1 /27 P O V E R T Y A N D I N E Q UA L I T Y Regional Poverty and Inequality Update Latin America and the Caribbean October 20251 Poverty and Equity Global Practice Main Poverty in Latin America and the Caribbean (LAC) is projected to fall to 25.2 percent in 2025, Messages with Brazil and Mexico contributing to most of this decline. The middle class is projected to reach 42.8 percent by the end of 2025, which would be its highest level in history. Vulnerability to poverty has maintained a steady level at 32 percent, with the vulnerable population in the Caribbean region being larger (36 percent) than in the rest of LAC. These trends have been driven by increased employment and labor income growth across most countries, followed by contributions from public transfers. Despite recent progress, LAC countries lag behind their counterparts in more dynamic upper- middle-income regions in long-term poverty reduction and economic growth. Monetary poverty is persistent in LAC, with most poor households (between 58 and 76 percent) remaining poor from year to year, with higher rates in poorer countries. Despite persistent structural weaknesses in the labor market, including a declining education premium that constrains wage growth, job creation and shifts toward higher-skilled occupations have helped households raise their incomes over time. Looking ahead, global economic uncertainty and domestic challenges point to weaker growth and slower poverty reduction through 2027. Addressing these challenges will require policies to unlock job creation in strategic sectors through structural reforms in capital markets, infrastructure, and institutions, and to expand opportunities for workers to transition into higher-skilled occupations through higher educational attainment, strengthened skills systems, and reduced barriers to job transitions. 1 This brief summarizes recent facts related to poverty and inequality in Latin America and the Caribbean (LAC) using the new wave of harmonized household surveys from the Socio-Economic Database for LAC (SEDLAC). This brief was produced by the Poverty Global Practice in the LAC Region of the World Bank. The core team included Karen Barreto, Luis Eduardo Castellanos Rodríguez, and Catalina García García under the leadership of Diana Sanchez Castro and Hernán Winkler and the guidance of Carlos Rodríguez Castelan. Ana Carolina Leguizamo provided administrative assistance. The team thanks José Andrée Camarena, Gustavo Canavire, Otavio Canozzi, Jacobus De Hoop, Jonathan Lain, Gaston Marinelli, Hugo Ñopo, Anna Luisa Paffhausen, Lourdes Rodríguez, Yuri Yamashita, Guillermo Vuletin, and the staff from the LAC Poverty team for valuable inputs and comments. Contact: lac_stats@worldbank.org. Most of the data featured in this brief can be found at the LAC Equity Lab. P O V E R T Y A N D I N E Q UA L I T Y 2 /27 1. Regional macroeconomic and poverty trends Gross domestic product (GDP) in Latin America and the The regional outlook for 2025 remains uncertain, as LAC Caribbean (LAC) grew by 2.2 percent in 2024, maintaining countries face persistent inflationary pressures, largely a pace similar to the previous year but remaining below driven by the services sector. A slower pace of interest rate that of most other regions. Most economies in LAC 2 cuts is prolonging financial stress for households and firms, expanded between 1.4 (Mexico) and 5.0 (Dominican while elevated budget deficits and debt service obligations Republic) percent. Argentina and Ecuador experienced are constraining fiscal space for investing in social contractions of 1.3 and 2.0 percent, respectively. infrastructure and expanding social protection programs. Trade policy uncertainty continues to disrupt supply chains During 2024, the poverty rate at the upper-middle- and increase import costs, while foreign direct investment income poverty line of $8.30 per day (2021 PPP) declined inflows have declined sharply across the region, particularly by 2.4 percentage points (p.p.) to 25.5 percent of the in Southern Cone economies. As a result, economic growth region’s population, its lowest level recorded.3 Positive is expected to remain below global averages during the labor market outcomes across most countries and 2025–2027 period, and the pace of poverty reduction is increased public transfers in Brazil explain this larger- expected to slow down significantly to just 0.3 p.p. per year, than-expected reduction in poverty.4 resulting in a poverty level of 25.2 percent in 2025 (figure 1). Figure 1 Poverty Reduction and GDP Growth in Latin America, 2016–27 10.0 10.0 8.0 7.3 8.0 Percentage Points (Poverty Reduction) 6.0 6.0 4.1 4.0 4.0 Percentage (GDP Growth) 2.5 2.7 2.3 2.2 2.3 2.0 3.9 1.6 2.0 2.0 0.7 2.4 0.0 0.0 1.1 0.0 0.7 0.5 0.4 0.3 0.3 0.3 -0.4 -2.0 -0.4 -1.3 -2.0 -4.0 -4.0 -6.0 -6.0 -6.6 -8.0 -8.0 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025e 2026f 2027f Poverty Reduction - Latin America & Caribbean GDP - Latin America & Caribbean Sources: SEDLAC (CEDLAS and World Bank), available at LAC Equity LAB and World Bank’s Macro Poverty Outlook (Annual Meetings 2025 edition). Note: LAC poverty data based on 18 countries: Argentina, Bolivia, Brazil, Chile, Colombia, Costa Rica, Dominican Republic, Ecuador, El Salvador, Guatemala, Haiti, Honduras, Mexico, Nicaragua, Panama, Paraguay, Peru, and Uruguay. In cases where data were unavailable, values have been estimated using microsimulations and then pooled to create regional estimates. Poverty reduction is calculated as the percentage change in poverty rates multiplied by -1 using the international poverty line of $8.30 per day, 2021 PPP. Positive values indicate poverty reduction. e = estimate; f = forecast 2 The macroeconomic analysis is based on Maloney et al. (2025). 3 The upper-middle-income poverty line was updated in June 2025 using the new 2021 PPPs; see appendix B for more details. 4 In October 2024, the poverty decline between 2022 and 2024 was initially forecast at 1.3 percentage points, significantly below the actual decline of 3.5 percentage points observed during that period, as documented here. P O V E R T Y A N D I N E Q UA L I T Y 3 /27 2. Poverty, Vulnerability, and the Middle Class LAC maintains the second-lowest regional poverty just 1.6 percent annually, significantly below the 2.7 rate globally at the poverty line of $8.30 per day, with percent global average. Every other region outperformed only Europe and Central Asia performing better, at 9.6 LAC, ranging from Sub-Saharan Africa with a rate of 2.0 percent in 2024. However, LAC’s poverty reduction has percent to South Asia and EAP, which saw more robust slowed down over the past fifteen years, particularly growth rates of 5.1 and 5.3 percent, respectively. when compared to East Asia and the Pacific (EAP) (figure Recent trends show a significant improvement for LAC. 2). In 2010, EAP’s poverty rate was nearly 30 p.p. higher Since 2022, poverty has declined faster in LAC than in than LAC’s (67.5 percent versus 39.6 percent). By 2022, most other regions, reaching its lowest point this century. the poverty levels of the two regions had converged, At the poverty line of $8.30 per day, poverty dropped 2.4 highlighting LAC’s sluggish progress. p.p. between 2023 and 2024 to 25.5 percent, the fastest This divergence reflects differences in economic growth decline globally. Our nowcasting model indicates a slight trajectories. Between 2016 and 2024, LAC experienced decrease to 25.2 percent in 2025.5 the weakest economic expansion globally, averaging Figure 2 Poverty Rate, $8.30 per Day (2021 PPP), by World Region for Selected Years, 2016–25 100.0 90.0 80.0 70.0 Percentage of Population 60.0 50.0 40.0 30.0 20.0 10.0 88.3 78.1 49.0 46.3 28.9 25.5 9.6 0.0 Sub-Saharan South Asia Middle East & World East Asia & LAC Europe & Africa North Africa Pacific Central Asia 2024 2016 2022 2025e Sources: SEDLAC (CEDLAS and World Bank) and World Bank Poverty and Inequality Platform (PIP). Note: LAC aggregate based on 18 countries with available SEDLAC microdata. In cases where data were unavailable, values have been estimated using microsimulations and then pooled to create regional estimates. For non-LAC regions, 2024–2025 values are estimated by the PIP nowcasting model. e = estimate. Poverty reduction between 2022 and 2024 was common 5.7 p.p., respectively (table 1). Other notable reductions across LAC countries, though concentrated in the region’s were seen in the Dominican Republic (-6.2 p.p.), Paraguay largest economies. Brazil and Mexico drove most of the (-5.3 p.p.), Costa Rica (-4.5 p.p.), and Colombia (-3.2 p.p.). decline, with poverty falling in those countries by 4.7 and 5 The nowcasting model follows Montoya, Olivieri, and Braga (2023). 