Poverty and inequality

Regional Poverty and Inequality UpdateLatin America and the Caribbean1

October 2026

Poverty and Equity Global Practice

Main Messages

  • Poverty fell to a record low in 2025 in Latin America and the Caribbean (LAC) amid a slower pace of poverty reduction and modest economic growth. An estimated 24 percent of people in LAC lived on less than $8.30 per day (2021 PPP). This share is projected to fall to 23.6 percent in 2026. The vulnerable, those above the $8.30 poverty line but living on less than $17 per day (2021 PPP), accounted for 31.4 percent of the population in 2025.
  • The middle class remained the region's largest group. The middle class, those living on $17 or more per day (2021 PPP), accounted for 44.5 percent of the population in 2025 and is projected to reach 45 percent in 2026.
  • Growth alone did not determine poverty outcomes, as countries with similar growth rates had different poverty changes. Mexico and Brazil achieved large poverty reductions despite modest growth over the 2017–25 period, while the Dominican Republic, El Salvador, and Honduras combined strong growth with substantial declines in poverty, and Argentina’s recent macroeconomic stabilization supported a poverty decline in 2025.
  • Inequality fell in the region, and in most countries, but remained high. The regional Gini index fell from 51.3 in 2017 to 49.3 in 2025, while 10 of 16 countries reduced both inequality and poverty over the same period.
  • Jobs and earnings have become the main engine of poverty reduction since 2022. Labor market gains accounted for about 76 percent of the poverty reduction between 2022 and 2025, double the share between 2017 and 2025 (about 37 percent). Public transfers and other non-labor income continue to play a key role in raising incomes among the poorest households.
  • A more educated, older workforce helped raise household incomes. Rising average earnings did not always translate into higher pay for comparable workers. The boost from a more educated workforce and favorable demographics may fade as the population ages.
  • Lasting poverty reduction will depend on more productive, better-paying jobs. Improving foundational learning and skills, raising productivity in sectors that employ low-income workers, maintaining targeted social protection, and expanding social insurance are priorities.

Poverty and inequality

2. Poverty Reduction, Middle-Class Expansion, and Persistent Vulnerability

LAC has the second-lowest poverty rate among developing regions, but East Asia and the Pacific (EAP) has nearly caught up. Only Europe and Central Asia (ECA) had a lower rate, at 4.3 percent in 2025 (refer to figure 2). EAP’s rate fell from 38.7 percent in 2017, when LAC’s rate was 32.6 percent, to 25.0 percent in 2025, just above LAC’s rate of 24.0 percent in the same year.4 During this same period, average GDP growth in EAP was 5.2 percent per year, compared with 1.8 percent in LAC. This growth gap partly explains the faster pace of poverty reduction in EAP. Among LAC countries, similar GDP growth rates were accompanied by different poverty outcomes (refer to appendix A), suggesting that how income reaches households matters alongside aggregate growth.

Figure 2. LAC has the second-lowest regional poverty rate
Poverty Rate, $8.30 per Day (2021 PPP), by World Region for Selected Years, 2017-25

Sources: SEDLAC (CEDLAS and World Bank) and World Bank Poverty and Inequality Platform (PIP).

Note:

Latin America and Caribbean 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. MENAAP stands for Middle East, North Africa, Afghanistan and Pakistan. For non-LAC regions, 2025 values are estimated by the PIP nowcasting model.

Poverty reduction over 2017–25 varied across LAC countries. From 2017 to 2025, poverty fell by 18.4 p.p. in El Salvador, 16.4 p.p. in Mexico, 13.0 p.p. in Honduras, and 9.8 p.p. in the Dominican Republic, compared with more than 8 p.p. regionwide (refer to figure 3). Despite progress, poverty in Honduras remained above 40 percent in 2025. Poverty also fell in Costa Rica (6 p.p.), although below the regional average. Peru recorded a modest reduction of 1.1 p.p. The trend was less favorable in Guatemala, Argentina, Ecuador, and Uruguay, which did not experience a net decline in poverty over the 2017–25 period. However, during the 2022–25 period, poverty declined in Argentina, Ecuador, and Uruguay (refer to appendix B). Guatemala couldn’t be assessed over this shorter period because its latest survey is from 2023.

Figure 3. Poverty fell in 12 of 16 LAC countries between circa 2017 and circa 2025, but at different speeds
Change in Poverty from circa 2017 to circa 2025

Source: SEDLAC (CEDLAS and World Bank).

Note:

LAC aggregate based on 18 countries with available SEDLAC microdata. These are World Bank harmonized estimates and do not correspond to official national figures. In cases where data were unavailable, values have been estimated using microsimulations and then pooled to create regional estimates. Andean region: Bolivia, Colombia, Ecuador, Peru. Central America: Costa Rica, Dominican Republic, El Salvador, Guatemala, Honduras, Nicaragua, Panama. Southern Cone: Argentina, Chile, Paraguay, Uruguay. ECH = Uruguay’s Continuous Household Survey (Spanish acronym); EPHC = Paraguay’s Continuous Permanent Household Survey (Spanish acronym).

a. Paraguay and Uruguay series are not strictly comparable between 2017 and 2025. Uruguay’s ECH became a rotating panel in 2022; Paraguay’s estimates for 2022 onward are based on the revised EPHC, which uses a new sampling frame and expansion factors derived from the 2022 Population Census, along with a revised valuation of in-kind transfers (Hambre Cero). Aggregates that depend on expanded totals, such as the wage bill and working-age population, are not comparable between 2017 and later years. Changes in rates and distributional indicators for 2017–25 should be interpreted with caution.

b. Data for Guatemala use 2014 as base year and 2023 as final year, the two most recent surveys available in SEDLAC.

c. In Peru, the national official poverty estimate is consumption based; the results shown here are based on the CEDLAS and World Bank harmonization based on income.

More Latin Americans joined the middle class, while one in three remains vulnerable to falling into poverty

LAC’s middle class continued expanding in 2025. The share of Latin Americans classified as middle class rose to 44.5 percent in 2025, from 34.4 percent in 2021, and is projected to increase further to 45 percent in 2026 (refer to figure 4).5 The income threshold separating vulnerability from middle-class status is estimated at $17 per person per day (2021 PPP), based on household panel data used to assess the risk of falling into poverty (World Bank 2025).

The share of the region’s population vulnerable to poverty has remained close to one-third since 2010. As the income distribution shifts toward higher incomes, movements from poverty into vulnerability and from vulnerability into the middle class can keep this share broadly stable, without implying that the same individuals remain vulnerable over time. In 2025, poverty and vulnerability together accounted for 55.4 percent of the population. In 2026, poverty is projected to fall to 23.6 percent and the middle class to expand to 45.0 percent. Yet the share of the vulnerable is expected to remain broadly unchanged at 31.4 percent, highlighting the challenge of translating poverty reduction into greater economic security.

