poverty rates, which was previously not possible using WDI (which only provides a split by urban/rural). However, for calculating the headcount number of poor, population estimates from the WDI will continue to be used to align with global reporting. This adjustment results in slightly higher extreme poverty estimates (0.02-0.03 percentage points), attributed to higher rural population estimates in MoH compared to WDI. Shift to multiple imputations The previous estimates for 2019-2021 were imputed based on 200 vectors following the Approach 1 methodology from Roy and van der Weide (2022). A main challenge of this imputation is the non-normality of the welfare indicator’s distribution in the NSS Consumption Expenditure Survey (CES) 2011, as a result of which a multiple imputation approach can produce a large bias in poverty estimation (Schenker and Taylor, 1996). To address this, Roy and van der Weide (2022) proposed 13 a new approach where the log of household expenditure per capita was imputed by drawing errors from an empirical distribution of regression residuals. The revised estimation introduced in this update adopts a more conventional strategy, which entails normalizing the dependent variable to align with the normality assumption before employing a standard multiple imputation protocol. The merits of this more conventional approach include its foundation in a broad spectrum of scholarly work and guidelines (Corral et al., 2022; Rubin, 1987; Schafer, 1999), its empirically tested and validated reliability both within and beyond the sample (Yoshida et al., 2022; Zhang et al., 2024), and the ease of implementation using standard Stata syntax. The revised estimation process was implemented using the following steps: (1) The initial step involved training the prediction model on Uniform Recall Period (URP)- based expenditure from NSS-CES 2011, the latest available unit-record data at the time of this update. MI assumes a normal distribution for the imputed parameter (log per capita consumption) and the error term. However, the distribution of the observed log per capita consumption deviates from this assumption. The team explored various methods, including Box and Cox (1964), Log shift, and Johnson (1949), to transform the distribution to normal. While the Box-Cox transformation predicted point estimates for poverty in rural areas within a 95% confidence interval (CI) of the actual rural poverty estimates, none of the transformations yielded in-sample point estimates for urban areas that lie within a 95% CI of the actuals. (2) Exploring an alternative specification, the team used log household consumption as the dependent variable instead of log per capita consumption. After imputing the log household consumption, the log per capita consumption and poverty rates are estimated. This transformation resulted in predicted poverty rates that fell within the 95% CI of the actual poverty estimates for both rural and urban areas without requiring any transformation. Consequently, the decision was made to use log household consumption as the dependent variable. (3) The predicted log per capita consumption distribution closely mirrored the actual distribution from 0.25th to the 99.75th percentile, except for a few outliers. These outliers, however, had a large impact on the predicted Gini coefficient. To resolve this, the team 14 winsorized 0.25% from both ends of the dependent variable. This enabled the calculation of inequality estimates without the loss of prediction accuracy in the poverty rates. (4) After successfully predicting in-sample poverty and inequality estimates in the training model (NSS-CES 2011), the team followed the same procedure to predict out-of-sample poverty and inequality estimates in the modeling dataset – Consumer Pyramids Household Survey (CPHS) 2019, 2020, and 2021. Additionally, rural/urban population shares from the MoH were used to derive national-level poverty estimates in CPHS. All estimates from CPHS follow official NSS-CES 2011 in using the uniform recall period (URP) method, with a 30-day recall period for consumption. (5) As mentioned above, if the dependent variables follow a normal distribution, the multiple imputation works well outside the training dataset to obtain reliable poverty estimates (Yoshida et al. 2022). Also, the revised methodology includes a sampling weight adjustment to correct the sampling bias of CPHS following Roy and Van der Weide (2022). Zhang et al. (2024) have further validated that integrating multiple imputations with sampling weight adjustments markedly diminishes the sampling bias associated with poverty estimates in surveys that rely on imputed welfare indicators. The table below shows the final changes in poverty and inequality estimates following both adjustments. The revised methodology results in higher extreme poverty rates (0.46-1 percentage points) compared to the previous estimates. Table 11 Changes to poverty and inequality estimates, India 2019-2021 Poverty headcount Poverty headcount Poverty headcount $2.15 $3.65 $6.85 Gini Index Sep Mar Sep Mar Sep Mar Sep Mar Country Year 2023 2024 2023 2024 2023 2024 2023 2024 India 2019 12.73 13.21 45.89 43.98 82.37 80.73 35.02 33.81 India 2020 14.72 15.46 49.68 48.26 84.01 82.98 34.77 33.77 India 2021 11.90 12.92 46.54 44.06 82.98 81.77 34.21 32.77 15 3.7. Iran 2013-2019 Estimates from 2013/14 to 2019/20 have been revised with a new consumption aggregate. The consumption aggregates were constructed using Iran’s Household Expenditure and Income Survey, which is collected annually by the Statistical Center of Iran and made publicly available on their website. The new consumption aggregate follows the most recent guidelines from the World Bank on how to construct consumption aggregates for poverty and inequality analysis (Mancini and Vecchi, 2022). These guidelines ensure that the consumption aggregate is consistent over time, accounting for health expenditure, among other best practices. The new consumption aggregate also incorporates temporal and spatial deflation within each survey, as documented in Amendola et al. (2023). This is particularly important in the context of Iran, which has experienced high year-on-year inflation. This revision leads to considerable changes, especially to the Gini index. Due to limited documentation on the previously used aggregate, it is impossible to fully understand the source of the differences. A possible explanation is the correction for temporal and spatial price differences which can significantly influence within-country inequality (and which is now included). For the new aggregate, the Gini coefficient for the nominal and deflated consumption aggregate can be compared (Fig. 2). Shifting from the deflated to the nominal aggregate reduces the gap between the old and new estimates by two-thirds. Table 12 Changes to poverty and inequality estimates, Iran 2013-2019 Poverty headcount Poverty Poverty $2.15 headcount $3.65 headcount $6.85 Gini Index Sep Mar Sep Mar Sep Mar Sep Mar Country Year 2023 2024 2023 2024 2023 2024 2023 2024 Iran 2013 0.31 0.29 2.95 2.75 19.39 20.90 37.36 33.97 Iran 2014 0.81 0.62 4.50 4.49 21.40 23.52 38.78 34.81 Iran 2015 0.74 0.40 4.52 4.04 22.03 23.00 39.47 35.26 Iran 2016 0.69 0.45 4.59 4.15 21.99 23.13 39.97 35.91 Iran 2017 0.56 0.33 4.03 3.71 20.03 21.57 40.80 36.69 Iran 2018 0.87 0.67 4.63 4.31 22.28 24.36 42.00 37.42 Iran 2019 0.92 0.73 5.71 5.62 26.24 29.10 40.94 36.48 16 Figure 2 Trends in the Gini coefficient for Iran by welfare type 3.8. Luxembourg Income Study (LIS) As in previous editions, welfare data for the following nine economies is drawn from the Luxembourg Income Study (LIS) published by the LIS Data Center: Australia, Canada, Germany, Israel, Japan, South Korea, United States, United Kingdom and Taiwan, China.5 Additionally, PIP includes some historical LIS data (typically before the early 2000s, prior to the existence of EU-SILC) for some European countries that currently use the EU-SILC.6 The break in comparability (between LIS and EU-SILC) is indicated through PIP’s main outputs.7 In all cases we use disposable income per capita in the form of 400 bins (see Chen et al., 2018 for more details). For this release, LIS data was downloaded on 12 December 2023. The following 19 country-years have been added to PIP, as they became available in LIS during the past year: 5 The term country, used interchangeably with economy, does not imply political independence but refers to any territory for which authorities report separate social or economic statistics. 