This chapter examines the progress made by Latin America and the Caribbean (LAC) in educational attainment over recent decades, as well as performance on the latest PISA assessment. Findings indicate that the performance of the LAC region remains lower than that of OECD countries, even though the socio-economic gap in the region is narrower in comparison to the OECD average. Consequently, what may appear as a reduced social divide is likely more indicative of the overall low quality of education systems in LAC. The chapter concludes by comparing investment levels in education within the region to those of other countries, revealing that some countries significantly outperform LAC despite similar spending levels. It emphasises that upward educational and social mobility in LAC is achievable through greater and more efficient resource allocation in education.
Social Mobility and Inequality in Latin America and the Caribbean
Abstract
Inequality in Latin America and the Caribbean (LAC), unlike other regions, is characterised by a significant concentration of wealth, a substantial gap between the wealthy and the middle class, and access to quality services limited to a small, affluent group at the top of the income distribution (Torche, 2014[1]). Education stands out as the most critical service for addressing these opportunity gaps. Expanding access to high-quality education - regardless of income or socio-economic background - would benefit not only individuals but also society as a whole.
Understanding the distinction between social inequality, equity in education, and social mobility is essential in this context. Social inequality refers to the uneven distribution of resources, while equity in education emphasises equal access to quality learning opportunities for all, regardless of background. Social mobility, on the other hand, is about ensuring that an individual’s socio-economic future is determined by their skills and efforts rather than their starting point in life. In LAC countries, both high inequality and low social mobility coexist, significantly limiting opportunities for those from disadvantaged backgrounds.
In highly unequal environments, limited access to quality education results in fewer skills, reduced labour market opportunities, and lower contributions to national productivity, economic growth, and tax revenues. For individuals, lower educational attainment is associated with worse material conditions throughout life, including higher poverty rates, lower employment prospects, increased informal employment, and lower earnings. Over the past decade, labour market polarisation due to digitalisation has exacerbated these disparities, particularly for those with an upper secondary education or less, as demand for middle-skill jobs has declined.
As highlighted in Chapter 2, the predominant pattern in educational mobility across Latin America and the Caribbean is stagnation. While the region has made significant strides in expanding primary and secondary education coverage over the past few decades, particularly benefiting households with low educational attainment, progress has been uneven. Even today, more than half of adults aged 25 and older in LAC have not completed upper secondary education, compared to less than 30% in OECD countries (OECD, 2021[2]).
Additionally, despite improvements in access observed in the first two decades of the 21st century, high dropout rates persist and remain one of LAC’s most pressing challenges (ECLAC, 2024[3]). Dropout rates are strongly correlated with socioeconomic status, with students from lower-income families more likely to leave school early to enter the workforce, hindering their pursuit of higher education and limiting their skill development. Alarmingly, dropout rates have risen further in the aftermath of the COVID-19 pandemic (ECLAC, UNESCO and UNICEF, 2022[4]).
However, it is important to note that educational attainment - often measured by years spent in school - does not necessarily reflect actual skill levels. This metric fails to provide a clear picture of cognitive abilities and competencies or how they compare to other countries. To address this gap, the OECD introduced the Programme for International Student Assessment (PISA) in 2000. PISA offers an internationally comparable measure of student performance, assessing 15-year-olds in areas like mathematics, science, and reading. It links data from students, teachers, schools, and education systems to analyse disparities in educational outcomes (OECD, 2023[5]).
The latest assessment was implemented in 2022, with a focus on mathematics. It was the first assessment to take place after the COVID-19 pandemic, when LAC had the longest school closures in the world with an average of 72 weeks (ECLAC, 2023[6]). This version of the assessment recorded participation from 14 LAC countries and almost 100,000 students in the region.
PISA 2022 revealed that, despite progress in the region, LAC students underperform in mathematics, reading and science when compared to the rest of the world. The assessment also revealed differences in performance due to socioeconomic status (a measure that includes parents’ educational attainment), in line with the high inequality levels observed in the region. But, more importantly, it revealed the generalised low performance across different groups when compared to OECD countries and lack of baseline proficiency among students. This is particularly relevant for the future of the region, as research indicates that PISA results are linked to success indicators, such as completing a tertiary degree or being in the labour force by age 25 - all indicators that correlate with economic growth (OECD, 2019[7]).
The LAC region’s limited access to high-quality education across socio-economic backgrounds has direct economic consequences, given the established link between cognitive skills and growth. For instance, improving PISA scores by 25 points could result in a 3% GDP increase (OECD, 2010[8]). However, only Peru (+26 pts), Colombia (+13 pts), and Brazil (+23 pts) have shown consistent gains from their initial assessments to the 2022 evaluation. Education's potential to drive social mobility is diminished when it reinforces rather than reduces income inequality.
Effective education systems are therefore essential for a skilled workforce. A skilled workforce promotes innovation, boosts productivity, and demands for better jobs (OECD, 2018[9]). Educational outcomes are strongly tied to various measures of well-being in adulthood, such as poverty rates, employment, and income. If educational success is independent of socio-economic background, it fosters social mobility and enables talented individuals even from disadvantaged backgrounds to reach their full potential. In other words, when success in education is not determined by socio-economic background or parental education, education becomes a key driver of both economic growth and social mobility (OECD, 2018[10]). A more inclusive and prosperous future for LAC depends on expanding access to high-quality education for all, which can promote social mobility and reduce entrenched inequities that have constrained the region's development for decades.