Figure 4. Poverty reached historic lows in 2025 as the middle class continued expanding
Poverty, Vulnerability, and Middle-Class Trends in LAC, 2000-28 (2021 PPP)

Source: SEDLAC (CEDLAS and World Bank).

Note:

The LAC aggregate is based on 18 Latin American and Caribbean countries with available SEDLAC microdata. When data were unavailable, values were estimated using microsimulations and then pooled to create regional estimates. Due to methodological changes in Mexico’s official household survey in 2016, there is a break in the LAC-18 aggregate.

Poverty and inequality

3. Socioeconomic Profiles

Children remain disproportionately represented among the poor. In 2025, children under 15 accounted for 36 percent of the poor, compared with 22 percent of the total population. People ages 15–64 represented 58 percent of the poor versus 67 percent of the total population, and those ages 65 and older, 6 percent versus 11 percent (refer to table 1). By contrast, children accounted for only 13 percent of the middle class and older adults for 16 percent.

Educational attainment improved across income groups, but gaps remain. In 2025, 67 percent of the poor had low educational attainment, compared with 47 percent of the total population (refer to table 1). Between 2017 and 2025, this share fell by 5.2 p.p. among the poor but by 7.4 p.p. overall, widening the gap. Gains in educational attainment provide a necessary foundation for economic mobility, but they are not sufficient on their own to secure higher and more stable incomes.

Informal employment remains widespread among poor workers, limiting their access to secure earnings and employment-based social protection. Nearly 75 percent of poor workers held informal jobs in 2025, compared with 46 percent of all workers (refer to table 1). These shares were 55 percent among vulnerable workers and 33 percent among middle-class workers. Between 2017 and 2025, informality increased among poor and vulnerable workers, and remained broadly stable among middle-class workers, yet it declined overall due to a composition effect.

Salaried employment remains substantially less common among the poor than in the overall population. In 2025, salaried workers represented 40 percent of poor workers, compared with 62 percent of the overall population (refer to table 1). Between 2017 and 2025, the share of economically active individuals who were unemployed or engaged in unpaid work declined across all income groups. However, gains in salaried employment occurred only among the middle class, where the share of salaried workers increased from 67 percent to 69 percent. By contrast, the share remained broadly unchanged among the vulnerable and declined slightly among the poor, from 41.0 percent to 40.4 percent. Self-employment remained slightly more prevalent among the poor, although its share increased across all income groups during the period (refer to table 1).

Table 1. Children and informal workers are overrepresented among the poor

Source: SEDLAC (CEDLAS and World Bank).

Note:

LAC aggregate based on 18 Latin American and Caribbean countries with available microdata. In cases where data were unavailable, values have been estimated using microsimulations and then pooled to create regional estimates. Educational attainment categories cover the entire population. Low education: 0–8 years of completed education; middle education: 9–13 years of completed education; high education: 14 or more years of completed education. Informality is measured as the share of workers who do not contribute to a pension system.

Poverty and inequality

4. Inequality fell but remained high

Inequality fell between 2017 and 2025, but LAC continues to be one of the world’s most unequal regions (refer to figure 5).8 The regional Gini index declined from about 51 to 49 (refer to figure 6), still above the World Bank’s 40-point threshold for high inequality (World Bank 2026b). In 2025, the richest 20 percent of households received 54 percent of total income, while the poorest 20 percent received only 4 percent. Inequality increased in Colombia, Ecuador, El Salvador, Argentina, and Uruguay. In Argentina, inflation, which remained elevated through 2024, disproportionately hurt poorer households, widening the income gap prior to the 2025 stabilization and poverty decline. In Colombia, earnings rose across the wage distribution, but more strongly among higher-paid workers. In El Salvador, strong growth, rising international remittances, and substantial poverty reduction coincided with higher inequality (World Bank 2026a). International remittances accounted for more than 27.5 percent of El Salvador's GDP in 2025 (World Bank 2026c), and 24.2 percent of households received remittances. As remittance flows surged after the pandemic, they raised the non-labor income of recipient households relative to that of poorer, nonrecipient households, widening income gaps. Uruguay’s small increase should be interpreted with caution, given the break in its survey series.

Figure 5. LAC continues to rank among the most unequal regions in the world
Income Distribution (Share) within LAC and LAC vs. Other Regions, Circa 2025

Sources: SEDLAC (CEDLAS and World Bank) and Poverty and Inequality Platform (PIP).

Note:

Percentages are based on income (LAC) or consumption (non-LAC regions). The LAC bar is a pooled aggregate of 18 Latin American and Caribbean countries with available SEDLAC microdata, whereas non-LAC regional bars are simple averages of country-level shares. MENAAP stands for Middle East, North Africa, Afghanistan and Pakistan. Andean subregion: Bolivia, Colombia, Ecuador, Peru. Central America subregion: Costa Rica, Dominican Republic, El Salvador, Guatemala, Honduras, Nicaragua, Panama. Southern Cone Subregion: Argentina, Chile, Paraguay, Uruguay. Where data are unavailable, values are estimated using microsimulations before regional aggregation. Brazil’s 2025 data are preliminary. For Mexico, 2024 is the most recent year available.

Poverty fell in most countries, but not always alongside lower inequality. From 2017 to 2025, 10 of 16 countries recorded declines in both poverty and inequality. From 2022 to 2025, poverty fell in 14 of 15 countries and inequality fell in 9 of them (refer to appendix C). Honduras and Mexico combined some of the largest poverty declines with reductions in the Gini index. By contrast, Colombia experienced a more modest poverty decline with an increase in inequality. In Guatemala, poverty rose even as the Gini index fell.

Figure 6. Inequality fell in LAC and in most countries in the region
Change in Gini from circa 2017 to circa 2025

Source: SEDLAC (CEDLAS and World Bank).