6 These additional pre-EUSILC surveys were introduced in the March 2020 update (Atamanov et al., 2020). 7 The comparability between surveys is indicated through the variables comparable_spell and survey_comparability available in the main outputs on the PIP’s website, Stata command and API. For more on comparability see PIP’s Methodological Manual. 17 • GBR (United Kingdom): 2021 • ISR (Israel): 2019, 2020, 2021 • ITA (Italy): 1977, 1978, 1979, 1980, 1981, 1982, 1983, 1984, 2002 • SWE (Sweden): 2001 • TWN (Taiwan): 2017, 2018, 2019, 2020, 2021 Finally, the following 16 country-years have been revised, as explained in more detail on the LIS website: • CAN (Canada): 1997 • FRA (France): 1984, 2000 • GBR (United Kingdom): 1995 • ISR (Israel): 2010, 2012, 2013 • ITA (Italy): 1986, 1987, 1989, 1991, 1993, 1995 • SWE (Sweden): 2000, 2002 • TWN (Taiwan): 2016 3.9. Mexico 1989-2020 • 1989-2014: SEDLAC’s harmonization of the income aggregate has changed for 1989- 2014. The change in the household income aggregate results from additional information in the questionnaire which allows computing the welfare aggregate as the average income received over the last six months instead of the last month. This revision was previously implemented for the 2016, 2018, and 2020 data. In this version, it has been extended to the entire data series to enhance comparability. For a more detailed explanation, refer to Castaneda et al. (2022). • 1996: The change mentioned for 1989-2014 was also applied in 1996. However, the magnitude of changes this year are larger due to a harmonization error in the previous version. Specifically, in the previous version, the income from the most distant month was mistakenly used instead of the income from the month immediately preceding the interview. Much of the difference in magnitude, as well as direction of the revision, for 1996 comes from this. Without the error, the old version of the 1996 data (last month approach) would give lower poverty, and the adjustment to the average of the last 6 months would not lead to a fall in the poverty rate. 18 • 2016, 2018 and 2020: The income aggregate for these years has been corrected. The income component from "work performed in the five months prior to last month" was erroneously excluded in the previous version of the data. This change result in higher income level and thus lower poverty rates than previously estimated and reported. Table 13 Changes to poverty and inequality estimates, Mexico 1989-2020 Poverty headcount Poverty headcount Poverty headcount $2.15 $3.65 $6.85 Gini Index Sep Mar Sep Mar Sep Mar Sep Mar Country Year 2023 2024 2023 2024 2023 2024 2023 2024 Mexico 1989 9.05 8.71 24.22 23.51 53.75 53.50 50.63 50.11 Mexico 1992 7.38 7.87 20.83 20.92 47.64 49.00 52.61 52.31 Mexico 1994 7.31 7.93 20.43 21.74 46.60 47.64 52.81 53.42 Mexico 1996 18.10 13.96 36.54 32.06 63.25 60.40 53.55 51.97 Mexico 1998 12.90 15.24 27.79 30.61 55.20 56.72 51.68 53.28 Mexico 2000 8.92 10.68 21.29 23.65 48.18 49.88 52.58 53.42 Mexico 2002 6.59 7.52 18.95 20.63 45.31 47.05 50.09 50.58 Mexico 2004 5.95 7.15 15.98 17.66 42.13 44.16 50.03 50.34 Mexico 2005 6.54 7.80 16.09 17.75 40.90 42.93 50.12 50.94 Mexico 2006 4.15 5.35 13.11 14.99 37.16 39.66 48.94 49.71 Mexico 2008 5.36 5.98 14.49 16.62 37.78 40.24 49.88 50.85 Mexico 2010 4.52 4.70 12.76 14.34 38.09 40.18 47.21 47.66 Mexico 2012 3.84 4.20 12.24 13.45 35.42 37.34 48.71 49.60 Mexico 2014 3.71 3.98 12.01 13.09 38.30 40.34 48.72 48.91 Mexico 2016 3.21 2.32 10.71 8.86 33.58 31.42 47.68 46.91 Mexico 2018 2.63 1.88 9.36 7.83 31.06 28.80 46.71 45.96 Mexico 2020 3.10 2.13 9.93 8.19 32.55 30.27 45.40 44.60 3.10. Peru 2018-2021 • 2018-2020: SEDLAC’s harmonization of the income aggregate has changed for 2018-2020. The change in the household income aggregate results from considering additional information provided by the questionnaire. Income from several cash transfers programs that were not previously considered (Programa Beca, Programa por Servicio Militar Voluntario, Profesor Contratado por el Estado and Propina de la Escuela de la PNP o FFAA) is now included. This income was mistakenly not added to the non-labor income variable in the previous version of the harmonized data, thus affecting total household income. This has now been corrected. 19 • 2021: In addition to the change mentioned for 2018-2020, income from new cash transfer programs (Bono Electricidad, Bono Niño, Bono ONP and Programa Social Contigo) is now included in the 2021 harmonized survey. Table 14 Changes to poverty and inequality estimates, Peru 2018-2021 Poverty headcount Poverty headcount Poverty headcount $2.15 $3.65 $6.85 Gini Index Sep Mar Sep Mar Sep Mar Sep Mar Country Year 2023 2024 2023 2024 2023 2024 2023 2024 Peru 2018 3.58 3.57 10.69 10.66 30.36 30.31 42.39 42.37 Peru 2019 3.00 2.98 9.66 9.62 28.86 28.79 41.56 41.53 Peru 2020 5.88 5.82 17.47 17.37 43.02 42.85 43.79 43.74 Peru 2021 2.87 2.77 10.24 9.99 33.74 33.42 40.25 40.12 3.11. Saint Lucia 2015 Up until now, PIP used an income-based welfare aggregate to produce internationally comparable poverty and inequality estimates for Saint Lucia using data from the latest Survey of Living Conditions and Household Budgets (SLC-HBS), which was conducted between November 2015 and July 2016. This was done to ensure comparability within the Latin America and Caribbean (LAC) region, where most countries use income-based poverty measures. The harmonization followed the SEDLAC methodology. In Saint Lucia, and many other countries in the Caribbean, official poverty measurement, however, uses a consumption-based welfare aggregate. To increase the availability of comparable poverty and inequality estimates for the Caribbean sub-region, a new harmonized consumption-based welfare aggregate for Caribbean countries with recent household surveys and available microdata has been constructed. As a result of this effort, the following changes will be made to the information previously shown in PIP for Saint Lucia: • The primary welfare measure for Saint Lucia will be consumption-based instead of income- based as has been the case so far. This is to align with the welfare measure used for national poverty monitoring, which is consumption-based. • The survey year will be given as 2015.78 instead of 2016, as has been the case so far. Data collection for the last Saint Lucia SLC-HBS started in 2015. During the update the team 20 realized that the survey year had not been specified correctly for the income-based measures used so far. This has now been corrected. • Minor corrections have been made to the income-based welfare aggregate that will be used as a secondary measure for poverty and inequality, including the following: a. Inclusion of imputed rent reported by households reporting “squatting” as a type of living arrangement. The previous income harmonization did not consider implicit rent for these types of households. The SEDLAC guidelines, however, recommend inclusion of implicit