4 /27 Table 1 Poverty Changes in LAC and Subregions, 2022–24 International Lower-Middle-Income Upper-Middle-Income Poverty Rate Poverty Rate Poverty Rate Country/Subregion $3.00 per day $4.20 per day $8.30 per day Change Change Change 2022 2024 2022 2024 2022 2024 (p.p.) (p.p.) (p.p.) Brazil 4.9 3.0 -1.9 8.5 6.2 -2.3 25.3 20.6 -4.7 Mexico 2.3 1.7 -0.6 5.7 4.2 -1.5 27.4 21.7 -5.7 Andean Subregion 6.6 6.8 0.2 12.2 12.1 -0.1 35.8 34.0 -1.8 Central America Subregion 8.0 7.3 -0.7 14.1 12.6 -1.5 36.8 34.3 -2.5 Southern Cone Subregion 1.2 0.9 -0.3 2.6 2.2 -0.4 12.1 12.0 -0.1 LAC 5.8 4.9 -0.9 9.9 8.6 -1.3 29.0 25.5 -3.5 Source: SEDLAC (CEDLAS and World Bank). Note: LAC aggregate based on 18 countries with available SEDLAC microdata. In cases where data were unavailable, values have been estimated using microsimulations and then pooled to create regional estimates. Andean Subregion: Bolivia, Colombia, Ecuador, and Peru; Central America Subregion: Costa Rica, Guatemala, Honduras, Nicaragua, Panama, El Salvador, and Dominican Republic. Southern Cone Subregion: Argentina, Chile, Paraguay, and Uruguay. Brazil and Mexico 2024 data are preliminary. Complete country data are presented in table C1. Vulnerability and Middle-Class Trends The vulnerability line, which was recently updated to $17.00 The share of vulnerable people in LAC increased steadily per day (2021 PPP), separates non-poor households into during 2000–14. In the last decade, the vulnerable class those who are vulnerable (facing a high probability of falling has remained relatively stable at around 32 percent. The into poverty) and those in the middle class (with a low middle class has continued to grow, increasing from 35.5 probability of falling into poverty). 6 percent of the population in 2016 to 42.3 percent in 2024 (figure 3). Figure 3 Poverty, Vulnerability, and Middle-Class Trends (2021 PPP) 60 50 42.3 42.8 Percentage of Population 40 32.2 32.0 30 25.5 25.2 20 10 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025e Poverty $8.30 per day Vulnerability $8.30-$17.00 per day Middle-Classs $17.00+ per day Source: SEDLAC (CEDLAS and World Bank). Note: The LAC aggregate is based on 18 countries with available SEDLAC microdata. When data were unavailable, values were estimated using microsimulations and then pooled to create regional estimates. The break in the LAC-18 series from 2014 onward is due to methodological changes in Mexico’s household survey in 2016. 6 Appendix B describes the methodology. 5 /27 The size of the vulnerable and the middle class varies Between 2022 and 2024, LAC’s middle-class expansion across countries: Uruguay and Chile show the lowest has been particularly pronounced in the Dominican vulnerability rates (20.4 percent in 2024 and 26.8 Republic and Costa Rica, with remarkable gains of 13.8 percent in 2022, respectively) alongside the highest and 8.9 p.p., respectively. The region’s largest economies, middle-class rates (73.7 and 67.5 percent, respectively), Brazil and Mexico, contributed significantly to regional while countries like El Salvador and Peru exhibit higher progress, with middle-class expansions of 5.5 and 6.4 vulnerability rates (40.2 and 39.4 percent, respectively) p.p., respectively. In the Caribbean, the middle class with smaller middle classes (29.9 and 24.4 percent, increased modestly from 44.7 to 46 percent, with poverty respectively). In the Caribbean, more than one-third of declining by 0.9 p.p.. However, it should be noted that the population (except in Saint Lucia) is vulnerable (figure these estimates are based on projections given the 4). This economic insecurity is fundamental considering limited availability of recent survey data. the Caribbean’s exposure to weather-related shocks: over three-quarters of households in Suriname, Belize, and Saint Lucia report exposure to disaster risks.7 Table 2 Poverty, Vulnerability, and Middle-Class Rates in LAC and Subregions, 2022–24 Upper Middle-Income Vulnerability Rate Middle-Class Rate Poverty Rate ($8.30-$17.00 per day) ($17.00+ per day) Country/Subregion ($8.30 per day) Change Change Change 2022 2024 2022 2024 2022 2024 (p.p.) (p.p.) (p.p.) Brazil 25.3 20.6 -4.7 30.5 29.6 -0.9 44.2 49.7 5.5 Mexico 27.4 21.7 -5.7 39.4 38.7 -0.7 33.2 39.6 6.4 Andean Subregion 35.8 34.0 -1.8 33.5 33.6 0.1 30.7 32.4 1.7 Central America Subregion 36.8 34.3 -2.5 33.8 32.6 -1.2 29.4 33.1 3.7 Southern Cone Subregion 12.1 12.0 -0.1 29.5 28.6 -0.9 58.4 59.4 1.0 LAC-18 29.0 25.5 -3.5 32.8 32.2 -0.6 38.2 42.3 4.1 Caribbean 19.2 18.3 -0.9 36.1 35.7 -0.4 44.7 46.0 1.3 Source: SEDLAC (CEDLAS and World Bank). Note: LAC aggregate based on 18 countries with available SEDLAC microdata. In cases where data were unavailable, values have been estimated using microsimulations and then pooled to create regional estimates. Subregions definitions as in table 1. Brazil and Mexico 2024 data are preliminary. Caribbean figures are based on six countries with available consumption microdata: Barbados (2016), Belize (2018), Grenada (2018), Jamaica (2021), Saint Lucia (2015), and Suriname (2022). The Caribbean estimates were projected to 2022 and 2024 using neutral distribution based on GDP growth (see Macro Poverty Outlook for methodological details) and then aggregated as a population-weighted average. Poverty, Vulnerability, and Middle Class in the Caribbean Recently harmonized consumption data from six countries (Barbados, Belize, Jamaica, Grenada, Saint Lucia, and Suriname) indicate that poverty rates in the Caribbean are generally below the upper-middle-income countries (UMIC) and LAC averages, in line with their relatively high GDP per capita. Poverty rates at $8.30 per day range from 7.8 percent in Saint Lucia to 22.9 percent in Jamaica (figure 4). 7 The analysis focused on the Caribbean is based on World Bank (forthcoming) and Anglade et al. (2024). 6 /27 Figure 4 Vulnerability to Falling into Poverty is common across the Caribbean 100.0 90.0 28.4 32.1 80.0 40.3 41.9 38.7 38.2 Percentage of Population 43.5 43.6 70.0 68.0 60.0 50.0 44.9 32.8 48.1 40.0 40.1 35.3 40.4 40.7 36.7 30.0 20.0 24.2 26.7 29.0 10.0 19.7 19.6 22.9 20.9 19.9 15.8 7.8 0.0 Barbados Belize Grenada Jamaica Saint Lucia Suriname UMIC LAC SIDS 2016 2018 2018 2021 2015 2022 Poverty $8.30 per day Vulnerability $8.30 - $17.00 per day Middle-Class $17.00+ per day Sources: Caribbean Consumption-Based Harmonization (CONLAC) for the six Caribbean countries; SEDLAC for LAC (2022); and PIP for UMIC and SIDS (2022). Note: For Caribbean countries, Barbados used the BSLC 2016 survey, Belize used the HBS 2018 survey, Grenada used the SLCHB 2018 survey, Jamaica used the JSLC survey 2021, Saint Lucia used the SLCHBS 2015 survey, and Suriname used the SSLC 2022 survey. Small Island Developing States (SIDS), all consumption-based, consist of Fiji (2019), Mauritius (2017), Seychelles (2018), Maldives (2019), Marshall Islands (2019), Tuvalu (2010), Tonga (2021), Nauru (2012), and the Dominican Republic (2022). P O V E R T Y A N D I N E Q UA L I T Y 7 /27 3. Poverty, Vulnerability, and Middle-Class Profiles Children bear a disproportionate burden of poverty in for long-term mobility, labor market conditions have LAC, while older adults are significantly better off. More prevented the poor from reaping the full economic than one in three people living in poverty are under 18 benefits of their educational progress, highlighting years old, a ratio unchanged over the past decade, and a disconnect between educational achievement and children represent only 13 percent of the middle class income gains in the region (Neidhöfer, Ciaschi, and (table 3). The pattern reverses for older populations: the Gasparini 2022). elderly comprise 15.5 percent of the middle class but Labor market informality, measured as the share of just 5.6 percent of those in poverty, likely reflecting their workers without pension contributions, has either greater asset accumulation and pension income over increased or remained unchanged across socioeconomic time. groups in the region between 2016 and 2024. Among Educational attainment in LAC has risen substantially the poor, 8 out of 10 workers hold informal jobs, across all income groups between 2016 and 2024, with compared to one-third among the middle class. the most pronounced gains occurring among the poor. Informal employment is generally linked to lower The share of individuals with at least a high school productivity, limited benefits, and weaker social diploma increased by 3.4 and 4.4 p.p. among the middle protection mechanisms, contributing to a vicious cycle class and vulnerable groups, respectively, compared to of poverty and poor-quality jobs. In addition, economic 5.8 p.p. among the poor. However, these educational development and poverty reduction go hand in hand gains have not translated into proportionate income with a shift from self-employment to salaried work. improvements due to declining returns to education However, the share of salaried workers remained in the labor market, which have reduced wage growth stagnant and even declined among the poor between (World Bank 2025b). While human capital is essential 2016 and 2024 in LAC. Table 3 Socioeconomic Profiles of the Poor, Vulnerable, and Middle Class in LAC, 2016–24 (percentage) International