Note:

LAC aggregate based on 18 Latin American and Caribbean countries with available SEDLAC microdata. These are World Bank harmonized estimates and do not correspond to official national figures. In cases where data were unavailable, values have been estimated using microsimulations and then pooled to create regional estimates. Andean subregion: Bolivia, Colombia, Ecuador, Peru. Central America subregion: Costa Rica, Dominican Republic, El Salvador, Guatemala, Honduras, Nicaragua, Panama. Southern Cone Subregion: Argentina, Chile, Paraguay, Uruguay. ECH = Uruguay’s Continuous Household Survey (Spanish acronym); EPHC = Paraguay’s Continuous Permanent Household Survey (Spanish acronym).

a. Paraguay and Uruguay series are not strictly comparable between 2017 and 2025. Uruguay’s ECH became a rotating panel in 2022; Paraguay’s estimates for 2022 onward are based on the revised EPHC, which uses a new sampling frame and expansion factors derived from the 2022 Population Census, along with a revised valuation of in-kind transfers (Hambre Cero). Aggregates that depend on expanded totals, such as the wage bill and working-age population, are not comparable between 2017 and later years. Changes in rates and distributional indicators for 2017–25 should be interpreted with caution.

b. Data for Guatemala use 2014 as base year and 2023 as final year, the two most recent surveys available in SEDLAC.

c. In Peru, official estimates are consumption-based; the results shown here are based on the CEDLAS and World Bank harmonization based on income.

The distribution and sources of income growth

Income grew fastest for the poorest households during the 2017–2025 period, helping reduce both poverty and regional inequality. Annualized per capita income rose by about 3.2 percent in the poorest decile, compared with 1.5 percent in the richest (refer to figure 7). This pattern was driven mainly by labor income, as it contributed between 1.2 and 1.9 p.p. across deciles, while public transfers and other non-labor income added about 1.8 p.p. in the poorest decile but only about 0.3 p.p. in the richest. Other non-labor income, including private transfers, imputed rent, capital income, and other nonclassifiable non-labor income, contributed an average of 0.4 p.p. across deciles.

Since 2022, labor income has driven faster, more evenly distributed income growth (refer to figure 7). In this period, annualized per capita income grew between 3.9 and 4.9 percent across deciles, with the largest gains around the middle of the distribution (4.8 percent in the fifth decile and 4.9 percent in the sixth). Unlike in the full 2017–2025 period, however, the poorest decile (4.5 percent) did not grow faster than the middle. The contribution of other non-labor income also increased, doubling to an average of 0.8 p.p. across deciles. The importance of this component increased, especially in the poorest decile, where its share of total income growth rose from 30 percent in 2017–2025 to 38 percent in 2022–2025.

Figure 7. Labor income drove faster and more evenly distributed income growth after 2022
a. LAC, 2017–25
b. LAC, 2022–25

Source: SEDLAC (CEDLAS and World Bank).

Note:

LAC aggregate based on 18 Latin American and Caribbean countries with available microdata. In cases where data were unavailable, values have been estimated using microsimulations and then pooled to create regional estimates. 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. Other non-labor income includes components such as private transfers, imputed rent, capital income, and other nonclassifiable non-labor income.

Poverty and inequality

5. Drivers of Poverty Changes

Labor market gains were the main individual contributor to poverty reduction in LAC, and their relative importance increased after 2022. During the 2017–2025 period, changes in employment and labor income accounted for about 37 percent of the region’s 8.4 p.p. decline in poverty (refer to figure 8)9, while nonlabor income sources together (public transfers, pensions, remittances, and other non-labor income) accounted for more than 53 percent, led by public transfers. Demographic change contributed the remaining 9 percent (with women making a meaningful contribution to poverty reduction, refer to box 2). By contrast, during the 2022–2025 period, the labor market share doubled, to about 76 percent of the 4.2 p.p. decline in poverty. In absolute terms, labor market gains over these three years were as large as those for the full 2017–25 period, suggesting that labor market losses from 2017 to 2022—a period that included the COVID-19 pandemic—were later recovered. Non-labor income contributed about 24 percent to poverty reduction, mainly through other non-labor income. Demographic change accounted for less than 1 percent.

Figure 8. Labor market changes drove poverty reduction in LAC, especially after 2022
Drivers of Poverty Changes, LAC, Circa 2017–25 and 2022–25

Source: SEDLAC (CEDLAS and World Bank).

Note:

LAC aggregate is based on 16 Latin American and Caribbean countries with available microdata; Nicaragua and Haiti are not included. Nicaragua is excluded because its most recent household survey dates from 2014. Estimates for later years are based on a predictive model and are not suitable for decomposition. Guatemala is excluded from the 2022–25 period due to data unavailability. “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.

At the country level, labor market gains were the main driver of poverty reduction in most countries, but differences among countries were meaningful. From 2017 to 2025, labor market gains accounted for most of the poverty reduction in the countries with the largest declines (El Salvador, Mexico, Honduras, and the Dominican Republic), as well as in Brazil and Paraguay (refer to figure 9). In Chile, Colombia, Costa Rica, and Peru, poverty reduction was driven mainly by non-labor income and demographic change, with pensions contributing notably to the declines in Colombia and Costa Rica. Where labor markets faltered, progress stalled. For instance, poverty rose in Argentina, Guatemala, and Uruguay, while in Bolivia and Panama non-labor income more than offset labor market losses. Between 2022 and 2025, labor market gains drove poverty reduction in most countries, accounting for most of the decline in Honduras, Colombia, Brazil, Mexico, Paraguay, and the Dominican Republic, whereas non-labor income accounted for more than 40 percent of the decline in El Salvador and Costa Rica (refer to appendix D). Poverty rose in Bolivia between 2022 and 2025, where labor market changes (a combination of changes in real labor incomes and employment) pushed it up. These differences across countries reveal the importance of country-specific diagnostics of the drivers and sustainability of poverty reduction.

Figure 9. Country-level labor market changes drove poverty reduction; where they faltered, progress stalled
Drivers of Poverty Changes, Countries and LAC, Circa 2017–25

Source: SEDLAC (CEDLAS and World Bank).

Note:

LAC aggregate is based on 16 Latin American and Caribbean countries with available microdata; Nicaragua and Haiti are not included. Nicaragua is excluded because its most recent household survey dates from 2014. Estimates for later years are based on a predictive model and are not suitable for decomposition. These World Bank harmonized estimates do not correspond to official national figures. In cases where data were unavailable, values have been estimated using microsimulations and then pooled to create regional estimates. “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. ECH = Uruguay’s Continuous Household Survey (Spanish acronym); EPHC = Paraguay’s Continuous Permanent Household Survey (Spanish acronym).

a. Paraguay and Uruguay series are not strictly comparable between 2017 and 2025. Uruguay’s ECH became a rotating panel in 2022; Paraguay’s estimates for 2022 onward are based on the revised EPHC, which uses a new sampling frame and expansion factors derived from the 2022 Population Census, along with a revised valuation of in-kind transfers (Hambre Cero). Aggregates that depend on expanded totals, such as the wage bill and working-age population, are not comparable between 2017 and later years. Changes in rates and distributional indicators for 2017–25 should be interpreted with caution.

b. Data for Guatemala use 2014 as base year and 2023 as final year, the two most recent surveys available in SEDLAC.

c. In Peru, the national official poverty estimate is consumption based; the results shown here are based on the CEDLAS and World Bank harmonization based on income.