rent for homeowners and non-market tenants (those living rent- free or at subsidized prices), which includes squatters. b. Inclusion of income of individuals who did not report any type of employment but did report labor income. c. Change in the factor used to convert income from secondary occupation from daily to monthly value. To convert from daily to monthly income, income from secondary occupation is now multiplied by 30.5 (close to 365/12). In the previous income harmonization, the conversion factor used was 30. The table below shows the updated and previous key poverty and equity estimates using the income-based welfare aggregate for Saint Lucia: Table 15 Changes to poverty and inequality estimates, Saint Lucia 2015 Poverty headcount Poverty headcount Poverty headcount $2.15 $3.65 $6.85 Gini Index Sep Mar Sep Mar Sep Mar Sep Mar Country Year 2023 2024 2023 2024 2023 2024 2023 2024 Saint 2015 5.12 5.22 11.71 11.74 25.34 25.26 51.23 51.24 Lucia 3.12. Samoa 2013 Household size has been corrected by applying filter questions to define a household member. This filter question relates to the occupancy of the household (hm1112), and we exclude those members that: (i) left the household during the past 12 months with no plans to return (hm1112 = 4) or (ii) died in the last 12 months (hm1112 = 5). This is equivalent to excluding household members not included in the household list (hm1100 = 2). Doing this in the individual dataset gives a household size equivalent to the “De facto household size” in the aggregate dataset (variable hb017), which conforms to the country’s definition of household membership. The 21 filter questions were not applied previously for the 2013 survey (but it was applied for the 2018 survey). With the changes, the population weights and per capita welfare were affected. Thus, the poverty rates and Gini index have changed slightly. Table 16 Changes to poverty and inequality estimates, Samoa 2013 Poverty headcount Poverty headcount Poverty headcount $2.15 $3.65 $6.85 Gini Index Sep Mar Sep Mar Sep Mar Sep Mar Country Year 2023 2024 2023 2024 2023 2024 2023 2024 Samoa 2013 1.15 1.15 10.51 10.51 43.27 43.28 38.73 38.74 3.13. Sao Tome and Principe 2017 The recent adjustment made to the welfare aggregate stemmed from the realization that household size calculations did not fully utilize questionnaire data. Specifically, an individual should not be considered as a household member if absent for more than 6 months or has no plans to stay in the household. In the previous version, there was incomplete use of the information from the questionnaire, as the variable which informs about plans to stay had not been used. This oversight resulted in inaccuracies in household size determination for a small number of households. The correction of this issue resulted in slight adjustments in per capita consumption for the affected households and their population weight, consequently leading to minor changes in poverty estimates. Table 17 Changes to poverty and inequality estimates, Sao Tome and Principe 2017 Poverty headcount Poverty headcount Poverty headcount $2.15 $3.65 $6.85 Gini Index Sep Mar Sep Mar Sep Mar Sep Mar Country Year 2023 2024 2023 2024 2023 2024 2023 2024 Sao Tome and 2017 15.55 15.72 44.84 45.03 79.73 79.70 40.75 40.73 Principe 3.14. Thailand 2017-2021 Household size has been corrected for these survey data years by excluding domestic workers from the household. This follows the definition of household size applied by the NSO for poverty measurement. The changes in household size affect the population weight and per capita consumption, and hence poverty rates and the Gini index change slightly. 22 Table 18 Changes to poverty and inequality estimates, Thailand 2017-2021 Poverty headcount Poverty headcount Poverty headcount $2.15 $3.65 $6.85 Gini Index Sep Mar Sep Mar Sep Mar Sep Mar Country Year 2023 2024 2023 2024 2023 2024 2023 2024 Thailand 2017 0.04 0.04 0.82 0.82 15.11 15.12 36.53 36.38 Thailand 2018 0.03 0.03 0.95 0.95 15.61 15.61 36.41 36.41 Thailand 2019 0.10 0.10 0.58 0.58 13.15 13.15 34.86 34.86 Thailand 2020 0.05 0.05 0.67 0.67 13.18 13.18 34.99 34.99 Thailand 2021 0.01 0.01 0.56 0.56 12.16 12.17 35.12 34.92 3.15. Uruguay 2000 Changes have been made to the variable that captures the years of completed education. This change impacts the indicator which flags “coherent” income observations (SEDLAC variable cohh=1). Only coherent observations are included in the sample. Changes to the poverty and inequality estimates are only detectable at the third decimal. Table 19 Changes to poverty and inequality estimates, Uruguay 2000 Poverty Poverty headcount Poverty headcount headcount $2.15 $3.65 $6.85 Gini Index Sep Mar Sep Mar Sep Mar Sep Mar Country Year 2023 2024 2023 2024 2023 2024 2023 2024 Uruguay 2000 0.66 0.66 2.45 2.45 12.48 12.48 42.91 42.91 3.16. Vietnam 2020 A coding error that resulted in incorrect sample size and weights being used has been corrected. This correction results in small changes in poverty and inequality estimates. Table 20 Changes to poverty and inequality estimates, Vietnam 2020 Poverty Poverty headcount Poverty headcount headcount $2.15 $3.65 $6.85 Gini Index Sep Mar Sep Mar Sep Mar Sep Mar Country Year 2023 2024 2023 2024 2023 2024 2023 2024 Vietnam 2020 0.65 0.65 3.78 3.78 18.73 18.72 36.81 36.79 23 4. Economy-years added Table A1 in the Appendix gives the complete list of new economy-years added to the PIP database. A total of 101 new economy-years were added. For several special cases, the methodologies for constructing the new data points are described below. 4.1. New surveys from the West African Economic and Monetary Union (WAEMU) Benin, Burkina Faso, Cote d’Ivoire, Guinea-Bissau, Mali, Niger, Senegal, Togo, and Chad all have data from new surveys, the second round of the Enquête Harmonisée sur le Conditions de Vie des Ménages conducted in 2021/22 (just 2022 for Chad). The surveys are highly comparable to those of the first round conducted in 2018/19 and so is the methodology to construct the nominal aggregate. As for the first round, no spatial deflation is applied for international poverty, and monthly temporal deflators based on the official CPI are applied. 4.2. Syria 2007, 2009, 2022 Prior to this March 2024 PIP update, poverty and inequality estimates from survey data were available for Syria until 2003. For the purposes of creating an aggregate for the Middle East and North Africa and the world, the 2003 distribution has been extrapolated forward for almost two decades, using the growth rate in final household consumption (HFCE) per capita. With this update, new survey data sources have been leveraged to provide poverty and inequality estimates for 2007, 2009 and 2022. The 2007 and 2009 survey estimates are based on grouped data, while the 2022 survey estimates are based on microdata. These new data points fill knowledge gaps in PIP about the evolution of poverty in Syria before and after the conflict started in 2011. In addition, poverty aggregates for the Middle East and North Africa and the world now include these new data points. The data source used for interpolating surveys in Syria was also changed, see Section 7 below. Grouped data for 2007 and 2009 The Central Bureau of Statistics, the statistical agency in Syria, conducted two nationally representative budget surveys in 2007 and 2009 (Household Income and Expenditure Survey - HIES). However, microdata were not publicly released. As described in Redaelli et al. (forthcoming), poverty estimates based on international poverty lines for 2007 and 2009 have been 24 estimated using information available on the website of Syria’s Central Bureau of Statistics (CBS) and the standard approach used in PIP to estimate poverty when only