Poverty Rate Vulnerability Rate Middle-Class Rate $8.30 per day $8.30–$17.00 per day $17.00+ per day 2016 2024 2016 2024 2016 2024 Age groups 0–14 37.0 36.0 25.4 24.7 14.5 12.9 15–64 58.2 58.5 67.3 66.8 72.7 71.6 65+ 4.8 5.6 7.3 8.5 12.9 15.5 Total 100 100 100 100 100 100 Education (a) Low education 65.4 59.6 50.0 45.6 33.4 30.0 High education 34.6 40.4 50.0 54.4 66.6 70.0 Total 100 100 100 100 100 100 8 /27 International Poverty Rate Vulnerability Rate Middle-Class Rate $8.30 per day $8.30–$17.00 per day $17.00+ per day 2016 2024 2016 2024 2016 2024 Informality (b) Informal workers (c) 80.4 83.4 54.8 58.6 31.9 32.4 Formal workers 19.6 16.6 45.2 41.4 68.1 67.6 Total 100 100 100 100 100 100 Type of employment Employer 4.4 4.5 3.1 3.2 6.2 5.2 Salaried worker 43.9 42.5 63.5 63.6 69.4 70.7 Self-employed 28.9 31.5 21.7 23.2 18.6 19.7 Unemployed or unpaid worker 22.8 21.5 11.7 10.0 5.9 4.3 Total 100 100 100 100 100 100 Source: SEDLAC (CEDLAS and World Bank). Note: LAC aggregate based on 18 countries with available microdata. In cases where data were unavailable, values have been estimated using microsimulations and then pooled to create regional estimates. (a) Low education: no formal education or completed primary education; high education: completed secondary or tertiary education. (b) Working people aged 15–64 years. (c) Workers without work-related pension insurance. For Argentina: salaried workers without pension insurance and unpaid workers without complete tertiary education. For Mexico: workers without work-related health insurance benefits. For Honduras, unpaid workers without tertiary education or with tertiary education are employed by small private companies, and salaried workers with limited education are employed by small private companies. The poor in the Caribbean have a similar profile to their connected. The share of poor people with primary or no counterparts in the rest of the LAC region. First, children education ranges from 12.5 percent in Jamaica to 72.4 represent between 29 and 46 percent of the poor. percent in Belize. Second, educational attainment and poverty are strongly P O V E R T Y A N D I N E Q UA L I T Y 9 /27 4. Inequality The LAC region ranks among the most unequal in the quintile receiving 8 to 9 percent. Inequality also varies world. The top 20 percent of households capture 54 8 within LAC: Brazil and Colombia stand out as the most percent of total income, while the poorest 20 percent unequal countries in the region, with the richest 20 receive only 4 percent (figure 5). By comparison, in all percent accounting for 55 to 59 percent of income. In other regions except Sub-Saharan Africa, the top quintile contrast, the top quintile’s share is lower in the Dominican claims 40 to 43 percent of income, with the bottom Republic, El Salvador, and Uruguay, at 45 to 46 percent. Figure 5 Income Distribution (Share) within LAC and LAC vs. Other Regions, Circa 2024 4% 4% 4% 5% 4% 6% 6% 8% 8% 8% 9% Income Share (percentage) 41% 42% 43% 46% 45% 42% 46% 50% 50% 53% 51% 55% 54% 53% 49% 49% 54% 48% 43% 42% 40% 40% Brazil Andean Central Southern Mexico LAC Sub-Saharan East Middle East Europe South Subregion America Cone Africa Asia & & North & Central Asia Subregion Subregion Pacific Africa Asia Top 20 Middle 60 Bottom 20 Gap Sources: SEDLAC (CEDLAS and World Bank) and Poverty and Inequality Platform (PIP). Note: Percentages based on income (LAC) or consumption (non-LAC regions). LAC aggregate based on 18 countries with available SEDLAC microdata. In cases where data were unavailable, values have been estimated using microsimulations and then pooled to create regional estimates. Subregions definitions as in table 1. Brazil and Mexico 2024 data are preliminary. The percentage share for the rest of the regions is an average of the shares of each country (income or consumption). In 2024, the Gini coefficient for LAC reached 49.1 America and the Southern Cone experienced a 2.1- and points, a value higher than the World Bank’s 40-point 1-point decline in the Gini coefficient, respectively. In the threshold for classifying a country as characterized by Caribbean countries, inequality is generally high: the Gini “high inequality” (figure 6). At the same time, it declined coefficient is close to or above the World Bank’s global by 2.3 points between 2016 and 2024. This regional threshold for high inequality in all the countries analyzed improvement was driven primarily by reductions in the apart from Barbados (where the Gini is 34). Similarly, two largest economies: Mexico (-4.3 points) and Brazil in all countries except Barbados, the top 10 percent of (-3.1 points). However, these gains were partially offset the population consumes nearly twice as much as the by increases in the Andean region (+1.3 points). Central bottom 40 percent. 8 Regional inequality comparisons should be made with caution, because inequality of income is typically higher than inequality of consumption (World Bank 2024). While LAC countries report income-based inequality measures, most other countries use consumption-based measures (World Bank, 2016, 77–80). In fact, wage inequality in Brazil and Colombia resembles levels in South Asian countries such as India and Sri Lanka, whereas other LAC countries align with East Asia and Pacific countries such as Thailand and the Philippines (World Bank 2025b). 10 /27 Figure 6 Gini Coefficients for Brazil, Mexico, LAC, and LAC Subregions, 2016, 2022, 2024 55.0 50.0 Gini Coefficient 45.0 40.0 35.0 50.3 49.1 49.0 47.9 43.4 42.6 30.0 Brazil LAC Andean Region Central America Southern Cone Mexico 2024 2016 2022 Source: SEDLAC (CEDLAS and World Bank). Note: LAC aggregate based on 18 countries with available SEDLAC microdata. Pooled Gini coefficient for LAC calculated using pooled microdata for all countries. In cases where data were unavailable, values have been estimated using microsimulations. Subregions definitions as in table 1. Brazil and Mexico 2024 data are preliminary. The decline in income inequality was largely driven by driver of growth among the middle- and top-income broad-based income growth, which was slightly higher deciles, accounting for 2.3-3.4 p.p. of total income growth at the bottom and middle deciles of the distribution from decile 3 through decile 10. In contrast, non-labor (figure 7). The middle-income deciles (3 to 6) recorded income played a more significant role among the poorest the fastest gains, averaging approximately 4.7 percent, 20 percent of households, contributing 2.4 to 2.8 p.p. of while the poorest decile grew by 4.4 percent and the top total income growth. decile by only 2.5 percent. Labor income was the main Figure 7 Income Growth by Decile, LAC, 2022–24 5.0 4.5 4.0 1.6 1.2 1.5 1.7 1.0 Growth Rate (Annualized) 2.0 3.5 1.0 2.4 3.0 2.8 2.5 0.2 2.0 3.4 3.2 3.3 1.5 3.1 2.9 2.9 2.7 2.1 2.3 1.0 1.6 0.5 - 1 2 3 4 5 6 7 8 9 10 Deciles of Per Capita Household Income Labor income Non-Labor Income Source: SEDLAC (CEDLAS and World Bank). Note: LAC aggregate based on 18 countries with available microdata. In cases where data were unavailable, values have been estimated using microsimulations and then pooled to create regional estimates. Brazil, Colombia, and Mexico 2024 data are preliminary. Growth rates for labor and non-labor income are weighted by each component’s share in total income; therefore, they represent contributions to total income growth rather than stand-alone growth rates for each component. P O V E R T Y A N D I N E Q UA L I T Y 11 /27 5. Poverty Drivers Jobs and public transfers were the main poverty these declines, to a lesser extent. The prepandemic reduction drivers in LAC between 2016 and 2024, years (2016–19) brought only a modest 1-p.p. decline in though their relative importance shifted across poverty, reflecting stagnant labor markets and limited subperiods (figure 8). Over the full period, labor market transfer expansion. During the pandemic (2019–22), improvements, through employment and earnings poverty fell slightly, primarily due to the swift and growth, accounted for 3.1 p.p. (around 37 percent) of sizable rollout of public transfers, while labor market the total 8.4-point decline in poverty. Public transfers conditions contributed little. The picture reversed in the contributed another 1.9 points, about one-quarter postpandemic recovery (2022–24): more than half of of the overall reduction. Pensions, the demographic the poverty reduction came from more jobs and higher transition, and remittances were also factors driving earnings, while transfers played only a minor role. Figure 8 Drivers of Poverty Changes in LAC, 2016-24 1.0 0.0 -0.1 0.1 -0.4 -0.4 -1.3 -1.0 -2.1 -0.1 -3.1 -2.0 -0.3 Percentage Points -3.0 -0.3 -0.8 -0.9 -4.0 -5.0 -1.9 -6.0 -0.3 -0.7 -7.0 -1.2 -8.0 -9.0 2016 - 2019 2019 - 2022 2022 - 2024 2016 - 2024 Labor Market Demography Public Transfers Remittances Pensions Other Non-labor Income Source: SEDLAC (CEDLAS and World Bank). Note: LAC aggregate varies by period based on available