Poverty and inequality

6. The Role of Demography and Education in Income Growth

LAC’s demographic transition is reshaping the age profile of its working-age population. Between 2017 and 2025, the share of people ages 15–39 in the overall working-age population, ages 15–64, declined by about 4 p.p., while the share of those ages 40–64 increased (refer to figure 10). The largest declines were among people ages 15–19 and 20–24, whose shares fell by 1.5 p.p. and 1.4 p.p., respectively, while the largest increase was among those ages 60–64 (1.3 p.p.).

The share of people ages 65 and older in LAC’s total population rose from 9 percent in 2017 to 11.4 percent in 2025. The largest increases were in Costa Rica (5.6 p.p.), Panama (3.9 p.p.), Honduras (3.8 p.p.), and Ecuador (3.5 p.p.). By 2025, the countries with the highest share of people ages 65 and older were Costa Rica (16.1 percent), Uruguay (15.4 percent), and Chile (14.0 percent). Together, these shifts point to an older workforce and, over time, rising old-age dependency.

Figure 10. Population aging, among the 15 to 64 years old, is advancing and educational attainment continues to rise
a. LAC age (5-year spans), ages 15–64, circa 2017 and circa 2025
b. LAC education, ages 15–64, circa 2017 and circa 2025

Source: SEDLAC (CEDLAS and World Bank).

Note:

Shares of the population ages 15–64 by five-year age group (panel a) and by educational attainment level (panel b, population ages 15–64). LAC is a population-weighted aggregate of countries with available microdata (refer to appendix G).

Alongside these demographic shifts, educational attainment increased. Between 2017 and 2025, the share of people ages 15–64 who had at least completed secondary education rose by 9.8 p.p. (refer to figure 10). The share who had completed tertiary education increased from 13.6 percent to 18.2 percent, while the share of all groups with less than completed secondary education fell, most notably for those with incomplete primary (from 17 percent to 11.6 percent).

The expanding groups are also those with the highest average labor income. Workers with high education earn about three times as much per month as those with low education, and workers ages 41–65 earn more than younger workers (refer to figure 11). Average monthly labor income per worker in LAC was nearly flat between 2017 and 2022, and then rose from $941 in 2017 to $1,052 in 2025 (2021 PPP). After 2022, the increase was generalized across women and men and across workers ages 15–24, 25–40, and 41–65. However, the changes differed by education level. From 2017 to 2025, average labor income rose by about 11 percent among workers with low education and 7 percent among those with medium level of education, but fell by 1.6 percent among highly educated workers, whose earnings dipped by 2022 and only partly recovered by 2025.

Figure 11. Labor earnings accelerated after 2022
Average labor income by demographic group in LAC, 2017, 2022, and 2025

Source: SEDLAC (CEDLAS and World Bank).

Note:

Mean monthly labor income per worker from all jobs, in 2021 PPP USD, for employed people ages 15–65. Educational attainment by years of education: low = 0-8 years; middle = 9-13 years; high = 14 or more years. LAC is a population-weighted aggregate of countries with available microdata (refer to appendix G). PPP = purchasing power parity; USD = United States dollar.

As the relative supply of educated workers grew, the wage premium associated with education declined, although returns to education remain meaningful (refer to figure 12). From 2017 to 2025, the number of highly educated individuals per low-educated person in the population ages 15–64 increased from 0.49 to 0.83, while the number of middle-educated individuals per low-educated person rose from 1.34 to 1.96. Over the same period, the average hourly earnings of highly educated workers fell from 3.19 times to 2.86 times those of low-educated workers and from 2.42 times to 2.27 times those of middle-educated workers. The earnings premium of middle-educated workers relative to low-educated workers also narrowed, from 1.32 to 1.26. Despite this compression, highly educated workers still earned nearly three times as much per hour as low-educated workers in 2025. Taken together, these trends suggest that educational upgrading contributed to rising average labor incomes by increasing the share of workers in higher-paying education groups, even as the earnings advantage associated with higher levels of education diminished. Persistent foundational learning gaps, as highlighted in PISA 2025, and weak demand for skilled labor among small, low-productivity firms with shallow technology adoption also weighed on returns to education (World Bank 2026a).

Figure 12. As the supply of relatively better educated workers rose, the education wage premium narrowed from 2017 to 2025
a. Relative supply of education groups
b. Wage premium by education group

Source: SEDLAC (CEDLAS and World Bank).

Note:

Low = 0–8 years (less than complete lower secondary); Middle = 9–13 years (complete lower secondary to complete upper secondary); High = 14+ years (more than upper secondary). Supply = share of each group in the population ages 15–64. Relative supply = ratio of shares. Wage premium = ratio of mean hourly earnings of the main job (wage converted to 2021 PPP). PPP = purchasing power parity.

Alongside the demographic and earnings changes, employment growth also contributed to the rise in total labor income. Overall, employment increased modestly, with larger gains among young people. Between the first quarters of 2017 and 2025, the employment rate rose from 57.1 percent to 58.9 percent, an increase of 1.8 percentage points (refer to figure 13). The gain came from lower unemployment (from 9.3 percent to 6.1 percent), as participation edged down slightly (from 63.0 percent to 62.7 percent). Moreover, low headline unemployment in LAC coexists with widespread informality and small-scale self-employment, masking persistent gaps in job quality (World Bank 2026a). Among young people, employment increased by 2.9 percentage points, from 38.6 percent to 41.5 percent. The youth unemployment rate fell from 20.4 percent to 13.8 percent, and the share of those not in employment, education, or training (NEET) declined from 23.0 percent to 19.1 percent. Even so, youth unemployment in 2025 remained more than twice the overall rate.

These changes in population, employment, and earnings add up to the total labor income the region generates, known as the wage bill. It can grow when there are more people of working age, when more of them have jobs, or when each worker earns more (refer to figure 14). From 2017 to 2025, the wage bill grew by about 2.8 percent per year driven by higher labor income per worker. It contributed 1.4 p.p. (half of the total), followed by the larger working-age population (0.9 p.p., about one-third), and the higher employment rate contributed the rest (0.5 p.p.) (refer to figure 14).10 During the 2022–25 period, wage bill growth accelerated to 5.9 percent per year, with labor income per worker contributing 4.3 p.p. (about 73 percent), while working-age population and employment rates contributed the remaining 1.6 p.p. Taken together, the two periods imply that the wage bill grew by only about 1 percent per year before 2022. The wage bill grew in most countries during 2022–25, but growth rates varied, reaching 9 percent per year or more in Costa Rica, Honduras, and the Dominican Republic, while the wage bill contracted in Bolivia.