grouped data are available (World Bank, 2024).8 For both 2007 and 2009, CBS publishes information on average household consumption by decile of the national welfare distribution. These deciles of household consumption were converted into per capita consumption by leveraging information on average household size by decile. An assumption of the conversion to per capita household income is that per capita consumption is not a function of household size within deciles, which appears reasonable. Furthermore, the conversion uses average household size by decile estimated on 2003 microdata, thus assuming no change between 2003 and 2007/2009. Survey data for 2022 Poverty estimates for 2022 rely on the Demographic and WASH survey conducted under the Humanitarian Needs Assessment Programme (HNAP) in the summer of 2022. Based on a collaboration between the World Bank and HNAP, the questionnaire of this survey was improved in order to better proxy consumption, compared to the earlier round in 2018 (see Redaelli et al. forthcoming). While the consumption aggregate based on the summer 2022 Demographic and WASH survey better captures household welfare compared to the previous HNAP survey rounds, the welfare aggregate remains incomplete, leaving out housing and durables.9 Due to differences in survey questionnaires, poverty estimates based on the 2022 Demographic and WASH survey, and estimates for 2007 and 2009 based on the Household Income and Expenditure Survey conducted by CBS are not comparable. 8 The link to the website is: https://web.archive.org/web/20151008101348/http:/www.cbssyr.sy/index-EN.htm. 9 It is to be noted that these challenges are not unique to Syria as the estimation of rental value of housing and consumer flow of durables remain one of the most difficult to measure components of the welfare aggregate (see Mancini and Vecchi, 2022) 25 4.3. Zambia 2022 In 2022 the Zambia Statistics Agency implemented a new Living Conditions Monitoring Survey (LCMS). This survey serves as the primary data for estimating official poverty and inequality figures, with technical assistance provided by the World Bank. The previous survey was conducted in 2015, and since then the country has undergone several economic- and weather-related shocks, including droughts, the COVID-19 pandemic, and an external debt default. All in all, this resulted in a 2.7 percent reduction in real GDP per capita over the 7-year period (ZamStats National Accounts). The changes introduced in the 2022 LCMS rendered it unsuitable for estimating comparable poverty and inequality figures, thus compromising the estimation of trends. To restore comparability, two estimation methods were rigorously employed to measure the poverty trend and one for assessing the inequality trend. Both strategies rely on identifying a subset of the consumption aggregate that is comparable across surveys and on the assumption that the underlying relationship between the comparable and non-comparable portions is stable over time. The comparable portion of consumption includes health, a subset of education, clothing, financial services, durables, and housing. In 2015, this comparable portion accounts for 33.7% of total consumption and has a correlation of 0.987 with total consumption. More concretely, the first method – Survey of Well-being via Instant and Frequent Tracking (SWIFT) – trains a consumption model using actual 2015 survey data and then uses the model to estimate the 2022 consumption aggregate. The methodology uses multiple imputation and machine learning techniques to train the model. The SWFT approach was first created in 2014 and has since been implemented in numerous countries to restore comparability between surveys, increase frequency of official statistics, conduct rapid poverty monitoring in crises contexts, among others (Yoshida et al., 2022). The second method follows the approach taken by Deaton (2003) to restore comparability between the National Sample Survey (NSS) rounds in India. This method consists of estimating the probability that a household falls below the poverty line. It does so by relying on the relationship between the comparable portion of household consumption and total household consumption. Both models also use additional household characteristics and food consumption dummies. Both approaches can be used to estimate poverty, but only the first produces estimates of inequality. 26 The SWIFT method is used to generate the final consumption aggregate for this PIP update, while the second method is used for validation purposes. For more technical details, see Zambia Statistics Agency and World Bank (2023). 5. Changes to CPI data The baseline source of CPI data has been updated to the IMF’s International Financial Statistics (IFS) as of 1 November 2023. Lakner et al. (2018) provide an overview of the various CPI series that are used in PIP. Table A2 in the Appendix to this note gives the up-to-date source of the deflator for all countries included in PIP as of the current update. 6. Changes to national accounts and population data We have incorporated new national accounts and population data from the latest vintages of our standard sources. The primary source of national accounts data is the February 2024 vintage of the World Development Indicators (WDI). As before, when WDI data are missing, data from the IMF’s World Economic Outlook (WEO), October 2023 version are used. Supplementary data from the Maddison Project Database (MPD), 2020 version are further used to fill missing observations. For a more complete series, national accounts data are chained on backward or forward using growth rates in WEO data, or MPD data, when WDI data are missing. The population data have also been revised to the February 2024 vintage of the World Development Indicators (WDI). Compared to the previous June 2023 vintage of WDI used for the previous PIP update, there have been small revisions to population data. In addition, a change has been made to the urban and rural population data used for India. The total population is still taken from WDI, but the urban and rural shares have been revised. Previously, these shares were taken from the WDI. Now, they reflect the shares estimated by the Ministry of Health of India from 2012-2022. Prior to 2012, for the surveys of 1993-94, 2004-05, 2009-10, and 2011-12, they reflect the urban and rural shares implicit in the survey data using the sum of weights of urban and rural households. In between these surveys and prior to 1993-94, the 27 urban and rural shares are interpolated and extrapolated using the urban and rural shares from WDI. The exact calculations are available at https://github.com/PIP-Technical-Team/aux_pop 7. Lining up methodology: revised extrapolation and interpolation rules The method used to extrapolate and interpolate welfare vectors to a reference year has been revised. Extrapolation and interpolation of survey-based estimates is necessary because the surveys measuring poverty are not fielded each year for all countries, but global and regional poverty estimates need to be reported for a common year. The prior extrapolation method shifted the welfare aggregates obtained from a household survey by a common scale factor equal to a measure of growth taken from national accounts data: real growth in Household Final Consumption Expenditure per capita (henceforth ‘growth in HFCE’) for countries outside of Sub- Saharan Africa, and growth in real GDP per capita (henceforth ‘growth in GDP’) for countries in Sub-Saharan Africa (Prydz et al., 2019). The new approach makes three changes (Table 21). The first major change introduces a “passthrough rate”, a factor that determines what share of growth from national accounts “passes through” to the survey welfare measure (Prydz et al., 2022; Ravallion, 2003). We adopt a passthrough rate of 0.7 for countries that use consumption to measure welfare. For example, if a country using consumption experienced growth in national accounts of 2 percent, household survey consumption is increased by only 1.4 percent. The selection of a 0.7 passthrough rate, and the decision to only apply this to countries that use a consumption aggregate is based on empirical evidence, which shows a systematic tendency of growth in mean consumption measured in surveys to consistently underestimate growth in national accounts (Mahler et al., 2022). There is no similar pattern for income. The second major change alters the measure of economic growth that is used to scale up survey- based estimates of welfare. In particular, we suggest using GDP growth for all countries classified as low income or lower middle-income, and HFCE growth for countries classified as upper middle- income and high-income. Empirical analysis shows that this change improves the accuracy of the extrapolation method (Mahler and Newhouse, 2024). 