comparable microdata. Core countries present across most periods include Argentina, Brazil, Costa Rica, Ecuador, Peru, and Uruguay, with additional countries added depending on data availability: 2016–19 (12 countries), 2019–22 (9 countries), 2022–24 (14 countries), and 2016–24 (13 countries). “Labor Market” includes labor income and share of employed people, and “Other Non-Labor Income” includes capital income, rentals, monetary and nonmonetary internal transfers, imputed rent, and other nonclassifiable non-labor income. Between 2016 and 2024, the drivers of poverty reduction Salvador. In most other countries, employment growth varied widely across LAC countries. The labor market’s and earnings gains were smaller (figure 9.1). contribution to poverty reduction at the regional level Following the pandemic, a labor market recovery was driven by the two biggest economies, Brazil and occurred in most LAC countries that played a key role in Mexico. Beyond these countries, the labor market played reducing poverty. In eight countries, employment and a notable role mainly in the Dominican Republic and El wage growth accounted for between 1.5 and 4 p.p. of 12 /27 poverty reduction, representing 48 to 94 percent of the the largest LAC economies may partly explain the labor total poverty decline (figure 9.2). Unlike in the pandemic market’s strong role in reducing poverty in the long period, public transfers had a limited impact during (2016–24) and short term (2022–24) (Engbom and Moser this time, except for Brazil, where they contributed 38 2022; World Bank 2025a). percent of poverty reduction. Rising minimum wages in Figure 9.1 Drivers of Poverty Rate Changes, LAC and Selected Countries, 2016–24 10.0 5.0 Percentage Points 0.0 -5.0 -10.0 -15.0 -20.0 Argentina Ecuador Guatemala Uruguay Peru Panama Costa Rica Honduras Bolivia LAC Brazil Dominican El Salvador Mexico (2014 - 2023) (2016 - 2023) Republic (2016 - 2023) (2017 - 2023) Labor Market Demography Public Transfers Remittances Pensions Other Non-labor Income Source: SEDLAC (CEDLAS and World Bank). Note: LAC aggregate based on 13 countries with available and comparable microdata: Argentina, Brazil, Costa Rica, Ecuador, Honduras, Mexico, Panama, Peru, and Uruguay, as well as Bolivia and El Salvador (2016–23), Dominican Republic (2017–24), and Guatemala (2014–23). Data from Uruguay for 2016 are not strictly comparable with data from 2024. Income definitions as in figure 8. Argentina has urban coverage only. Figure 9.2 Drivers of Poverty Changes, LAC and Selected Countries, 2022–24 4.0 2.0 0.0 Percentage Points -2.0 -4.0 -6.0 -8.0 Ecuador Argentina Panama Uruguay Peru Bolivia El Salvador Colombia Honduras LAC Costa Rica Brazil Paraguay Mexico Dominican (2023-2024) (2022-2023) (2022-2023) (2023-2024) Republic Labor Market Demography Public Transfers Remittances Pensions Other Non-labor Income Source: SEDLAC (CEDLAS and World Bank). Note: LAC aggregate based on 14 countries with available and comparable microdata: Argentina, Brazil, Colombia, Costa Rica, Dominican Republic, Ecuador, Mexico, Paraguay, Peru, and Uruguay, as well as Panama and Honduras (2023–24) and Bolivia and El Salvador (2022–23). Income definitions as in figure 8. Argentina has urban coverage only. P O V E R T Y A N D I N E Q UA L I T Y 13 /27 6. Jobs and Poverty Transitions Household income and poverty are shaped not just by Poverty in the study period was persistent in LAC, as people’s current jobs, but also by the paths they have more than half and up to three-quarters of initially poor taken in recent years, for example, whether they recently households remained poor, ranging from 58 percent in lost or found a job or moved to a high-skill job. Cross- Argentina to 76 percent in Brazil.10 Notably, poverty was sectional data, which take only a snapshot in time, cannot more persistent in countries with higher poverty rates, capture these dynamics. To address this limitation, this suggesting a link between low short-term socioeconomic section draws on panel data from household surveys mobility and poverty. Similarly, most non-poor households in five LAC countries—Argentina, Brazil, the Dominican remained non-poor, ranging from 83 percent of panel Republic, El Salvador, and Peru—covering the periods households in El Salvador and Peru to over 90 percent in 2015–19 and 2021–23. 9 Argentina, Brazil, and the Dominican Republic (figure 10). Figure 10. Shares of Households Changing Poverty Status, 1-Year Transition Averages, 2016–23 Transition averages for initially poor households 100 45.0 38.3 90 40.0 Share of poor households 24 37 80 26 35.0 Average Poverty Rate 70 34.0 30.0 (percentage) 42 43 60 25.0 50 27.3 20.0 40 20.6 15.0 30 20 13.9 10.0 10 5.0 58 76 57 74 63 0 0.0 Argentina Brazil Dominican Republic Peru El Salvador Remained poor Escaped poverty Poverty rate (av. 2016-2023) Transition averages for initially non-poor households 100 45.0 8 10 8 90 17 17 40.0 Share of poor households 80 Average Poverty Rate 35.0 70 38.3 30.0 (percentage) 60 34.0 25.0 50 13.9 27.3 20.0 40 30 20.6 15.0 20 10.0 10 92 90 92 83 83 5.0 0 0.0 Argentina Brazil Dominican Republic Peru El Salvador Remained Non-Poor Fell Into Poverty Poverty rate (av. 2016-2023) Source: SEDLAC (CEDLAS and World Bank). Note: Calculations based on one-year panels covering 2017–18, 2018–19, and 2022–23. The poverty headcount corresponds to the average between 2016 and 2023. 9 See appendix D1 for detailed information on the panel construction. The calculations are based on one-year panels for specific periods in each country, depending on data availability: Argentina (2016–19 and 2021–23), Brazil (2016–19 and 2022–23), Dominican Republic (2017–19 and 2021–23), El Salvador (2018–19 and 2021–23), and Peru (2015–19 and 2021–23). 10 These are the average conditional transition rates for one-year panels covering 2017–18, 2018–19, and 2022–23, as these are the periods for which all five countries have available panel data. Nevertheless, the country-level conditional transition rates remain almost unchanged when using all the countries’ corresponding available data. 14 /27 Jobs are a key pathway out of poverty and provide While employment is a critical pathway out of poverty, economic protection. When household heads transition the results reveal important nuances about job quality from unemployment or inactivity into work, households and poverty transitions. When the household head exhibit a 13.8 p.p. higher probability of escaping (typically the primary income earner) transitions into or poverty than poor households who did not experience out of employment, the probability of the household’s this transition. Likewise, the same transition into 11 crossing the poverty or vulnerability threshold changes employment increases the likelihood of reaching middle by approximately 13 to 14 p.p. Given that this represents class status by 13.4 p.p. for households initially under a substantial shift in employment status for the the vulnerability line of $17.00 per day (2021 PPP). In household head, the relatively modest magnitude of contrast, losing a job increases the likelihood of falling the effect suggests that such transitions often involve into poverty by 14.1 p.p. and increases the chances of movement from one low-quality job to another, yielding leaving the middle class by 13.5 p.p. (figure 11). With only marginal improvements in household welfare. some variations, these patterns are observed in all five Indeed, workers who cycle in and out of employment countries. disproportionately transition from or to informal and lower-skilled occupations, compared with the average worker in the labor market (table D4).12 Figure 11 Marginal Effects of Job Gain and Loss on the Probability of Changing Poverty or Middle-Class Status in LAC 20.0 15.0 Percentage Points 10.0 5.0 13.8 13.4 - 2.8 - 1.4 14.1 13.5 0.0 - 4.1 - 4.3 -5.0 Job Gain Job Loss Out of Poverty Enters Middle Class Into Poverty Leaves Middle Class Source: SEDLAC (CEDLAS and World Bank). Note: The LAC aggregate is based on five countries with available panel microdata for 2021–23: Argentina, Brazil, the Dominican Republic, El Salvador, and Peru. Calculations are based on one-year transition panels. For Brazil, only the 2022–23 panel data are used. All estimations include year and country– region fixed effects, with errors clustered at the country–region level. Skill levels are defined according to the ILO occupational classification: high skill (managers, professionals, technicians and associate professionals), medium skill (clerical support workers, service and sales workers, skilled agricultural, forestry and fishery workers, craft and related trades workers, plant and machine operators), and low skill (elementary occupations). The regression is at the household level. Occupational mobility within employment plays a crucial accountant or from waiter to restaurant manager) can role in poverty and vulnerability dynamics. Occupational propel households toward greater economic security. upgrading (moving to positions that require higher Conversely, occupational downgrading (shifting to skills, such as advancing from bookkeeping assistant to lower-skilled roles, such as moving from cook to kitchen 11 Country-level and pooled regional regressions were implemented on the panel data to examine how labor market transitions are associated with people’s poverty and vulnerability status. See appendix D for detailed information on the regression analysis methodology. These findings represent correlational rather than causal relationships, because the analysis is constrained by the endogenous nature of the dependent variables. Job transitions are likely linked to unobservable characteristics that affect poverty and vulnerability status, which can bias the estimated coefficients. 