Figure 13. Youth unemployment fell faster than total unemployment
a. Labor market rates, LAC
b. Youth labor market rates, LAC

Sources: LABLAC and SEDLAC (CEDLAS and World Bank).

Note:

Labor market rates for the population ages 15 and older (panel a) and for youth ages 15–24 (panel b). Employment and participation rates are shares of the population in each age group; unemployment rates are shares of the labor force; the NEET rate is the share of youth neither employed nor studying. LAC is an employment-weighted aggregate of 11 Latin America and Caribbean countries: Argentina, Bolivia, Brazil, Chile, Colombia, Costa Rica, Ecuador, Mexico, Panama, Peru, and Uruguay. Ecuador and Panama have annual data, so the same value is assigned to each quarter. Data for Argentina and Peru cover urban areas only (Lima and Callao for Peru). LABLAC = Labor Database for Latin America and the Caribbean; NEET = not in education, employment or training.

Part of the rise in labor income per worker reflects changes in the composition of the workforce. Average earnings can increase simply because the workforce has more educated workers or more people in their peak earning years, even if pay for similar workers stays the same. To separate these effects, figure 15 decomposes household per capita income growth into contributions from income sources (labor income, remittances, transfers, pensions, and other non-labor income such as imputed rent), employment rates, demographic composition (age and sex), and educational attainment, within age–sex–education groups.11 Box 3 presents the methodological details.

Figure 14. Rising labor incomes have become the main driver of wage bill growth in LAC
a. Shapley decomposition of wage bill growth, circa 2017–25
b. Shapley decomposition of wage bill growth, circa 2022–25

Source: SEDLAC (CEDLAS and World Bank).

Note:

The wage bill is the total monthly labor income of people ages 15–64 (2021 PPP USD). It equals the working-age population, times the employment rate, times mean labor income per worker. Its annual log growth is therefore the exact sum of the annual log growth of these three factors (Shapley decomposition). Bars show each factor’s contribution; diamonds show the annualized wage bill growth. LAC is a population-weighted aggregate of countries with available microdata (refer to appendix G). USD = United States dollars. ECH = Uruguay’s Continuous Household Survey (Spanish acronym); EPHC = Paraguay’s Continuous Permanent Household Survey (Spanish acronym).

a. Paraguay and Uruguay series are not strictly comparable between 2017 and 2025. Uruguay’s ECH became a rotating panel in 2022; Paraguay’s estimates for 2022 onward are based on the revised EPHC, which uses a new sampling frame and expansion factors derived from the 2022 Population Census, along with a revised valuation of in-kind transfers (Hambre Cero). Aggregates that depend on expanded totals, such as the wage bill and working-age population, are not comparable between 2017 and later years. Paraguay is excluded from panel a (2017-25 period) due to a series break. Changes in rates and distributional indicators for 2017–25 should be interpreted with caution.

b. Data for Guatemala use 2014 as base year and 2023 as final year, the two most recent surveys available in SEDLAC.

c. In Peru, the national official poverty estimate is consumption based; the results shown here are based on the CEDLAS and World Bank harmonization based on income.

d. Data for Panama and Honduras use 2023 as the base year for circa 2022.

Rising education and demographic shifts accounted for most of the growth in household income. Averaged across deciles, education explained about 54 percent of growth, and demography (age, sex) about 30 percent, over the 2017–25 period (refer to figure 15). For mean per capita income, which gives more weight to higher-income households, education accounted for about 70 percent of the annual increase of 2 percent, demographic shifts for about 32 percent, and income within groups for about 17 percent. The residual, net of a small positive contribution from employment rates, offset about 19 percent (refer to appendix F).

Sources of growth differed along the distribution (refer to figure 15). Education contributed a similar amount in every decile, about 1.2 p.p. to 1.5 p.p. per year. Its relative importance increased with income, from about 32 percent of the 3.8 percent growth in the poorest decile to about 95 percent of the 1.5 percent growth in the richest.12 It was the largest single contributor from the fourth decile upward. Demography also mattered more at the bottom, contributing about 1 p.p. per year in the poorest decile, against about 0.5 p.p. in the richest. In the two poorest deciles, however, income within groups was the largest contributor (about 2.6 p.p. and 1.7 p.p. per year, respectively). Its contribution faded to nearly zero in the top deciles.

Figure 15. Decomposing household income growth by decile, circa 2017–25

Source: SEDLAC (CEDLAS and World Bank).

Note:

Microsimulation following Bourguignon et al. (2005) and Bourguignon et al. (2008). Each component measures the change in household per capita income in a counterfactual scenario that changes only that factor to its value in the other year and holds all other factors constant. The factors are age–sex shares (demography), education shares within age–sex groups (education), employment rates, and labor income within age–sex–education groups. Each LAC decile is a population-weighted average of the corresponding national deciles, rather than a decile of a pooled regional distribution. Growth in each LAC decile therefore reflects growth within the corresponding national deciles and changes in countries’ population shares, which enter the residual. LAC is a population-weighted aggregate of countries with available microdata (refer to appendix G). Household per capita income is measured in monthly 2021 PPP USD. PPP = purchasing power parity. USD = United States dollar.

The boost from a more educated workforce and favorable demographics may fade as the population ages. Between 2017 and 2025, more people entered higher-earning working ages, and the share of children fell, raising average household income even though earnings of comparable workers grew slowly. As old-age dependency rises, this demographic boost could weaken, and sustaining income growth will depend more on earnings gains within groups.

Poverty and inequality

7. Conclusion

Poverty in LAC reached a record low in 2025, but the pace of progress is slowing. Labor income has become increasingly important to poverty reduction, supported by employment growth, educational upgrading, demographic change, and more recently, higher earnings. Inequality fell, but LAC remained one of the world's most unequal regions.

Labor market gains supported poverty reduction but have not reached every household. From 2022 to 2025, jobs and earnings drove much of the region’s income growth and poverty reduction, but in the poorest decile, transfers and other non-labor income contributed about as much as labor earnings to income growth. Informal work also remains widespread among poor workers, with about three in four holding informal jobs in 2025, and the share of salaried workers among the poor has not increased. Social protection therefore remains essential.

Much of the rise in total labor income, or the wage bill, between 2017 and 2025 came from higher labor income per worker. About one third of total labor income growth came from a larger working-age population, and the rest reflected a higher employment rate. From 2022 to 2025, labor income per worker became more important for wage bill growth, accounting for more than 70 percent.