28 Table 21 Summary of changes in lining up methodology Old method New method • 1 (full passthrough) for all • 0.7 for consumption aggregates Passthrough rate countries • 1 for income aggregates • HFCE for upper middle-income and Measure of economic • HFCE outside of Sub- high-income countries Saharan Africa growth • GDP for low and lower middle- • GDP in Sub-Saharan Africa income countries Handling of missing HFCE data (when • Use GDP for entire • Use GDP only for years where HFCE extrapolation is missing preferred) A final smaller change relates to the handling of missing HFCE data. This change is easiest understood through a hypothetical example. Suppose we want to extrapolate welfare for a country from 2016 to 2021 and 2022, and the preferred national accounts data for the country is HFCE. Suppose further that HFCE for this country is available 2016-2021 but not for 2022, while GDP is available all years. Previously, the 2016-2021 extrapolation would rely on HFCE while the 2016- 2022 extrapolation would rely on GDP. Now, for the 2022 extrapolation, we extrapolate using HFCE until 2021 and then using GDP to 2022. This increases the use of the preferred national accounts variable and avoids a kink in the extrapolated welfare vectors between 2021 and 2022. All changes apply both to interpolations and extrapolations. The impact of the new interpolation and extrapolation rule is small globally, but meaningful at the regional level, as explained in more detail in Mahler and Newhouse (2024). 7.1. Interpolation of poverty data for Syria The interpolation of poverty data for Syria over the period 2009-2022 uses the growth rate of nominal GDP (LCU) deflated by the CPI instead of the growth rate of real GDP (which would be deflated using the GDP deflator). Interpolation of poverty estimates based on the standard approach suggests a relatively steady poverty increase over the period 2009-2022. Given the significant changes to Syria’s economy post-conflict (collapse in domestic production, increase in informality, and reliance on imports, including staples), the quality of GDP deflator estimates, which have 2000 as the base year, has progressively deteriorated, affecting the reliability of GDP estimates in constant prices typically used for interpolation (Redaelli et al., forthcoming). Using 29 the growth rate of nominal GDP (LCU) deflated by CPI results in trends that are better aligned with economic activity as for example proxied by nightlight data emissions. 8. Comparability database Since September 2019, we provide metadata on comparability of poverty estimates within countries over time. The assessment of comparability is country-dependent and relies on the accumulation of knowledge from past and current Bank staff in the countries, as well as close dialogue with national data producers with knowledge of survey design and methodology (see (Atamanov et al., 2019) for more information on reasons that break comparability). More information about the comparability database and how to use it is available at https://worldbank.github.io/PIP-Methodology/welfareaggregate.html#comparability. The PIP website also indicates comparability in its main output. 30 9. References Abu-Ismail, K., Abdel-Gadir, A., El-Laithy, H., 2011. Poverty and Inequality in Syria (1997- 2007) (No. 15), Arab Development Challenges REport. UNDP. Amendola, N., Mancini, G., Redaelli, S., Vecchi, G., 2023. Price adjustments and poverty measurement. World Development Indicators. World Bank, Washington, DC. Atamanov, A., Castaneda Aguilar, R.A., Diaz-Bonilla, C., Jolliffe, D., Lakner, C., Mahler, D.G., Montes, J., Moreno Herrera, L.L., Newhouse, D., Nguyen, M.C., Prydz, E.B., Sangraula, P., Tandon, S.A., Yang, J., 2019. September 2019 PovcalNet Update, Global Poverty Monitoring Technical Note 10. Washington, D.C. https://doi.org/10.1596/32478 Atamanov, A., Castaneda Aguilar, R.A., Fujs, T.H., Dewina, R., Diaz-Bonilla, C., Mahler, D.G., Jolliffe, D., Lakner, C., Matytsin, M., Montes, J., 2020. March 2020 PovcalNet Update: What’s New. Global Poverty Monitoring Technical Note 11. Box, G.E.P., Cox, D.R., 1964. An Analysis of Transformations. Journal of the Royal Statistical Society., Series B (Methodological) 26, 211–252. Castaneda, R.A.A., Eilertsen, A., Fujs, T., Lakner, C., Mahler, D.G., Nguyen, M.C., Schoch, M., Tetteh Baah, S.K., Viveros, M., Wu, H., 2022. April 2022 global poverty update from the World Bank. URL https://blogs.worldbank.org/opendata/april-2022-global-poverty- update-world-bank (accessed 5.26.22). Castaneda, R.A.A., Fujs, T., Lakner, C., Mahler, D.G., Nguyen, M.C., Tetteh Baah, S.K., 2020. September 2020 global poverty update from the World Bank: New annual poverty estimates using the revised 2011 PPPs. URL https://blogs.worldbank.org/opendata/ Corral, P., Molina, I., Cojocaru, A., Segovia, S., 2022. Guidelines to Small Area Estimation for Poverty Mapping. World Bank. https://doi.org/10.1596/37728 Datt, G., 1998. Computational tools for poverty measurement and analysis. IFPRI FCND Discussion Paper 50. Deaton, A., 2003. Adjusted Indian Poverty Estimates for 1999-2000. Economic and Political Weekly 38, 322–326. Deaton, A., Zaidi, S., 2002. Guidelines for Constructing Consumption Aggregates for Welfare Analysis, Living Standards Measurement Study Working Paper. World Bank, Washington, DC. Fernandez, J., Olivieri, S., Wambile, A., 2023. Reconstructing poverty and inequality trends in Bangladesh: A survey-to-survey imputation approach. World Bank, Washington, D.C. Johnson, N.L., 1949. Systems of Frequency Curves Generated by Methods of Translation. Biometrika 36, 149–176. https://doi.org/10.2307/2332539 Lakner, C., Mahler, D.G., Nguyen, M.C., Azevedo, J.P., Chen, S., Jolliffe, D., 2018. Consumer price indices used in global poverty measurement. Global Poverty Monitoring Technical Note 8. Mahler, D.G., Castañeda Aguilar, R.A., Newhouse, D., 2022. Nowcasting Global Poverty. The World Bank Economic Review 36, 835–856. https://doi.org/10.1093/wber/lhac017 Mahler, D.G., Newhouse, D., 2024. Changes to the Extrapolation Method for Global Poverty Estimation. Global Poverty Monitoring Technical Note 35. Mancini, G., Vecchi, G., 2022. On the Construction of a Consumption Aggregate for Inequality and Poverty Analysis. World Bank. Prydz, E.B., Jolliffe, D., Serajuddin, U., 2022. Disparities in Assessments of Living Standards Using National Accounts and Household Surveys. Review of Income and Wealth 68, S385–S420. https://doi.org/10.1111/roiw.12577 31 Prydz, E.B., Jolliffe, D.M., Lakner, C., Mahler, D.G., Sangraula, P., 2019. National Accounts Data Used in Global Poverty Measurement. Global Poverty Monitoring Technical Note