12 These findings are consistent with those from Donovan, Lu, and Schoellman (2023), who find that labor market turnover is higher in developing countries but is mostly limited to transitions in and out of lower-quality jobs. 15 /27 helper or from cashier to shelf stocker) can push Households whose heads experience occupational households toward vulnerability or poverty. In LAC, when upgrading show increased chances of escaping poverty in household heads experience occupational upgrading, all five studied countries, with estimates ranging from 1.3 households show a 3.3 p.p. higher probability of escaping p.p. in Argentina to 14.4 p.p. in El Salvador.13 In contrast, poverty (figure 12). Similarly, occupational upgrading occupational downgrading shows a more limited is associated with a 3.1 p.p. increase in the likelihood association with falling into poverty, with statistically of reaching middle-class status for initially vulnerable significant increases observed only in Brazil and Peru households. Conversely, occupational downgrading at 2.4 and 3.4 p.p. respectively. Similarly, occupational appears to be linked to deteriorating economic downgrading appears linked to leaving the middle class outcomes, with households experiencing a 1.8 p.p. higher primarily in Brazil, where it increases this probability by probability of falling into poverty when the household 3.5 p.p.14 head moves to a lower-skilled position. Figure 12 Marginal Effects of Job Upgrading and Downgrading on the Probability of Changing Poverty or Middle-Class Status in LAC 7.0 5.0 3.3 3.1 3.0 1.8 1.8 Percentage Points 1.0 1.0 0.7 0.2 0.5 -1.0 -3.0 -5.0 Occupational Upgrading Occupational Downgrading Out of Poverty Enters Middle Class Into Poverty Leaves Middle Class Source: SEDLAC (CEDLAS and World Bank). Note: The LAC aggregate is based on five countries with available panel microdata for the 2021–23 period: Argentina, Brazil, the Dominican Republic, El Salvador, and Peru. Calculations are based on one-year transition panels. For Brazil, only the 2022–23 panel data are used. All estimations include year and country–region fixed effects, with errors clustered at the country–region level. Skill levels are defined according to the ILO occupational classification: high skill (managers, professionals, technicians and associate professionals), medium skill (clerical support workers, service and sales workers, skilled agricultural, forestry and fishery workers, craft and related trades workers, plant and machine operators), and low skill (elementary occupations). Regression is at the household level. 13 See appendix D for more detailed information. 14 These findings complement those of Menezes-Filho and Narita (2025), who analyze labor market turnover using longitudinal data sets from five LAC countries. Although the authors do not examine associations with poverty and middle-class-status transitions, they find that job-to-job changes generally increase wages, which should be expected to be linked with poverty reduction and middle-class expansion. They also report that workers who change occupations experience smaller wage gains than those who remain in the same occupation, but they do not distinguish between occupational upgrading and downgrading when examining wage effects. P O V E R T Y A N D I N E Q UA L I T Y 16 /27 7. Conclusions The recent labor market recovery across the LAC have constrained LAC’s growth for decades. On the region has been a central driver of the process of demand side, easing bottlenecks in strategic sectors poverty reduction observed since 2022, outpacing most such as agribusiness, tourism, and renewable energy other regions. While this represents an encouraging can unlock new sources of employment, including for development, questions remain about the sustainability low-skilled workers, while a more predictable regulatory of this recent progress. Persistent stagnation in labor environment and stronger competition frameworks productivity over the past decade points to structural encourage investment and innovation. On the supply limitations, particularly in attracting investment into side, raising the quality of education at all levels and dynamic, higher-value sectors that are better positioned forging tighter linkages between formal education to generate quality employment. These constraints, curricula and the private sector are critical for enabling coupled with heightened global uncertainty arising adaptation to technological change and expanding from factors ranging from disruptions in international opportunities for workers. Equally important are reforms trade to the transformative impacts of artificial that deepen capital markets, improve infrastructure, intelligence, underscore the risk that current gains in modernize tax systems, and strengthen institutions to poverty reduction could stall without deeper reforms to better manage the risks inherent in entrepreneurship strengthen productivity and resilience. and innovation (Maloney et al. 2025). These policies will be key to transforming the current recovery into a more To sustain momentum, policies to generate more and inclusive and sustainable path of long-term growth and better jobs must address both immediate labor market poverty reduction. challenges and the deeper structural barriers that 17 /27 References Anglade, Boaz, Emilia Cucagna, Jacobus de Hoop, and Anna Luisa Paffhausen. 2024. “Disaster Risk Preparedness of Households in the Caribbean.” International Journal of Disaster Risk Reduction 115: 104956. Donovan, Kevin, Will Jianyu Lu, and Todd Schoellman. 2023. “Labor Market Dynamics and Development.” Quarterly Journal of Economics 138 (4): 2287–2325. Engbom, Niklas, and Christian Moser. 2022. “Earnings Inequality and the Minimum Wage: Evidence from Brazil.” American Economic Review 112 (12): 3803–47. Fernandez, Jaime, Sergio Olivieri, and Diana Sanchez. 2023. “A Methodology for Updating International Middle- Class Lines for the Latin American and Caribbean Region.” Policy Research Working Paper 10447, World Bank, Washington, DC. Ferreira, Francisco H. G., Julián Messina, Jamele Rigolini, Luis F. López-Calva, Maria Ana Lugo, and Renos Vakis. 2013. Economic Mobility and the Rise of the Latin American Middle Class. Washington, DC: World Bank. Foster, Elizabeth Mary, Dean Mitchell Jolliffe, Gabriel Lara Ibarra, Christoph Lakner, and Samuel Kofi Tetteh Baah. 2025. “Global Poverty Revisited Using 2021 PPPs and New Data on Consumption.” Policy Research Working Paper 11137, World Bank, Washington, DC. International Labour Organization. 2025. “ Correspondence between broad skill levels and ISCO classifications [Table]”. In Classification of occupations: Skill levels. ILOSTAT. Instituto Nacional de Estadística y Censos (INDEC). 2025. “Incidencia de la pobreza y la indigencia en 31 aglomerados urbanos.” Technical Report 9 (237), INDEC, Buenos Aires, AR. Jenkins, Stephen P. 2017. “Pareto Models, Top Incomes and Recent Trends in UK Income Inequality.” Economica 84 (334): 261–89. López-Calva, Luis F., and Eduardo Ortiz-Juarez. 2014. “A Vulnerability Approach to the Definition of the Middle Class.” Journal of Economic Inequality 12: 23–47. Lustig, Nora. 2019. “The ‘Missing Rich’ in Household Surveys: Causes and Correction Approaches.” Working Paper 75, Commitment to Equity, Tulane University, Department of Economics, New Orleans, LA. Maloney, William, Guillermo Vuletin, Pablo Garriga, and Raul Morales. 2025. Transformational Entrepreneurship for Jobs and Growth. Washington, DC: World Bank. Menezes-Filho, Naercio, and Renata Narita. 2025. “Labor Market Turnover and Inequality in Latin America.” Oxford Open Economics 4 (Issue Supplement_1): i349–i375. Montoya, Kelly, Sergio Olivieri, and Cicero Braga. 2023. “Considering Labor Informality in Forecasting Poverty and Inequality: A Microsimulation Model for Latin American and Caribbean Countries.” Policy Research Working Paper 10497, World Bank, Washington, DC. Neidhöfer, Guido, Matías Ciaschi, and Leonardo Gasparini. 