Earnings growth within education groups was modest, and the earnings advantage of highly educated workers narrowed. However, as the share of highly educated workers rose, the average labor income in the region increased across demographic groups. Changes in the age and sex structure of the population also supported growth, accounting for about 30 percent of household per capita income growth. As population ages and old-age dependency rises, productivity gains will matter more for sustained income growth.

Sustained poverty reduction requires higher productivity and better-paying work, alongside targeted social protection. Immediate support is particularly important for poor families with children, who remain overrepresented among the poor. Over time, raising labor productivity where low-income workers already work, easing moves to better jobs, and reducing barriers faced by women and young people can broaden progress. Better foundational learning, skills, and conditions for productive private investment and firm technology adoption will help turn gains in educational attainment into sustained earnings growth (World Bank 2026a).

Poverty and inequality

8. Appendixes

Appendix A. Similar growth rates, different poverty outcomes

At the country level, similar growth rates were accompanied by very different poverty outcomes in both the 2017–25 and 2022–25 periods. Figure A.1 plots annualized real GDP growth in the horizontal-axis against poverty reductions (in p.p.) in the vertical-axis, for 2017–25 (refer to panel a) and 2022–25 (refer to panel b). Higher growth was weakly associated with larger poverty reductions, but with considerable variation among countries. In 2017–25 period, Mexico (16.4 p.p.) and El Salvador (18.4 p.p.) recorded substantial poverty declines at growth rates above the region’s average, about 0.9 percent and 2.6 percent per year, respectively. Peru (1.1 p.p.), Chile (4.5 p.p.), and Brazil (9.8 p.p.) grew at similar rates of about 2 percent per year but had very different outcomes. Costa Rica and Guatemala both grew about 3.6 percent per year; poverty fell by about 6 p.p. in Costa Rica and increased by about 2.5 p.p. in Guatemala.

Between 2022 and 2025, the pattern in which poverty fell was far from uniform among countries, although the variation was less than in the full 2017–25 period. Honduras (9.9 p.p.), El Salvador (8.7 p.p.), and Costa Rica (7.4 p.p.) recorded the largest reductions, with growth of about 3.3 percent to 4.5 percent. Colombia and Mexico, however, achieved reductions of about 6 p.p. to 7 p.p. with growth below 2 percent. By contrast, poverty remained broadly unchanged in Panama, despite growth of about 4.7 percent—a rate similar to Costa Rica's, while poverty increased in Bolivia. This pattern suggests that the distribution of growth and the income channels reaching households matter alongside aggregate output.

Figure A.1. Similar growth rates led to different poverty outcomes in both 2017–25 and 2022–25
a. 2017–25: Wide dispersion at every level of growth
GDP Growth and Poverty Reduction, circa 2017-25
b. 2022–25: Moderate dispersion persisted
GDP Growth and Poverty Reduction, circa 2022-25

Sources: SEDLAC (CEDLAS and World Bank) and World Development Indicators (WDI).

Note:

LAC aggregate based on 18 Latin American and Caribbean countries with available SEDLAC microdata. In cases where data were unavailable, values have been estimated using microsimulations. 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. ECH = Uruguay’s Continuous Household Survey (Spanish acronym); EPHC = Paraguay’s Continuous Permanent Household Survey (Spanish acronym).

a. Paraguay and Uruguay series are not strictly comparable between 2017 and 2025. Paraguay’s EPHC uses a new sampling frame and expansion factors derived from the 2022 Population Census; Uruguay’s ECH became a rotating panel in 2022.

b. For Guatemala, the comparison uses 2014 as base year and 2023 as final year, corresponding to the most recent surveys available in SEDLAC.

c. In Peru, the national official poverty estimate is consumption based; the results shown here are based on the CEDLAS and World Bank harmonization based on income.

Appendix B. Change in poverty between 2022 and 2025

Between 2022 and 2025, poverty declined in 14 of 15 countries, in all three subregions, and in the LAC region (refer to figure B.2). Honduras, El Salvador, and Costa Rica recorded the largest reductions, although their starting poverty levels differed considerably (about 53 percent, 33 percent, and 17 percent, respectively). Poverty fell by 9.9 p.p. in Honduras, 8.7 p.p. in El Salvador, and 7.4 p.p. in Costa Rica, where it dropped by about two-fifths. Colombia, Mexico, the Dominican Republic, Brazil, and Paraguay also recorded declines above the regional average. Reductions were more modest in Peru and Ecuador, and in Chile and Uruguay, where poverty was already the lowest in the region. Poverty also fell in Argentina, Ecuador, and Uruguay, the three countries in this sample where it had risen during the 2017–25 period, although only Argentina's decline was sizable. By contrast, despite a reduction in the 2017–25 period overall, poverty increased in Bolivia (by 2.4 p.p.) between 2022 and 2025.

Figure B.2. Poverty declined in 14 of 15 LAC countries between circa 2022 and circa 2025
Change in Poverty from circa 2022 to circa 2025

Source: SEDLAC (CEDLAS and World Bank).

Note:

LAC aggregate based on 18 Latin American and Caribbean countries with available SEDLAC microdata. These are World Bank harmonized estimates and do not correspond to official national figures. In cases where data were unavailable, values have been estimated using microsimulations and then pooled to create regional estimates. Andean subregion: Bolivia, Colombia, Ecuador, Peru. Central America subregion: Costa Rica, Dominican Republic, El Salvador, Guatemala, Honduras, Nicaragua, Panama. Southern Cone Subregion: Argentina, Chile, Paraguay, Uruguay. ECH = Uruguay’s Continuous Household Survey (Spanish acronym); EPHC = Paraguay’s Continuous Permanent Household Survey (Spanish acronym).

a. Paraguay and Uruguay series are not strictly comparable between 2017 and 2025. Uruguay’s ECH became a rotating panel in 2022; Paraguay’s estimates for 2022 onward are based on the revised EPHC, which uses a new sampling frame and expansion factors derived from the 2022 Population Census, along with a revised valuation of in-kind transfers (Hambre Cero). Aggregates that depend on expanded totals, such as the wage bill and working-age population, are not comparable between 2017 and later years. Changes in rates and distributional indicators for 2017–25 should be interpreted with caution.

c. In Peru, the national official poverty estimate is consumption-based; the results shown here are based on the CEDLAS and World Bank harmonization based on income.