Series, Global Poverty Monitoring Technical Note Series. Ravallion, M., 2003. Measuring Aggregate Welfare in Developing Countries: How Well Do National Accounts and Surveys Agree? The Review of Economics and Statistics 85, 645– 652. Redaelli, S., Infanzon, M., Moreno Herrera, L., forthcoming. Assessing the extent of monetary poverty in Syria after a decade of conflict. World Bank, Washington, DC. Roy, S.S., van der Weide, R., 2022. Poverty in India Has Declined over the Last Decade But Not As Much As Previously Thought (World Bank Policy Research Working Paper Series No. 9994), no. 9994. World Bank, Washington, DC. Rubin, D.B., 1987. Multiple imputation for nonresponse in surveys. Wiley, New York. Schafer, J.L., 1999. Multiple imputation: a primer. Stat Methods Med Res 8, 3–15. https://doi.org/10.1177/096228029900800102 Schenker, N., Taylor, J.M.G., 1996. Partially parametric techniques for multiple imputation. Computational Statistics & Data Analysis 22, 425–446. https://doi.org/10.1016/0167- 9473(95)00057-7 World Bank, 2024. Poverty and Inequality Platform Methodology Handbook [WWW Document]. URL https://datanalytics.worldbank.org/PIP-Methodology/ (accessed 3.15.24). Yoshida, N., Takamatsu, S., Yoshimura, K., Aron, D.V., Chen, X., Malgioglio, S., Shivakumaran, S., Zhang, K., 2022. The Concept and Empirical Evidence of SWIFT Methodology [WWW Document]. World Bank. URL https://documents.worldbank.org/en/publication/documents- reports/documentdetail/099547109302235758/IDU04a3a086c0a2da04853084b10e85525 3105f9 (accessed 3.13.24). Zambia Statistics Agency, World Bank, 2023. Estimating a consistent Poverty and Inequality trend in Zambia: 2015-2022 Poverty and Inequality Trend Methodological Note. Zhang, K., Takamatsu, S., Yoshida, N., 2024. Correcting Sampling and Nonresponse Bias in Phone Survey Poverty Estimation Using Reweighting and Poverty Projection Models, Policy Research Working Papers. World Bank, Washington, D.C. 32 10. Appendix 10.1. Complete list of new country-years Table A1. Economies-years added in March 2024 PIP update Economy Year Survey Name Argentina 2022 EPHC-S2 Armenia 2022 ILCS Austria 2021 EU-SILC Burundi 2020 EICVMB Belgium 2021 EU-SILC Benin 2021 EHCVM Burkina Faso 2021 EHCVM Bulgaria 2021 EU-SILC Brazil 2022 PNADC-E1 Switzerland 2019 EU-SILC Switzerland 2020 EU-SILC Chile 2022 CASEN Côte d’Ivoire 2021 EHCVM Cameroon 2021 ECAM-V Congo, Dem. Rep. 2020 EGI-ODD Colombia 2022 GEIH Cyprus 2021 EU-SILC Czech Republic 2021 EU-SILC Denmark 2021 EU-SILC Dominican Republic 2022 ECNFT-Q03 Spain 2021 EU-SILC Estonia 2021 EU-SILC Finland 2021 EU-SILC France 2021 EU-SILC United Kingdom 2021 FRS-LIS Guinea-Bissau 2021 EHCVM Greece 2021 EU-SILC Grenada 2018 SLCHB Croatia 2021 EU-SILC Hungary 2021 EU-SILC Indonesia 2023 SUSENAS Ireland 2021 EU-SILC Iran, Islamic Rep. 2011 HEIS Iran, Islamic Rep. 2012 HEIS Iran, Islamic Rep. 2020 HEIS Iran, Islamic Rep. 2021 HEIS Iran, Islamic Rep. 2022 HEIS Israel 2019 HES-LIS 33 Israel 2020 HES-LIS Israel 2021 HES-LIS Italy 1977 SHIW-LIS Italy 1978 SHIW-LIS Italy 1979 SHIW-LIS Italy 1980 SHIW-LIS Italy 1981 SHIW-LIS Italy 1982 SHIW-LIS Italy 1983 SHIW-LIS Italy 1984 SHIW-LIS Italy 2002 SHIW-LIS Italy 2021 EU-SILC Jamaica 2018 JSLC Jamaica 2021 JSLC Kazakhstan 2019 HBS Kazakhstan 2020 HBS Kazakhstan 2021 HBS Kyrgyz Republic 2021 KIHS Lithuania 2021 EU-SILC Luxembourg 2021 EU-SILC Latvia 2021 EU-SILC Mexico 2022 ENIGHNS Mali 2021 EHCVM Montenegro 2019 SILC-C Montenegro 2020 SILC-C Montenegro 2021 SILC-C Mongolia 2022 HSES Mauritania 2019 EPCV Malaysia 2021 HIS Niger 2021 EHCVM Netherlands 2021 EU-SILC Panama 2023 EH Peru 2022 ENAHO Poland 2020 EU-SILC Poland 2021 EU-SILC Portugal 2021 EU-SILC Romania 2021 HBS, EU-SILC Senegal 2021 EHCVM Serbia 2021 EU-SILC Suriname 2022 SSLC Slovak Republic 2020 EU-SILC Slovak Republic 2021 EU-SILC Slovenia 2021 EU-SILC Sweden 2001 HIS-LIS Sweden 2021 EU-SILC 34 Syrian Arab Republic 2007 HIES Syrian Arab Republic 2009 HIES Syrian Arab Republic 2022 HNAP Chad 2022 EHCVM Togo 2021 EHCVM Tonga 2021 HIES Tunisia 2021 NSHBCSL Türkiye 2020 SILC-C Türkiye 2021 SILC-C Taiwan, China 2017 FIDES-LIS Taiwan, China 2018 FIDES-LIS Taiwan, China 2019 FIDES-LIS Taiwan, China 2020 FIDES-LIS Taiwan, China 2021 FIDES-LIS Uruguay 2022 ECH Uzbekistan 2022 HBS Viet Nam 2022 VHLSS Zambia 2022 LCMS-VIII 35 10.2. CPI data sources Table A2 lists the source of CPI used for each economy-year reported in PIP. The columns in the table are defined as follows: • Code: The 3-letter economy code used by the World Bank: https://datahelpdesk.worldbank.org/knowledgebase/articles/906519-world- bankcountryand-lending-groups • Economy: Name of economy • Year(s): Welfare reporting year, i.e., the year for which the welfare has been reported. If the survey collects income for the previous year, it is the year prior to the survey. • CPI period: Common time period to which the welfare aggregates in the survey have been deflated. The letter Y denotes that the CPI period is identical to the year column. When the welfare aggregate has been deflated to a particular month within the welfare reporting year, the month is indicated by a number between 1 and 12, preceded by an M, and similarly with a Q for quarters. The letter W indicates that a weighted CPI is used, as described in equation 1 in (Lakner et al., 2018). • CPI source: Source of the deflator used. The source is given by the abbreviation, the frequency of the CPI, and the vintage; e.g. IFS-M-202311 denotes the monthly IFS database version November 2023. For economy-specific deflators, the description is given in the text or further details are available upon request. 36 Table A2. Source of temporal deflators used in PIP March 2024 update Code Economy Survey Year(s) CPI period Source HBS 2000 W IFS-M-202311 AGO Angola IBEP-MICS 2008 W IFS-M-202311 IDREA 2018 W IFS-M-202311 EWS 1996 Y IFS-M-202311 LSMS 2002-2012 Y IFS-M-202311 ALB Albania HBS 2014-2020 Y IFS-M-202311 SILC-C 2017-2019 (prev. year)Y IFS-M-202311 United Arab HIES 2014 W IFS-M-202311 ARE Emirates 2019 Y IFS-M-202311 EPH 1980-1987 Y NSO 1991-2002 M9 NSO ARG Argentina - urban EPHC-S2 2003-2022 M7-M12 NSO 2007-2014 M7-M12 Private estimates ARM Armenia ILCS ALL Y IFS-M-202311 IHS-LIS 1981 Y IFS-A-202311 IDS-LIS 1985 Y IFS-A-202311 AUS Australia SIHCA-LIS 1989 Y IFS-A-202311 SIH-LIS 1995-2018 Y IFS-A-202311 SIH-HES-LIS 2004-2016 Y IFS-A-202311 ECHP-LIS 1994-2000 Y IFS-M-202311 AUT Austria MC-LIS 1995 Y IFS-M-202311 EU-SILC 2004-2022 (prev. year)Y IFS-M-202311 SLC 1995 Y IFS-M-202311 AZE Azerbaijan HBS 2001-2005 Y IFS-M-202311 EDCM 1992 Y IFS-M-202311 EP 1998 W IFS-M-202311 BDI Burundi QUIBB 2006 Y IFS-M-202311 ECVMB 2013 W IFS-M-202311 EICVMB 2020 W IFS-M-202311 SEP-LIS 1985-1997 Y IFS-M-202311 PSBH-ECHP- BEL Belgium LIS 1995-2000 Y IFS-M-202311 EU-SILC 2004-2022 (prev. year)Y IFS-M-202311 QUIBB 2003 Y IFS-M-202311 EMICOV 2011 W IFS-M-202311 BEN Benin 2015 Y IFS-M-202311 EHCVM 2018 M10 IFS-M-202311 2021 M11 IFS-M-202311 EP-I 1994 W IFS-M-202311 BFA Burkina Faso EP-II 1998 Y IFS-M-202311 ECVM 2003-2009 Y IFS-M-202311 37 EMC 2014 Y IFS-M-202311 EHCVM 2018 M9 IFS-M-202311 2021 M10 IFS-M-202311 HHES 1983-1985 W WEO-A-202310 1988-1991 W IFS-A-202311 BGD Bangladesh 1995 W Survey HIES 2000-2022 Y Survey HBS 1989 Y IFS-A-202311 1992-1994 Y IFS-M-202311 BGR Bulgaria IHS 1995-2001 Y IFS-M-202311 MTHS 2003-2007 Y IFS-M-202311 EU-SILC 2007-2022 (prev. year)Y IFS-M-202311 Bosnia and LSMS 2001-2004 Y WEO-A-202310 BIH Herzegovina HBS 2007-2011 Y IFS-M-202311 FBS 1993-1995 Y IFS-M-202311 BLR Belarus HHS 1998-2020 Y IFS-M-202311 LFS 1993-1999 Y IFS-A-202311 BLZ Belize HBS 1995 Y IFS-A-202311 SLC 1996 Y IFS-A-202311 Bolivia - urban EPF 1990 W IFS-M-202311 EIH 1992 M11 IFS-M-202311 Bolivia ENE 1997 M11 IFS-M-202311 ECH 1999 M10 