2022. “Intergenerational Mobility of Economic Well- Being in Latin America.” Working Paper 303, CEDLAS, Buenos Aires, AR. World Bank. 2016. Poverty and Shared Prosperity 2016: Taking on Inequality. Washington, DC: World Bank. World Bank. 2024. Poverty, Prosperity, and Planet Report 2024: Pathways out of the Polycrisis. Washington, DC: World Bank. World Bank. 2025a. Mexico Poverty and Equity Assessment. Washington, DC: World Bank. World Bank. 2025b. Regional Jobs Update: Insights from Labor Force Surveys from Latin America and the Caribbean. Washington, DC: World Bank. World Bank. Forthcoming.“Shared Metrics, Shared Progress: Insights from Harmonized Data on Poverty and Inequality in the Caribbean.” 18 /27 Appendix Appendix A. Introducing the New 2021 Purchasing Power Parity (PPP) in Regional Estimates The World Bank revised its international poverty lines to poverty line, the regional rate increased from 3.9 percent reflect changes in living costs and consumption patterns to 5.3 percent in 2023. Poverty at the upper-middle- worldwide, using updated 2021 purchasing power parity income threshold rose from 25.1 percent to 27.9 percent. (PPP) rates and improved national poverty data. The This statistical adjustment translates to approximately 17 updated poverty lines are $3.00 per day for low-income million additional people classified as poor, including 8 countries, $4.20 per day for lower-middle-income million newly considered extremely poor. countries, and $8.30 per day for upper-middle-income While these methodological updates result in higher countries (2021 PPPs), replacing the previous thresholds poverty rates and numbers, the underlying poverty of $2.15, $3.65, and $6.85 per day, respectively (Foster et trends across the region remain consistent with the 2017 al. 2025).15 PPPs. This represents a methodological improvement These revisions present a slightly more challenging rather than a deterioration in living conditions, providing picture for LAC. Under the new $3.00 per day extreme a more accurate snapshot of poverty in the region. Figure A1 Impact of 2021 PPP Update on Poverty Rates in LAC, circa 2023 100.0 15.0 90.0 10.0 80.0 Poverty Rate (percentage) Poverty Change (p.p.) 70.0 5.0 60.0 50.0 0.0 40.0 -5.0 30.0 20.0 -10.0 10.0 0.0 -15.0 Haiti (2012) Honduras Guatemala Nicaragua (2014) Colombia Peru Ecuador El Salvador LAC Mexico (2022) Jamaica (2021) Paraguay Suriname (2022) Panama Belize (2018) Grenada (2018) Barbados (2016) Brazil Dominican Republic Bolivia Argentina Costa Rica Uruguay Chile (2022) Saint Lucia (2015) Poverty $8.30 per day, 2021 PPP Change 2017 PPP vs. 2021 PPP Source: SEDLAC (CEDLAS and World Bank). Note: LAC aggregate based on 18 countries in the region for which microdata (Income) are available. When data is unavailable, values have been estimated using microsimulations and then pooled to create regional estimates. Argentina only has urban coverage. Colombia 2023 data are preliminary and do not necessarily coincide with other sources or World Bank documents. Gray circles show the change in p.p. between 2017 PPP and 2021 PPP rates (right axis). 15 Three key factors shaped this revision: enhanced PPP conversion factors that better capture cross-country cost-of-living differences, improved household survey methodologies in low-income countries that provide more-accurate welfare measurements, and updated national poverty lines across middle-income countries. These adjustments incorporate data from the 2021 International Comparison Program, which captured post- COVID-19 price levels. 19 /27 Appendix B. Definition of a Vulnerability to Poverty Line Using 2021 PPP In 2013, the World Bank report Economic Mobility and the and the middle-class line as the 99th percentile income Rise of the Latin American Middle Class (Ferreira et al. 2013) among those same households. This methodology first introduced vulnerability and middle-class lines to established new regional lines of $14 per day for distinguish between the non-poor with economic security vulnerability and $81 per day for middle-class status (that is, the middle class) and those above poverty (2017 PPP). thresholds who may still face high risks of falling into The new vulnerability line update to 2021 PPPs uses poverty due to economic shocks (that is, the vulnerable the López-Calva and Ortiz-Juarez (2014) methodology population). The report adopted the methodology of applied to panel data from five LAC countries covering López-Calva and Ortiz-Juarez (2014), which uses panel the periods 2015–19 and 2021–23: Argentina, Brazil, data from Chile, Mexico, and Peru to identify income Dominican Republic, El Salvador, and Peru.17 This levels associated with a 10 percent probability of falling methodology was selected as the primary approach into poverty. This approach led to a regional vulnerability because it offers two key advantages over synthetic line of $10 per day (2005 PPP), the average across the panels and simple extrapolation methods: (1) it uses three countries, and a middle-class poverty line of recent data from several LAC countries with broad $50 per day to split the middle and upper class. When regional representation, encompassing the region’s international poverty lines were updated to 2011 PPPs in largest economy, the Southern Cone, the Andean region, 2014, the vulnerability and middle-class lines were simply and Central America; and (2) the panel data enable direct recalculated using the country-level extrapolation factors observation of actual transitions into and out of poverty from 2005 PPPs to 2011 PPPs. rather than simulated ones. The methodology’s three- With the adoption of 2017 PPPs, Fernandez, Olivieri, stage procedure relates different levels of per capita and Sanchez (2023) reestimate these thresholds using a household income to probability of falling into poverty synthetic panel methodology across 15 LAC countries. 16 conditional on observable household characteristics These authors’ approach defined the vulnerability line as (figure B1).18 the median income of households who fell into poverty 16 Note that the results for the vulnerability line depend on the probability level chosen for the analysis. The 10 percent probability of falling into poverty was retained because the panel data for the selected 5 countries suggest this is still the approximate proportion of non-poor households who will fall into poverty in the subsequent period. 17 See appendix D for detailed information on the panel construction. The calculations are based on one-year panels for specific periods in each country, depending on data availability: Argentina (2016–19 and 2021–23), Brazil (2016–19 and 2022–23), Dominican Republic (2017–19 and 2021–23), El Salvador (2018–19 and 2021–23), and Peru (2015–19 and 2021–23). 18 This update does not consider an upper threshold for the middle class in LAC. This is because household surveys do not typically capture top incomes very well (Jenkins 2017; Lustig 2019). In fact, tax records reveal substantial discrepancies with household surveys at high income levels. For instance, in Brazil mean income above the 99th percentile is $26,477 in tax records versus $12,538 in household surveys (2021 PPP). Peru shows similar disparities, with $21,129 versus $5,438 for the same percentile (2021 PPP). These discrepancies also appear in income concentration measures. In Brazil, the top 1 percent holds 27.4 percent of total income according to tax records versus 12.4 percent in household surveys. Peru exhibits comparable patterns, with 27.0 percent versus 7.7 percent for the top income percentile. 20 /27 Figure B1 Probability of Falling into Poverty vs. Household per Capita Income per Day 90.0 80.0 Probability of falling into poverty (percentage) 70.0 60.0 50.0 40.0 30.0 20.0 10.0 0.0 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0 50.0 Household per capita income per day (2021 ppp) Brazil Peru Argentina Dominican Republic El Salvador Source: SEDLAC (CEDLAS and World Bank). Note: Calculations follow López-Calva and Ortiz-Juarez (2014) and are based on one-year panels for specific periods by country: Argentina (2016–19, 2021–23), Brazil (2016–19, 2022–23), Dominican Republic (2017–19, 2021–23), El Salvador (2018–19, 2021–23), and Peru (2015–19, 2021–23). The proposed methodology for updating the vulnerability As an additional robustness check, we implemented the line has some limitations. First, only five countries in synthetic panels methodology of Fernández, Olivieri, and the region have panel data suitable for estimation, Sanchez (2023). This approach yields a vulnerability line including the variables needed to construct a harmonized estimate of $16.5 per day (2021 PPP) for the five countries measure of household income per capita for poverty with panel data and $16.9 per day when using cross-sectional analysis. Second, the results depend on the set of control data for 15 LAC economies. Additionally, simple extrapolation variables selected for the estimation. However, sensitivity using conversion rates from 2017 PPPs to 2021 PPPs analysis demonstrates that the results remain robust produces an average vulnerability line of $15.9 per day for when varying the household characteristics included in the five selected countries and $15.5 per day when applied to the analysis. all 15 LAC countries with available cross-sectional data. Table B1 Vulnerability Lines Update for 5 LAC Countries Average of the Vulnerability Line for 5 LAC Countries Real Panels Country PPPs Extrapolation Synthetic Panels (LC-OJ 2014) Brazil $16.0 $16.6 $18.8 Argentina $15.3 $21.1 $19.2 Peru $15.3 $14.0 $17.2 Dominican Republic $17.5 $17.0 $15.6 El Salvador $15.2 $14.0 $14.9 Simple average $15.9 $16.5 $17.1 All available countries in SEDLAC 2021 PPPs $15.5 $16.9 Source: SEDLAC (CEDLAS and World Bank). Note: Real panels correspond to simple average of 1-year gap panels between 2016–19 and 2022–23. Values represent daily income thresholds in 2021 PPP. PPP extrapolation estimates based on updating previous vulnerability lines using PPP adjustment factors from 2017 to 2021 PPPs. Synthetic panels methodology presents 2-year panel averages excluding 2020. 