Appendix C. Comparing poverty and inequality reduction in the 2017–25 and 2022–25 periods

Figure C.1. Poverty and inequality fell together in most countries in both periods
a. Poverty and inequality fell together in 10 of 16 countries between circa 2017 and circa 2025
Poverty and Gini Reductions, circa 2017-25
b. Poverty and inequality fell together in 9 of 15 countries between circa 2022 and circa 2025
Poverty and Gini Reductions, circa 2022-25

Source: SEDLAC (CEDLAS and World Bank).

Note:

The LAC aggregate is based on 18 Latin American and Caribbean countries with available SEDLAC microdata. Where data are unavailable, values are estimated using microsimulations. Poverty reduction equals the initial poverty rate minus the final rate, expressed in percentage points, using the international poverty line of $8.30 per person per day (2021 PPP). Gini reduction equals the initial Gini index minus the final index, expressed in Gini points on a 0–100 scale. Positive values indicate reductions; negative values indicate increases. The panels report cumulative, nonannualized changes over circa 2017–25 and circa 2022–25, respectively. Period lengths and axis ranges differ, so visual distances are not directly comparable across panels. PPP = purchasing power parity. ECH = Uruguay’s Continuous Household Survey (Spanish acronym); EPHC = Paraguay’s Continuous Permanent Household Survey (Spanish acronym).

a. Paraguay’s and Uruguay’s series have comparability limitations over 2017–25. (Paraguay’s EPHC weights were revised using the 2022 Population Census, and Uruguay’s ECH adopted a rotating panel in 2022.)

b. For Guatemala, the comparison uses 2014 as base year and 2023 as final year, corresponding to the most recent surveys available in SEDLAC.

c. In Peru, the national official poverty estimate is consumption based; the results shown here are based on the CEDLAS and World Bank harmonization based on income.

Appendix D. Drivers of poverty change by country, 2022–25

Figure D.1. Labor market gains drove poverty reduction in most countries after 2022
Drivers of Poverty Changes, Countries and LAC, Circa 2022–25

Source: SEDLAC (CEDLAS and World Bank).

Note:

LAC aggregate is based on 16 Latin American and Caribbean countries with available microdata; Nicaragua and Haiti are not included. Nicaragua is excluded because its most recent household survey dates from 2014. Estimates for later years are based on a predictive model and are not suitable for decomposition. Argentina, Bolivia, Brazil, Colombia, Costa Rica, Ecuador, and El Salvador (2022–25); Panama, and Honduras (2023–25); Chile, Dominican Republic, Mexico, Paraguay, Peru, and Uruguay (2022–24). Microdata for 2025 are preliminary. “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. ECH = Uruguay’s Continuous Household Survey (Spanish acronym); EPHC = Paraguay’s Continuous Permanent Household Survey (Spanish acronym).

a. Paraguay and Uruguay series are not strictly comparable between 2017 and 2025. Uruguay’s ECH became a rotating panel in 2022; Paraguay’s estimates for 2022 onward are based on the revised EPHC, which uses a new sampling frame and expansion factors derived from the 2022 Population Census, along with a revised valuation of in-kind transfers (Hambre Cero). Aggregates that depend on expanded totals, such as the wage bill and working-age population, are not comparable between 2017 and later years. Changes in rates and distributional indicators for 2017–25 should be interpreted with caution.

c. In Peru, the national official poverty estimate is consumption based; the results shown here are based on the CEDLAS and World Bank harmonization based on income.

Appendix E. Educational upgrading and demographic transition, 2017–25

Figure E.1. The LAC population became more educated and older, a tailwind for income growth
a. Educational upgrading in LAC, circa 2017–25
b. Demographic transition in LAC, circa 2017–25
c. Change in the share of the population ages 65 and older in LAC countries, circa 2017–25

Source: SEDLAC (CEDLAS and World Bank).

Note:

Data for people ages 15–64. The LAC is a population-weighted aggregate of countries with available microdata (refer to appendix G).

a. Paraguay and Uruguay series are not strictly comparable between 2017 and 2025. Uruguay’s ECH became a rotating panel in 2022; Paraguay’s estimates partly reflect the reweighting that occurred in 2022 in the revised EPHC, which uses a new sampling frame and expansion factors derived from the 2022 Population Census, along with a revised valuation of in-kind transfers (Hambre Cero).

b. Data for Guatemala use 2014 as base year and 2023 as final year, the two most recent surveys available in SEDLAC.

c. In Peru, the national official poverty estimate is consumption based; the results shown here are based on the CEDLAS and World Bank harmonization based on income.

Appendix F. Contributions to mean household per capita income growth by country

Figure F.1. Decomposing household income growth: composition effects (demography, education) versus income gains within groups, LAC region and LAC countries, circa 2017–25.

Source: SEDLAC (CEDLAS and World Bank).

Note:

Microsimulation following Bourguignon et al. (2005) and Bourguignon et al. (2008). Each component measures the change in mean household per capita income in a counterfactual scenario that changes only that factor to its value in the other year and holds all other factors constant. The factors are age–sex shares (demography), education shares within age–sex groups (education), employment rates, and labor income within age–sex–education groups. Household per capita income is measured in monthly 2021 PPP US dollars. PPP = purchasing power parity.

The LAC mean for Latin America and the Caribbean is a population-weighted average of national means. Growth in the LAC mean therefore reflects growth in national means and changes in countries’ population shares, which enter the residual. LAC is a population-weighted aggregate of countries with available microdata (refer to appendix G).

a. Paraguay and Uruguay series are not strictly comparable between 2017 and 2025. Uruguay’s ECH became a rotating panel in 2022; Paraguay’s estimates for 2022 onward are based on the revised EPHC, which uses a new sampling frame and expansion factors derived from the 2022 Population Census, along with a revised valuation of in-kind transfers (Hambre Cero).

b. Data for Guatemala use 2014 as base year and 2023 as final year, the two most recent surveys available in SEDLAC.

c. In Peru, the national official poverty estimate is consumption based; the results shown here are based on the CEDLAS and World Bank harmonization based on income.

Appendix G. Country data availability in SEDLAC

Table G.1. Reference years for circa period
CountryCirca 2017Circa 2022Circa 2025CoverageIn LAC aggregateComment
Argentina201720222025UrbanYes
Bolivia201720222025NationalYes
Brazil201720222025NationalYes
Chile201720222024NationalYes
Colombia201720222025NationalYes
Costa Rica201720222025NationalYes
Dominican Republic201720222024NationalYes
El Salvador201720222025NationalYes
Ecuador201720222025NationalYes
Guatemala20142023NationalNoOnly the 2014 and 2023 surveys are available.
Honduras201720232025NationalYes
Mexico201620222024NationalYes
Panama201720232025NationalYes
Paraguay201720222024NationalYesThe EPHC was introduced in 2017 alongside the EPH. A break in comparability occurs in 2022 due to the transition to annual EPHC estimates and revised survey weights.
Peru201720222024NationalYes
Uruguay201720222024NationalYesECH became a rotating panel in 2022.