IFS-M-202311 BOL 2000 M11 IFS-M-202311 EH 2001-2005 M11 IFS-M-202311 ECH 2004 M10 IFS-M-202311 EH 2006-2016 M10 IFS-M-202311 2017-2021 M11 IFS-M-202311 PNAD 1981-2011 M9 IFS-M-202311 BRA Brazil PNADC-E1 2012-2022 Y IFS-M-202311 PNADC-E5 2020-2021 Y IFS-M-202311 BLSS 2003-2017 Y Previous WDI/IFS BTN Bhutan 2022 M1-M8 Previous WDI/IFS HIES 1985-2002 W IFS-M-202311 BWA Botswana CWIS 2009 W IFS-M-202311 BMTHS 2015 W IFS-M-202311 EPCM 1992 W IFS-M-202311 Central African CAF ECASEB 2008 Y IFS-M-202311 Republic EHCVM 2021 M5 IFS-M-202311 SCF-LIS 1971-1995 Y IFS-M-202311 CAN Canada SLID-LIS 1996-2011 Y IFS-M-202311 CIS-LIS 2012-2019 Y IFS-M-202311 CHE Switzerland SIWS-LIS 1982 Y IFS-M-202311 38 NPS-LIS 1992 Y IFS-M-202311 IES-LIS 2000-2002 Y IFS-M-202311 EU-SILC 2007-2021 (prev. year)Y IFS-M-202311 CASEN 1987 Y IFS-M-202311 CHL Chile 1990-2022 M11 IFS-M-202311 CRHS-CUHS 1981-2011 Y NSO CHN China CNIHS 2012-2020 Y NSO EPAM 1985-1988 W IFS-M-202311 EP 1992 W IFS-M-202311 CIV Côte d'Ivoire ENV 1995-2015 Y IFS-M-202311 EHCVM 2018 M10 IFS-M-202311 2021 M11 IFS-M-202311 ECAM-I 1996 Y IFS-M-202311 ECAM-II 2001 Y IFS-M-202311 CMR Cameroon ECAM-III 2007 Y IFS-M-202311 ECAM-IV 2014 Y IFS-M-202311 ECAM-V 2021 M10 IFS-M-202311 E123 2004-2012 W IFS-M-202311 COD Congo, Dem. Rep. EGI-ODD 2020 Y WEO-A-202310 ECOM 2005 Y IFS-M-202311 COG Congo, Rep. 2011 W IFS-M-202311 Colombia - urban ENH 1980-1988 Y IFS-M-202311 1989-1991 M11 IFS-M-202311 COL Colombia 1992-2000 M11 IFS-M-202311 ECH 2001-2005 M11 IFS-M-202311 GEIH 2008-2022 M11 IFS-M-202311 EIM 2004 Y IFS-M-202311 COM Comoros EESIC 2013 Y IFS-M-202311 IDRF 2001 W IFS-M-202311 CPV Cabo Verde QUIBB 2007 W IFS-M-202311 IDRF 2015 Y IFS-M-202311 ENH 1981-1986 Y IFS-M-202311 EHPM 1989 Y IFS-M-202311 CRI Costa Rica 1990-2009 M7 IFS-M-202311 ENAHO 2010-2022 M7 IFS-M-202311 CYP Cyprus EU-SILC ALL (prev. year)Y IFS-M-202311 MC-LIS 1992-2002 Y IFS-M-202311 CZE Czech Republic CM 1993 Y IFS-M-202311 EU-SILC 2005-2022 (prev. year)Y IFS-M-202311 DEU Germany LIS ALL Y IFS-M-202311 EDAM 2002-2013 Y IFS-M-202311 DJI Djibouti 2017 M5 IFS-M-202311 DNK Denmark LM-LIS 1987-2000 Y IFS-M-202311 39 EU-SILC 2004-2022 (prev. year)Y IFS-M-202311 ENGSLF 1986-1989 Y IFS-M-202311 ICS 1992 M6 IFS-M-202311 Dominican ENFT 1996 M2 IFS-M-202311 DOM Republic 1997 M4 IFS-M-202311 2000-2016 M9 IFS-M-202311 ECNFT-Q03 2017-2022 Y IFS-M-202311 EDCM 1988 Y IFS-M-202311 DZA Algeria ENMNV 1995 Y IFS-M-202311 ENCNVM 2011 W IFS-M-202311 Ecuador - urban EPED 1987 Y IFS-M-202311 Ecuador ECV 1994 M6-M10 IFS-M-202311 Ecuador - urban EPED 1995 M11 IFS-M-202311 ECU 1998 M6 IFS-M-202311 (prev. Ecuador ECV 1999 year)M10-M9 IFS-M-202311 EPED 2000 M11 IFS-M-202311 ENEMDU 2003-2022 M11 IFS-M-202311 HIECS 1990-2012 W IFS-M-202311 EGY Egypt, Arab Rep. 2015 Y IFS-M-202311 2017-2019 W IFS-M-202311 HBS-LIS 1980-1990 Y IFS-M-202311 ESP Spain ECHP-LIS 1993-2000 Y IFS-M-202311 EU-SILC 2004-2022 (prev. year)Y IFS-M-202311 HIES 1993-1998 Y IFS-M-202311 EST Estonia HBS 2000-2004 Y IFS-M-202311 EU-SILC 2004-2022 (prev. year)Y IFS-M-202311 Ethiopia - rural HICES 1981 W IFS-M-202311 ETH Ethiopia 1995-2010 W IFS-M-202311 2015 M12 IFS-M-202311 IDS-LIS 1987-2000 Y IFS-M-202311 FIN Finland EU-SILC 2004-2022 (prev. year)Y IFS-M-202311 FJI Fiji HIES ALL W IFS-M-202311 TIS-LIS 1970-1990 Y IFS-M-202311 FRA France TSIS-LIS 1996-2002 Y IFS-M-202311 EU-SILC 2004-2022 (prev. year)Y IFS-M-202311 Micronesia, Fed. Sts. - urban CPH 2000 Y IFS-A-202311 FSM Micronesia, Fed. Sts. HIES 2005-2013 Y IFS-A-202311 GAB Gabon EGEP ALL Y IFS-M-202311 FES-LIS 1968-1993 Y IFS-M-202311 GBR United Kingdom FRS-LIS 1994-2021 Y IFS-M-202311 GEO Georgia HIS ALL Y IFS-M-202311 40 GLSS-I 1987 W IFS-M-202311 GLSS-II 1988 W IFS-M-202311 GLSS-III 1991 W IFS-M-202311 GHA Ghana GLSS-IV 1998 W IFS-M-202311 GLSS-V 2005 W Survey GLSS-VI 2012 W Survey GLSS-VII 2016 W Survey ESIP 1991 Y WEO-A-202310 EIBC 1994 W WEO-A-202310 GIN Guinea EIBEP 2002 W WEO-A-202310 ELEP 2007-2012 Y IFS-M-202311 EHCVM 2018 W IFS-M-202311 HPS 1998 Y IFS-M-202311 GMB Gambia, The HIS 2003 W IFS-M-202311 IHS 2010-2020 W IFS-M-202311 ILJF 1991 Y IFS-M-202311 ICOF 1993 Y IFS-M-202311 ILAP-I 2002 Y IFS-M-202311 GNB Guinea-Bissau ILAP-II 2010 Y IFS-M-202311 EHCVM 2018 W IFS-M-202311 2021 M11 IFS-M-202311 ECHP-LIS 1995-2000 Y IFS-M-202311 GRC Greece EU-SILC 2004-2022 (prev. year)Y IFS-M-202311 GRD Grenada SLCHB 2018 M4 IFS-M-202311 ENSD 1986 W IFS-M-202311 1989 Y IFS-M-202311 GTM Guatemala ENIGF 1998 M8 IFS-M-202311 ENCOVI 2000 M6-M11 IFS-M-202311 2006-2014 M7 IFS-M-202311 GLSMS 1992 W WEO-A-202310 GUY Guyana 1998 Y IFS-M-202311 Honduras - urban ECSFT 1986 Y IFS-M-202311 Honduras EPHPM 1989 Y IFS-M-202311 HND 1990-1993 M5 IFS-M-202311 1994 M9 IFS-M-202311 1995-2019 M5 IFS-M-202311 HBS 1988-2010 Y IFS-M-202311 HRV Croatia EU-SILC 2010-2022 (prev. year)Y IFS-M-202311 ECVH 2001 M5 IFS-M-202311 HTI Haiti ECVMAS 2012 M10 IFS-M-202311 HBS 1987-2007 Y IFS-M-202311 HUN Hungary HHP-LIS 1991-1994 Y IFS-M-202311 THMS-LIS 1999 Y IFS-M-202311 41 EU-SILC 2005-2022 (prev. year)Y IFS-M-202311 SUSENAS 1984-1999 Y IFS-M-202311 IDN Indonesia 2000-2007 M2 IFS-M-202311 2008-2023 M3 IFS-M-202311 M7-(next NSS 1977 year)M6 NSO 1983 Y NSO IND India M7-(next NSS-SCH1 1987-2011 year)M6 NSO M4-(next CPHS 2015-2021 year)M3 NSO SIDPUSS-LIS 1987 Y IFS-M-202311 LIS-ECHP-LIS 1994-2000 Y IFS-M-202311 IRL Ireland SILC-LIS 2002 Y IFS-M-202311 EU-SILC 2004-2022 (prev. year)Y IFS-M-202311 SECH 1986 Y IFS-A-202311 1990-1998 Y IFS-M-202311 HEIS 2005-2009 W IFS-M-202311 IRN Iran, Islamic Rep. M4-(next 2011-2021 year)M3 IFS-M-202311 M4-(next 2022 year)M3 WEO-A-202310 IHSES 2006 W COSIT IRQ Iraq 2012 Y COSIT ISL Iceland EU-SILC ALL (prev. year)Y IFS-M-202311 ISR Israel HES-LIS ALL Y IFS-M-202311 SHIW-LIS 1977-2002 Y IFS-M-202311 ITA Italy EU-SILC 2004-2022 (prev. year)Y IFS-M-202311 SLC 1988 M9 IFS-M-202311 M11-(next 1990-1993 year)M3 IFS-M-202311 JAM Jamaica 1996 M5-M8 IFS-M-202311 1999 M6-M8 IFS-M-202311 2002-2004 M6 IFS-M-202311 JSLC 2018-2021 Y IFS-M-202311 HEIS 1986 W IFS-M-202311 JOR Jordan 1992-1997 Y IFS-M-202311 2002-2010 W IFS-M-202311 JPN Japan JHPS-LIS ALL Y IFS-M-202311 HBS 1993-2021 Y IFS-M-202311 KAZ Kazakhstan LSMS 1996 Y IFS-M-202311 WMS-I 1992 Y NSO WMS-II 1994 Y NSO KEN Kenya WMS-III 1997 Y NSO IHBS 2005-2015 W NSO 42 KCHS 2020 M6 NSO KGZ Kyrgyz Republic KPMS 1998 Y IFS-M-202311 HBS 2000-2003 Y IFS-M-202311 KIHS 2004-2021 Y IFS-M-202311 HIES 2006 Y IFS-M-202311 KIR Kiribati 2019 W IFS-M-202311 HIES-FHES- KOR Korea, Rep. LIS ALL Y IFS-M-202311 LECS 1992 W IFS-A-202311 LAO Lao PDR 1997 W IFS-M-202311 2002-2018 W Survey LBN Lebanon HBS 2011 (next year)M5 IFS-M-202311 CWIQ 2007 Y IFS-M-202311 LBR Liberia HIES 2014-2016 Y IFS-M-202311 St. Lucia LSMS 1995 Y IFS-M-202311 LCA SLCHBS 2015 M11 IFS-M-202311 LFSS 1985 Y IFS-M-202311 HIES 1990 W IFS-M-202311 SES 1995 W IFS-M-202311 LKA Sri Lanka HIES 2002 Y IFS-M-202311 2006-2012 W IFS-M-202311 2016-2019 Y IFS-M-202311 HBS 1986 W WEO-A-202310 NHECS 1994 W WEO-A-202310 LSO Lesotho HBS 2002 W IFS-M-202311 CMSHBS 2017 M8 IFS-M-202311 HBS 1993-2008 Y IFS-M-202311 LTU Lithuania EU-SILC 2005-2022 (prev. year)Y IFS-M-202311 PSELL-LIS 1985-1993 Y IFS-M-202311 PSELL-ECHP- LUX Luxembourg LIS 1994-2001 Y IFS-M-202311 SEP-SILC-LIS 2002 Y IFS-M-202311 EU-SILC 2004-2022 (prev. year)Y IFS-M-202311 HBS 1993-2009 Y IFS-M-202311 LVA Latvia EU-SILC 2005-2022 (prev. year)Y IFS-M-202311 ECDM 1984 W IFS-M-202311 MAR Morocco ENNVM 1990-2006 W IFS-M-202311 ENCDM 2000-2013 W IFS-M-202311 MDA Moldova HBS ALL Y IFS-M-202311 EB 1980 Y IFS-M-202311 MDG Madagascar EPM 1993 W IFS-M-202311 43 1997-2010 Y IFS-M-202311 ENSOMD 2012 W IFS-M-202311 HIES 2002-2009 W IFS-M-202311 MDV Maldives 2016 Y IFS-M-202311 2019 M11 IFS-M-202311 ENIGH 1984-2014 M8 IFS-M-202311 MEX Mexico ENIGHNS 2016-2022 M8 IFS-M-202311 