21 /27 Appendix C Additional Figures and Tables Figure C1 Drivers of Inequality Changes by LAC Country, 2022–24 2.5 2.0 1.5 1.0 Percentage points 0.5 0.0 -0.5 -1.0 -1.5 -2.0 -2.5 Dominican Argentina El Salvador Panama Bolivia Peru Ecuador Paraguay Colombia Uruguay Mexico LAC Honduras Costa Rica Brazil Republic (2022-2023) (2023-2024) (2022-2023) (2023-2024) Labor Market Demography Public Transfers Remittances Pensions Other Non-labor Income Source: SEDLAC (CEDLAS and the World Bank). Note: LAC aggregate for 2022–24 is based on 14 countries with available and comparable microdata for both years: Argentina, Brazil, Colombia, Costa Rica, Dominican Republic, Ecuador, Mexico, Paraguay, Peru, and Uruguay, as well as Panama and Honduras (2023–24) and Bolivia and El Salvador (2022–23). Income definitions as in figure 7. Argentina has urban coverage only. Figure C2 Drivers of Inequality Changes by LAC Country, 2016–24 3.0 2.0 1.0 Percentage points 0.0 -1.0 -2.0 -3.0 -4.0 -5.0 Uruguay Ecuador Argentina El Salvador Panama Costa Rica Brazil Guatemala LAC Bolivia Dominican Peru Honduras Mexico (2016-2023) (2014-2023) (2016-2023) Republic (2017-2024) Labor Market Demography Public Transfers Remittances Pensions Other Non-labor Income Source: SEDLAC (CEDLAS and the World Bank). Note: LAC aggregate for 2016–24 is based on 13 countries with available and comparable microdata for both years: Argentina, Brazil, Costa Rica, Ecuador, Honduras, Mexico, Panama, Peru, and Uruguay, as well as Bolivia and El Salvador (2016–23), Dominican Republic (2017–24), and Guatemala (2014–23). The 2016 data for Uruguay are not strictly comparable with 2024. Income definitions as in figure 7. Argentina has urban coverage only. 22 /27 Table C1 Poverty Rates and Projections by LAC Countries, 2022–25e (percentage) International Lower-Middle-Income Upper-Middle-Income Country Poverty Rate Poverty Rate Poverty Rate $3.00 per day $4.20 per day $8.30 per day 2022 2023 2024 2025e 2022 2023 2024 2025e 2022 2023 2024 2025e Argentina 1.3 1.2 1.0 0.9 2.9 3.1 2.8 2.6 13.9 16.4 15.2 14.7 (urban) Bolivia 3.3 2.8 3.0 3.9 5.7 5.1 5.5 6.5 18.5 16.5 17.2 18.4 Brazil* 4.9 3.8 3.0 3.0 8.5 7.5 6.2 6.0 25.3 23.4 20.6 20.3 Chile 0.5 0.5 0.5 0.5 0.9 0.9 0.9 0.9 5.7 5.7 5.5 5.3 Colombia*, ** 9.3 8.6 8.5 8.3 15.8 15.2 14.7 14.2 40.2 39.1 37.0 36.2 Costa Rica 1.6 1.5 1.3 1.3 4.0 3.4 2.6 2.6 17.1 15.3 12.6 12.1 Dominican 1.0 1.3 0.8 0.7 3.2 3.0 2.0 1.9 20.2 16.9 14.0 13.4 Republic Ecuador 4.4 4.7 7.3 7.3 9.2 10.0 12.1 12.0 30.9 30.8 32.6 32.1 Guatemala 9.7 9.6 9.4 17.6 17.6 17.2 47.3 47.0 46.1 Honduras 17.0 15.7 14.0 26.6 23.5 21.1 53.3 49.8 46.1 El Salvador 5.2 4.6 4.9 5.2 9.6 8.6 8.9 9.0 32.7 29.9 29.9 29.9 Mexico* 2.3 1.7 1.7 5.7 4.2 4.2 27.4 21.7 21.8 Nicaragua 8.7 7.9 8.0 7.7 14.4 13.1 14.4 13.6 40.4 37.9 39.2 38.0 Panama 3.7 3.1 3.1 7.2 6.8 6.9 19.8 19.8 19.5 Paraguay 3.2 2.4 2.1 2.0 7.2 5.8 4.5 4.4 25.8 22.8 20.5 19.3 Peru 4.8 5.9 5.1 5.0 10.7 11.7 10.7 10.3 37.8 38.3 36.2 35.3 Uruguay 0.2 0.2 0.2 0.2 0.7 0.7 0.5 0.5 6.3 6.6 5.9 5.8 LAC 5.8 5.3 4.9 4.9 9.9 9.4 8.6 8.5 29.0 27.9 25.5 25.2 Source: SEDLAC (CEDLAS and the World Bank). Note: e = estimate. Highlighted cells indicate microsimulated data. For country-specific details on data comparability periods, refer to the comparability dashboard in the LAC Equity Lab (LEL). * Data for Brazil, Colombia, and Mexico in 2024 are preliminary. ** The 2023 data for Colombia were recently revised by the national statistics office and do not necessarily coincide with other sources or World Bank documents. Appendix D Using Panels for Five LAC Countries to Examine the Relationship between Labor Market Transitions and Socioeconomic Mobility The results presented in section 6 examine the D1. Panel Data Construction relationship between labor market transitions and Argentina’s EPHC survey employs a 2-2-2 rotating panel socioeconomic mobility, as well as the process of design with 25 percent panel renewal each quarter. updating the vulnerability line using PPP 2021. These Panel households are visited in two consecutive quarters, findings are derived from a regression analysis using skipped for two quarters, and then visited again for cross-country panel data from five LAC countries: Peru, two additional quarters. This design allows tracking Brazil, Argentina, Dominican Republic, and El Salvador. of 25 percent of the total sample year-to-year, with The data set combines these data covering different harmonized variables available through SEDLAC data. periods for each country: 2015–19 and 2021–23 for Peru, 2016–19 and 2021–23 for Argentina, 2017–19 and Brazil’s PNAD-C survey uses a 5-quarter continuous 2021–23 for Dominican Republic, 2016–19 and 2022–23 rotation scheme that tracks households for over for Brazil, and 2018–19 and 2021–23 for El Salvador. 15 months. While the public microdata lack unique individual or household identifiers, they provide 23 /27 key variables that enable the construction of these D2. Methodology for the Poverty and Jobs Transition identifiers. However, variables related to housing Regressions (Section 6) materials and ownership are not collected during the The estimation employs linear probability models using fifth visit. Because these variables are necessary for dummy variables that indicate labor market transitions: constructing imputed rent estimations—which are job entry, job exit, occupational upgrading, and required for calculating per capita household income occupational downgrading. Occupational upgrading and used in international poverty rate estimates—we applied downgrading are defined based on the ILO occupational the imputed rent from the first visit to the fifth visit. This classification described in table D5. Here occupational approach is feasible because households must remain upgrading is a binary variable equal to 1 if a worker in the same dwelling throughout the panel period to be moves from an occupation at skill level 1 to one at skill included. It is important to note that official international level 2 or 3, or from an occupation at skill level 2 to one poverty rates reported by the World Bank are calculated at skill level 3. Conversely, occupational downgrading is using only the first visit (except for rates reported in 2020 a binary variable equal to 1 when a worker transitions in and 2021, when the first visit data were not released). the opposite direction, from an occupation at a higher Dominican Republic’s ECNFT survey utilizes a 5-quarter skill set to one with a lower skill level. continuous rotation scheme, allowing year-to-year The model includes household and household’s head tracking of 25 percent of the total household survey characteristics measured at baseline as controls: urban sample, with harmonized variables available through residence, age group, gender, partnership status SEDLAC data. (whether the household head reports being married or El Salvador’s EHPM survey lacks an official panel design. having a partner present at the time of survey), education However, collaboration between the Oficina Nacional levels, household size, number of children, and number de Estadística y Censos (ONEC) and the El Salvador of workers in the household. poverty team at the World Bank enabled construction The specification incorporates country and subnational of a panel by using settlement geographical identifiers regional fixed effects to control for time-invariant and testing consistency of individual-level characteristics country-specific characteristics and period fixed effects such as age to analyze socioeconomic and labor market to account for common temporal shocks affecting all transitions among household heads. We acknowledge El countries. To ensure balanced representation across Salvador’s Country Poverty team for their collaboration in countries, we rescale the country-level weights so that constructing this panel data set. the within-country distribution is preserved and each Peru’s ENAHO survey provides the longest and most country’s observations sum to 100. Robust standard comprehensive panel data, tracking households for one errors are clustered at the country-region level to to five years during 2015–19 and 2019–24. The Instituto account for potential correlation in outcomes within Nacional de Estadística y Informática (INEI; National geographic areas. Institute of Statistics and Computing) of Peru provides Country-level regressions follow the same methodology preconstructed panel data sets that facilitate mapping to as the pooled regression, using survey weights, region- SEDLAC harmonized data sets. In this brief, we use only level fixed effects, and robust standard errors clustered the one-year panels, but we conduct a robustness check at the subnational regional level. using the longer panel structure, which is available upon request. 