Source: SEDLAC (CEDLAS and World Bank).

Note:

ECH = Encuesta Continua de Hogares (Continuous Household Survey, Uruguay); EPHC = Encuesta Permanente de Hogares Continua (Continuous Permanent Household Survey, Paraguay); LAC = Latin America and the Caribbean.

Poverty and inequality

References

  • Aguilar-Gomez, Sandra, Jacobus De Hoop, Hugo Ñopo, Julian Pedrazzi, and Javier Romero. forthcoming. “Women Lead the Way: A New Era of Poverty Reduction in Latin America and the Caribbean”.
  • Bourguignon, François, Francisco H. G. Ferreira, and Philippe G. Leite. 2008. “Beyond Oaxaca–Blinder: Accounting for Differences in Household Income Distributions.” Journal of Economic Inequality 6 (2): 117–48.
  • Bourguignon, François, Francisco H. G. Ferreira, and Nora Lustig, eds. 2005. “The Microeconomics of Income Distribution Dynamics in East Asia and Latin America”. World Bank and Oxford University Press.
  • Deville, Jean-Claude, and Carl-Erik Särndal. 1992. “Calibration Estimators in Survey Sampling.” Journal of the American Statistical Association 87 (418): 376–82.
  • DiNardo, John, Nicole M. Fortin, and Thomas Lemieux. 1996. “Labor Market Institutions and the Distribution of Wages, 1973–1992: A Semiparametric Approach.” Econometrica 64 (5): 1001–44.
  • Hovhannisyan, Shoghik, Veronica Montalva, Tyler Remick, Carlos Rodriguez-Castelán, and Kersten Stamm. 2022. Global Job Quality: Evidence from Wage Employment across Developing Countries. Washington, DC: World Bank Group.
  • Shorrocks, Anthony F. 2013. “Decomposition Procedures for Distributional Analysis: A Unified Framework Based on the Shapley Value.” Journal of Economic Inequality 11 (1): 99–126.
  • World Bank. 2021. Regional Poverty and Inequality Report: The Gradual Rise and Rapid Decline of the Middle Class in Latin America and the Caribbean (May 2021). World Bank Group.
  • World Bank. 2024. Regional Poverty and Inequality Update Latin America and the Caribbean (October 2024). World Bank Group.
  • World Bank. 2025. Regional Poverty and Inequality Update: Latin America and the Caribbean. World Bank.
  • World Bank. 2026a. Harnessing the AI Transition. Latin America and the Caribbean Economic Update (October 2026). World Bank.
  • World Bank. 2026b. Poverty, Prosperity, and the Planet Report 2026: Jobs at the Center of Poverty Eradication. World Bank Group.
  • World Bank. 2026c. World Bank Open Data. [Online]. Available at: https://data.worldbank.org/country/el-salvador [Accessed 7 October 2026].
  1. This update summarizes the main trends of poverty and inequality in Latin America in the period 2017–25 using harmonized surveys from the Socio-Economic Database for Latin America and the Caribbean (SEDLAC) and the Labor Database for Latin America and the Caribbean (LABLAC) created by the World Bank and the Centro de Estudios Distributivos, Laborales y Sociales (CEDLAS; Center for Distributive, Labor and Social Studies). See appendix G for country-specific periods. This update was produced by the Poverty and Equity Global Practice in the Latin America and Caribbean Region of the World Bank. The core team included Luis Eduardo Castellanos Rodriguez, Hector Daniel Segura Jimenez, and Diana Sanchez Castro, under the leadership of Eduardo A. Malásquez, Lourdes Rodriguez Chamussy and the guidance of Carlos Rodriguez Castelán. Box 1 was prepared by Isabella Contreras and Jonathan Lain. Box 2 is based on Sandra Aguilar-Gomez, Jacobus De Hoop, Hugo Ñopo, Julian Pierino Pedrazzi and Javier Romero (forthcoming). The team thanks Sofia Bell for her valuable contributions to the final version of the document, as well as the staff from the LAC Poverty team for their insightful comments. Contact: lac_stats@worldbank.org. Most of the data featured in this brief can be found at the LAC Equity Lab. ↩
  2. The decline in poverty from 33 percent in 2017 to 24 percent in 2025 corresponds to about 41 million fewer people living in poverty in the region. ↩
  3. Income thresholds for LAC are as follows: poor (less than $8.30 per day, 2021 PPP); vulnerable ($8.30 per day to $17 per day); middle class and above ($17 per day and over) (World Bank 2025). ↩
  4. The LAC aggregate is based on 18 countries in the region for which SEDLAC microdata is available. In cases where data were unavailable, values have been estimated using microsimulations and then pooled to create regional estimates. ↩
  5. Due to methodological changes in Mexico’s official household survey in 2016, there is a break in the LAC-18 aggregate. ↩
  6. The World Bank has harmonized household survey data from six Caribbean countries, five of which have information on household consumption and employment: Barbados, Grenada, Jamaica, Saint Lucia, and Suriname. The analysis does not cover Haiti. ↩
  7. The JQI is based on Hovhannisyan et al. (2022) and tracks four key dimensions: income, benefits (based on health insurance or retirement benefits), satisfaction (based on whether people have a second job), and security (based on tenure and contractual arrangements). Satisfaction refers to whether the worker is satisfied with his or her job. Job satisfaction is likely lower among workers with two jobs, assigning them a value of zero for this dimension. The analysis focuses on benefits, satisfaction, and security because welfare is measured using consumption rather than income in the selected Caribbean countries, so income data are less comprehensive and reliable. ↩
  8. International comparisons warrant caution because LAC measures inequality using income, whereas most other regions use consumption, which typically yields lower inequality estimates. ↩
  9. For this decomposition, the LAC aggregate is based on 16 countries with available microdata. ↩
  10. The analysis uses a Shapley decomposition to separate growth in the wage bill (that is, total labor income, including self-employment) into three components: changes in working-age population, employment rate, and average earnings per worker. ↩
  11. Figure 15 presents a decomposition using counterfactual microsimulations following Bourguignon, Ferreira, and Lustig (2005) and Bourguignon, Ferreira, and Leite (2008). ↩
  12. Growth at decile level might differ from figure 7, as the countries with available information for implementing the Bourguignon-type decomposition differ from the ones used to implement the Shapley decomposition. ↩