MHL Marshall Islands HIES 2019 W WEO-A-202310 HBS 1998-2008 Y IFS-M-202311 MKD North Macedonia SILC-C 2010-2020 (prev. year)Y IFS-M-202311 EMCES 1994 Y IFS-A-202311 EMEP 2001 W IFS-M-202311 MLI Mali ELIM 2006-2009 W IFS-M-202311 EHCVM 2018-2021 M10 IFS-M-202311 MLT Malta EU-SILC ALL (prev. year)Y IFS-M-202311 MPLCS 2015 M1 IFS-M-202311 MMR Myanmar MLCS 2017 Q1 IFS-M-202311 HBS 2005-2014 Y IFS-M-202311 MNE Montenegro SILC-C 2013-2022 (prev. year)Y IFS-M-202311 LSMS 1995-1998 Y IFS-M-202311 HIES-LSMS 2002 W IFS-M-202311 MNG Mongolia HSES 2007 W IFS-M-202311 2010-2022 Y IFS-M-202311 NHS 1996 W WEO-A-202310 MOZ Mozambique IAF 2002 W WEO-A-202310 IOF 2008-2019 W IFS-M-202311 EPCV 1987 Y IFS-M-202311 EP 1993 Y IFS-M-202311 MRT Mauritania EPCV 1995-2008 W IFS-M-202311 2014 Y IFS-M-202311 2019 M11 IFS-M-202311 HBS 2006 W IFS-M-202311 MUS Mauritius 2012-2017 Y IFS-M-202311 IHS-I 1997 W IFS-M-202311 IHS-II 2004 W Survey MWI Malawi IHS-III 2010 W Survey IHS-IV 2016 M4 Survey IHS-V 2019 M4 Survey HIS 1984-1997 Y IFS-M-202311 (prev. MYS Malaysia year)M7- (prev. 2004 year)M12 IFS-M-202311 44 (prev. year)M7- (prev. 2007 year)M10 IFS-M-202311 2009 W IFS-M-202311 2012-2016 Y IFS-M-202311 HIESBA 2019 W IFS-M-202311 HIS 2022 W IFS-M-202311 NHIES 1993 W WEO-A-202310 NAM Namibia 2003-2015 W IFS-M-202311 ENBCM 1992-2007 W IFS-M-202311 EPCES 1994 W IFS-M-202311 ENCVM 2005 Y IFS-M-202311 NER Niger ECVMA 2011-2014 Y IFS-M-202311 EHCVM 2018 M10 IFS-M-202311 2021 M11 IFS-M-202311 NCS 1985 W IFS-M-202311 1992-1996 Y IFS-M-202311 LSS 2003 W IFS-M-202311 GHSP-W1 2010 M3-M4 IFS-M-202311 NGA Nigeria GHSP-W2 2012 M3-M4 IFS-M-202311 GHSP-W3 2015 M3-M4 IFS-M-202311 (next year)M3-(next LSS 2018 year)M4 IFS-M-202311 EMNV 1993 M2 NSO 1998 M6 NSO NIC Nicaragua 2001 M6 IFS-M-202311 2005-2009 M8 IFS-M-202311 2014 M8-M10 IFS-M-202311 AVO-LIS 1983-1990 Y IFS-M-202311 NLD Netherlands SEP-LIS 1993-1999 Y IFS-M-202311 EU-SILC 2005-2022 (prev. year)Y IFS-M-202311 IDS-LIS 1979-2000 Y IFS-M-202311 NOR Norway EU-SILC 2004-2020 (prev. year)Y IFS-M-202311 MHBS 1984 W IFS-M-202311 LSS-I 1995 W IFS-M-202311 NPL Nepal LSS-II 2003 W IFS-M-202311 LSS-III 2010 W IFS-M-202311 NRU Nauru HIES 2012 W IFS-M-202311 HIES 1987 Y IFS-M-202311 1990-1998 W IFS-M-202311 PAK Pakistan IHS 1996 W IFS-M-202311 PIHS 2001 M6 IFS-M-202311 45 HIES 2004-2018 (next year)M1 IFS-M-202311 EMO 1979-1989 Y IFS-M-202311 PAN Panama 1991 M7 IFS-M-202311 EH 1995-2023 M7 IFS-M-202311 ENNIV 1985 W IFS-M-202311 1994 Y IFS-M-202311 PER Peru ENAHO 1997-2002 Q4 IFS-M-202311 2003 M5-M12 IFS-M-202311 2004-2022 Y IFS-M-202311 PHL Philippines FIES ALL Y IFS-M-202311 HIES 1996 Y IFS-A-202311 PNG Papua New Guinea 2009 W IFS-A-202311 HBS 1985-1987 Y IFS-A-202311 HBS-LIS 1986 Y IFS-A-202311 POL Poland HBS 1989-2019 Y IFS-M-202311 HBS-LIS 1992-1999 Y IFS-M-202311 EU-SILC 2005-2022 (prev. year)Y IFS-M-202311 PRT Portugal EU-SILC ALL (prev. year)Y IFS-M-202311 EH 1990 M7 IFS-M-202311 1995 M8-M11 IFS-M-202311 EIH 1997 (next year)M2 IFS-M-202311 EPH 1999 M9 IFS-M-202311 EIH 2001 M3 IFS-M-202311 EPH 2002 M11 IFS-M-202311 PRY Paraguay 2003 M9 IFS-M-202311 2004 M10 IFS-M-202311 2005 M11 IFS-M-202311 2006 M12 IFS-M-202311 2007-2008 M10 IFS-M-202311 2009 M11 IFS-M-202311 2010-2022 M10 IFS-M-202311 West Bank and PECS 2004-2011 Y IFS-M-202311 PSE Gaza 2016 W IFS-M-202311 HBS 1989 Y Milanovic (1998) MC 1992 Y IFS-M-202311 HIS 1994-1999 Y IFS-M-202311 ROU Romania IHS-LIS 1995-1997 Y IFS-M-202311 IHS 1998-2000 Y IFS-M-202311 HBS 2001-2021 Y IFS-M-202311 EU-SILC 2007-2022 (prev. year)Y IFS-M-202311 HBS 1993-2020 Y IFS-M-202311 RUS Russian Federation VNDN 2015-2019 (prev. year)Y IFS-M-202311 RWA Rwanda - rural ENBCM 1984 W IFS-M-202311 46 Rwanda EICV-I 2000 W IFS-M-202311 EICV-II 2005 W IFS-M-202311 EICV-III 2010 (next year)M1 IFS-M-202311 EICV-IV 2013 (next year)M1 IFS-M-202311 EICV-V 2016 (next year)M1 IFS-M-202311 NBHS 2009 Y IFS-M-202311 SDN Sudan 2014 M11 IFS-M-202311 EP 1991 W IFS-M-202311 ESAM 1994 W IFS-M-202311 ESAM-II 2001 W IFS-M-202311 SEN Senegal ESPS-I 2005 W IFS-M-202311 ESPS-II 2011 W IFS-M-202311 EHCVM 2018 M9 IFS-M-202311 2021 M11 IFS-M-202311 SLB Solomon Islands HIES ALL W IFS-M-202311 HEEAS 1989 W WEO-A-202310 SLE Sierra Leone SLIHS 2003 W WEO-A-202310 2011-2018 Y IFS-M-202311 EHPM 1989 Y IFS-M-202311 M10-(next 1991 year)M4 IFS-M-202311 SLV El Salvador 1995-1999 Y IFS-M-202311 2000-2007 M12 IFS-M-202311 2008-2022 M11 IFS-M-202311 LSMS 2002 Y IFS-M-202311 SRB Serbia HBS 2003-2019 Y IFS-M-202311 EU-SILC 2013-2022 (prev. year)Y IFS-M-202311 NBHS 2009 Y IFS-M-202311 SSD South Sudan (prev. HFS-W3 2016 year)M7 IFS-M-202311 São Tomé and IOF 2000 W IFS-M-202311 STP Principe 2010-2017 Y IFS-M-202311 SUR Suriname - urban EHS 1999 Y IFS-M-202311 Suriname SSLC 2022 Y IFS-M-202311 MC-LIS 1992-1996 Y IFS-M-202311 SVK Slovak Republic HBS 2004-2009 Y IFS-M-202311 EU-SILC 2005-2022 (prev. year)Y IFS-M-202311 IES 1987-1993 Y IFS-M-202311 HBS-LIS 1997-1999 Y IFS-M-202311 SVN Slovenia HBS 1998-2003 Y IFS-M-202311 EU-SILC 2005-2022 (prev. year)Y IFS-M-202311 HIS-LIS 1975-2002 Y IFS-M-202311 SWE Sweden EU-SILC 2004-2022 (prev. year)Y IFS-M-202311 47 SWZ Eswatini HIES ALL W IFS-M-202311 HES 1999 W IFS-M-202311 HBS 2006 W IFS-M-202311 SYC Seychelles 2013 Y IFS-M-202311 2018 W IFS-M-202311 HIES 1996-2007 W IFS-M-202111 Syrian Arab SYR 2009 Y IFS-M-202111 Republic HNAP 2022 Y IFS/IMF/Economist/EIU ECOSIT-II 2003 Y IFS-M-202311 ECOSIT-III 2011 Y IFS-M-202311 TCD Chad EHCVM 2018 W IFS-M-202311 2022 M2 IFS-M-202311 QUIBB 2006-2015 Y IFS-M-202311 TGO Togo EHCVM 2018-2021 M10 IFS-M-202311 THA Thailand SES ALL Y IFS-M-202311 TLSS 1999 Y WEO-A-202310 2003-2007 Y Survey TJK Tajikistan HBS 2004 Y Survey TLSS 2009 Y IFS-M-202311 HSITAFIEN 2015 Y IFS-M-202311 TKM Turkmenistan LSMS 1998 Y WEO-A-202310 TLSS 2001 Y WEO-A-202310 TLS Timor-Leste TLSLS 2007-2014 Y IFS-M-202311 HIES 2000 W IFS-M-202311 TON Tonga 2009-2021 Y IFS-M-202311 Trinidad and SLC 1988 Y IFS-M-202311 TTO Tobago PHC 1992 Y IFS-M-202311 HBCS 1985 Y IFS-A-202311 1990 Y IFS-M-202311 TUN Tunisia LSS 1995-2000 Y IFS-M-202311 NSHBCSL 2005-2021 W IFS-M-202311 HICES 1987-2019 Y IFS-M-202311 TUR Turkey SILC-C 2018-2022 (prev. year)Y IFS-M-202311 TUV Tuvalu HIES 2010 Y IFS-A-202311 TWN Taiwan, China FIDES-LIS ALL Y WEO-A-202310 HBS 1991 W IFS-A-202311 2000 W IFS-M-202311 TZA Tanzania 2007 Y IFS-M-202311 2011-2018 W IFS-M-202311 HBS 1989 Y WEO-A-202310 NIHS 1992 W WEO-A-202310 UGA Uganda 1996-1999 W IFS-M-202311 UNHS 2002-2019 W IFS-M-202311 48 HS 1992-1993 Y IFS-M-202311 UKR Ukraine HIES 1995-1996 Y IFS-M-202311 HLCS 1999-2020 Y IFS-M-202311 Uruguay - urban ENH 1981-1989 Y IFS-M-202311 (prev. ECH 1992-2005 year)M12 IFS-M-202311 URY (prev. Uruguay 2006-2022 year)M12 IFS-M-202311 (prev. ECH-S2 2021 year)M12 IFS-M-202311 CPS-LIS 1963-2001 Y IFS-M-202311 USA United States CPS-ASEC-LIS 2002-2021 Y IFS-M-202311 HBS 1998-2003 Y WEO-A-202310 UZB Uzbekistan 2022 Y IFS-M-202311 EHM 1981-1989 Y NSO VEN Venezuela, RB 1992-2006 M12 NSO VLSS 1992 W WEO-A-202310 VNM Vietnam 1997 W IFS-M-202311 VHLSS 2002-2022 M1 IFS-M-202311 HIES 2010 Y IFS-A-202311 VUT Vanuatu NSDP 2019 W IFS-A-202311 HIES 2002-2008 Y IFS-M-202311 WSM Samoa 2013 W IFS-M-202311 XKX Kosovo HBS ALL Y IFS-M-202311 HBS 1998 Y IFS-M-202311 YEM Yemen, Rep. 2005 W IFS-M-202311 2014 Y IFS-M-202311 KIDS 1993 Y IFS-M-202311 HIES 2000 W IFS-M-202311 ZAF South Africa IES 2005-2010 (next year)M6 IFS-M-202311 LCS 2008 W IFS-M-202311 2014 (next year)M6 IFS-M-202311 HBS 1991-1993 Y IFS-M-202311 LCMS-I 1996 Y IFS-M-202311 LCMS-II 1998 Y IFS-M-202311 LCMS-III 2002 W IFS-M-202311 ZMB Zambia LCMS-IV 2004 W IFS-M-202311 LCMS-V 2006 W IFS-M-202311 LCMS-VI 2010 Y IFS-M-202311 LCMS-VII 2015 Y IFS-M-202311 LCMS-VIII 2022 Y IFS-M-202311 ICES 2011 Y IFS-M-202311 ZWE Zimbabwe PICES 2017-2019 Y Survey 49