24 /27 Table D1 Transition Regression for 5 LAC Countries (Pooled) Going out of the Going into the Falling into Poverty Escaping Poverty Middle Class Middle Class 0.0276*** 0.138*** -0.0409*** 0.134*** Job gain -0.00736 -0.0129 -0.00697 -0.0143 0.141*** -0.0140** 0.135*** -0.0426*** Job loss -0.0119 (0.00636) -0.0156 -0.00952 0.00986 0.0328*** 0.00155 0.0305** Occupational upgrading -0.00634 -0.00853 -0.0165 -0.0121 0.0179** 0.00656 0.0182 0.00505 Occupational downgrading -0.00739 -0.0075 -0.0121 -0.00641 Observations 161,040 161,040 161,040 161,040 Adjusted R-squared 0.052 0.062 0.032 0.037 Source: Computations based on SEDLAC (CEDLAS and World Bank). Note: LAC aggregate is based on five countries with available panel microdata for the 2021–23 period: Argentina, Brazil, Dominican Republic, El Salvador, and Peru. Calculations are based on one-year transition panels. For Brazil, only the 2022–23 panel data are used. All estimations include a constant and year and country–region fixed effects, with errors clustered at the country–region level. Skill levels are defined according to the ILO occupational classification: high skill (managers, professionals, technicians and associate professionals), medium skill (clerical support workers, service and sales workers, skilled agricultural, forestry and fishery workers, craft and related trades workers, plant and machine operators), and low skill (elementary occupations). Demographic controls include residence area (urban/rural), age group, gender, an indicator for whether the household head has a partner in the household, household head’s level of education, number of household members, number of children in the household, and number of workers in the household. *** p < 0.01, ** p < 0.05, * p < 0.1. Table D2 Transition Regression for 5 LAC countries (Country Level) Going out of the Going into the Variables Falling into Poverty Escaping Poverty Middle Class Middle Class Argentina -0.0307*** 0.106*** -0.0449*** 0.0848*** Job gain -0.00674 -0.0112 -0.0083 -0.00752 0.114*** -0.0153** 0.135*** -0.0585*** Job loss -0.00468 -0.00484 -0.0155 -0.00342 0.00764 0.0129*** -0.0279 0.0146 Occupational upgrading -0.00569 -0.00229 -0.0195 -0.00809 -0.000933 -0.00464* -0.0121 0.00498 Occupational downgrading -0.00509 -0.00198 -0.0159 -0.00524 Observations 23,497 23,497 23,497 23,497 Brazil -0.0339*** 0.155*** -0.0267** 0.129*** Job gain -0.00342 -0.00736 -0.00939 -0.0247 0.165*** -0.0129 0.122*** -0.0287*** Job loss -0.00711 -0.0096 -0.0253 -0.00388 -0.00221 0.0209** -0.00321 0.0461*** Occupational upgrading -0.00297 -0.00597 -0.00555 -0.00503 0.0235*** -0.0016 0.0353*** -0.00265 Occupational downgrading -0.00408 -0.00416 -0.00371 -0.0053 Observations 112,894 112,894 112,894 112,894 25 /27 Going out of the Going into the Variables Falling into Poverty Escaping Poverty Middle Class Middle Class Dominican Republic -0.0182 0.135** -0.0496* 0.213*** Job gain -0.0266 -0.0521 -0.0258 -0.0177 0.140*** 0.0181 0.163*** -0.0403 Job loss -0.0374 -0.0224 -0.0471 -0.0276 0.0254 0.0833** 0.0543 0.0944** Occupational upgrading -0.016 -0.0359 -0.0599 -0.0391 0.032 0.0554 0.0467 -0.0256 Occupational downgrading -0.0202 -0.0498 -0.0418 -0.0228 Observations 2,233 2,233 2,233 2,233 El Salvador -0.015 0.0925*** -0.0303 0.0883** Job gain -0.0267 -0.00733 -0.02 -0.0301 0.159*** 0.0193 0.00274 -0.021 Job loss -0.0193 -0.0386 -0.0115 -0.0206 0.000585 0.144** -0.0514* 0.0335 Occupational upgrading -0.0399 -0.033 -0.0209 -0.0165 0.00289 0.0763** -0.0167 0.00616 Occupational downgrading -0.0137 -0.0249 -0.0102 -0.0136 Observations 5,092 5,092 5,092 5,092 Peru -0.0469*** 0.130*** -0.0568*** 0.115*** Job gain -0.0137 -0.0157 -0.0162 -0.0337 0.157*** -0.0272** 0.0989*** -0.0740*** Job loss -0.0219 -0.0118 -0.0273 -0.0175 0.0127 0.0280** -0.00952 -0.0120** Occupational upgrading -0.0168 -0.0109 -0.009 -0.00554 0.0339* -0.00345 0.0111 0.007 Occupational downgrading -0.0183 -0.0123 -0.0153 -0.0158 Observations 17,324 17,324 17,324 17,324 Source: Computations based on SEDLAC (CEDLAS and World Bank). Note: Calculations are based on one-year transition panels using 2021–23 data for Argentina. All estimations include a constant and year and country– region fixed effects, with errors clustered at the country–region level. Skill levels are defined according to the ILO occupational classification: high skill (managers, professionals, technicians and associate professionals), medium skill (clerical support workers, service and sales workers, skilled agricultural, forestry and fishery workers, craft and related trades workers, plant and machine operators), and low skill (elementary occupations). Demographic controls include residence area (urban/rural), age group, gender, an indicator for whether the household head has a partner in the household, household head’s level of education, number of household members, number of children in the household, and number of workers in the household. *** p < 0.01, ** p < 0.05, * p < 0.1. 26 /27 Table D3 Labor Market Transitions in Panel Data for 5 LAC Countries (percentage) Dominican Argentina Brazil Peru El Salvador Republic 2021–23 2022–23 2021–23 2021–23 2022–23 Labor market transitions Gaining job 4.6 6.9 4.2 5.6 5.8 Losing job 5.5 8.0 5.8 5.6 5.7 Occupational upgrading 8.4 4.5 4.6 7.8 6.2 Occupational downgrading 9.5 4.6 3.7 7.9 6.7 Conditional on falling into poverty Job gain 4.6 4.5 3.4 2.4 4.3 Job loss 14.4 27.9 19.4 12.3 17.2 Occupational upgrading 9.8 3.0 5.6 8.1 6.3 Occupational downgrading 9.2 5.0 5.3 9.2 6.9 Conditional on escaping poverty Job gain 16.5 23.9 13.2 11.0 13.2 Job loss 2.4 5.4 4.5 3.7 4.8 Occupational upgrading 10.5 4.7 8.2 9.3 11.8 Occupational downgrading 8.5 3.3 5.6 7.2 8.9 Conditional on falling into vulnerability Job gain 2.9 3.1 1.9 2.0 4.6 Job Loss 13.0 20.2 14.4 12.3 7.0 Occupational upgrading 6.1 4.0 6.2 6.4 3.6 Occupational downgrading 8.3 6.4 4.9 8.5 6.5 Conditional on going to middle class Job gain 9.4 16.6 10.5 11.6 10.5 Job loss 2.0 4.8 3.7 2.3 5.1 Occupational upgrading 10.3 6.4 7.6 6.7 8.3 Occupational downgrading 10.7 3.9 2.9 8.4 7.2 Source: Computations based on SEDLAC (CEDLAS and World Bank). Note: Calculations are based on one-year transition panels using 2021–23 data for Argentina, the Dominican Republic, El Salvador, and Peru. For Brazil, only the 2022–23 panel is used. Table D4 Informality, Social Protection and Employment (percentage) New Job Previous Job All Workers (Job gainers) (Job leavers) (Changers and non-changers) Job structure (5-country averages) Productive informality Formal 30.0 31.9 48.9 Informal 69.8 68.1 51.1 Legal informality Formal 18.9 24.1 41.7 Informal 81.1 75.7 58.3 27 /27 New Job Previous Job All Workers (Job gainers) (Job leavers) (Changers and non-changers) Job structure (5-country averages) Skill level Low skill 28.6 23.9 18.9 Medium skill 61.1 64.2 62.5 High skill 10.3 12.0 18.6 Employment type Employer 3.7 3.3 5.2 Salaried worker 46.2 46.8 55.0 Self-employed 47.6 47.6 37.9 Unpaid 2.5 2.3 1.9 Source: Computations based on SEDLAC (CEDLAS and World Bank). Note: Calculations are based on one-year transition panels using 2021–23 data for Argentina, Dominican Republic, El Salvador, and Peru. For Brazil, only the 2022–23 panel data are used. The productivity-based definition considers as informal those workers who have salaried jobs in small firms (with fewer than five employees), are self-employed without education beyond high school, or are unpaid family workers. Legal informality is defined as workers who do not have work-related pension insurance. For Argentina, these are salaried workers who do not receive work-related pension insurance and non- salaried workers without complete tertiary education. Table D5 ILO Occupational Classification Major Code Major Label Associated Skill Level 1 Managers 2 Professionals Skill levels 3 and 4 (high) 3 Technicians and Associate Professionals 4 Clerical Support Workers 5 Services And Sales Workers 6 Skilled Agricultural, Forestry and Fishery Workers Skill level 2 (medium) 7 Craft and Related Trades Workers 8 Plant and Machine Operators and Assemblers 9 Elementary Occupations Skill level 1 (low) Source: International Labour Organization (2025). Learn more: LAC EQUITY LAB