Key challenges, including low upper-secondary completion rates and high dropout rates, must be addressed. The COVID-19 pandemic worsened existing inequalities, and the growing demand for skilled workers due to digitalisation heightens the urgency. Leaders in the region must prioritise equal distribution of high cognitive skills to stimulate growth.
Education is undeniably one of the key drivers of upward mobility but unequal access - because of socio-economic status and financial barriers - can undermine public efforts. The following section examines current trends in educational attainment across the region, followed by an analysis of the 2022 PISA results in Latin America and the Caribbean, highlighting variations by socio-economic background which is reflective of parental education. The chapter concludes with a comparative review of regional investment in education relative to OECD countries.
In terms of educational attainment, Latin America and the Caribbean has experienced important progress. In primary education, most countries in the region have achieved universal coverage. The net enrolment rate - reflecting the proportion of students enrolled at the appropriate educational level for their age - for primary education in the region stands at 97.1%, similar to 98.9% in OECD member countries. However, 9.6 million children remained out of school in 2022 - surpassing the historical low of 2015 (UNESCO, 2024[11]). More importantly, as Figure 3.1 indicates, the proportion of out-of-school children as a share of the corresponding age group increases with age (OECD, 2022[12]).
Note: 2019/2020 or latest available. Data are not available for Primary school age in Peru. The LAC average refers to Argentina, Brazil, Chile, Colombia, Costa Rica, Dominican Republic, Ecuador, Mexico, Paraguay, Peru, and Uruguay where available.
Source: Gender Equality in Peru: Towards a Better Sharing of Paid and Unpaid Work
Despite policy reforms aimed at increasing compulsory education duration, upper secondary education coverage in LAC lags behind that of OECD countries (OECD, 2023[13]). The net enrolment rate in the region is 78.14%, 14 percentage points lower than the OECD average of 92.8% (Figure 3.2). In the same vein, completion rates for secondary education are 15 percentage points behind the OECD average, standing at 65% in LAC compared to 80% in OECD countries (Arias Ortiz et al., 2024[14]).
Worryingly, strides in educational attainment began decelerating before 2020. Experts note a decline in effective policies to reduce school dropout rates, which remain a significant challenge across Latin America and the Caribbean (ECLAC, UNESCO, UNICEF, 2024[15]). This issue is particularly critical for social mobility, which is influenced by various interconnected factors such as income levels, rural residency, indigenous status, and more importantly, parental education (Figure 3.3). Evidence from Argentina indicates that dropout rates are heightened by income shocks within households, with the impact being more pronounced in families with lower-educated earners and limited resources (OECD, 2017[16]).
As a result, exclusion rates show a strong negative correlation with household income. Among those of upper-secondary school age, 17.7% of students living in extreme poverty in urban areas and 25.2% in rural areas are out of school. In contrast, only 3.2% of high-income students in urban areas and 14.8% in rural areas face similar exclusion (ECLAC, 2024[3]).
Disparities within the education system manifest early, becoming particularly pronounced at the secondary level through dropout rates. When students leave the system, they forfeit the opportunity to pursue higher education, which exacerbates existing inequalities by limiting their job prospects. Those who do not complete secondary education are more likely to experience unemployment, live in poverty, and become single parents. At the aggregate level, dropout limits the skills of the country’s workforce and the share of highly productive workers who would contribute more through tax revenues if they had completed secondary education (Lyche, 2010[17]; OECD, 2017[16]).
Given these impacts, policies aimed at reducing school dropout should be viewed as investments rather than expenses. Notably, dropout rates can often be predicted based on demographic characteristics and early warning signs, such as absenteeism and grade repetition, making preventive measures feasible.
Note: ISO3 country codes. Total years of compulsory education can include pre-primary education, as is the case in El Salvador. Countries ranked in descending order of total years of compulsory education.
Source: (UNESCO, 2020), Global education monitoring report, 2020: Inclusion and education: all means all.
Countries in Latin America can promote inclusive growth by implementing policies to both prevent school dropout and reintegrate children and youths who are already out of the system. Intuitively, these measures should vary according to educational level. At the primary education level, the focus should be on ensuring a smooth transition into the educational system by using pre-emptive measures such as involving parents, addressing problematic behaviour, and providing therapeutic programs. At the lower secondary level (the last two years of primary in systems that extend from first to eighth grade), it is essential to identify at-risk students to facilitate their transition to upper secondary education. This group often includes students who have repeated grades, with Latin America exhibiting some of the highest grade repetition rates in PISA assessments – over 40% of students in Colombia have reported repeating at least one grade. Research links grade repetition to school disengagement, contributing to increased dropout rates; thus, policymakers should consider strategies to reduce this practice.
Another relevant measure is introducing substance abuse topics at this age, which, despite not having a direct impact on school dropout, can help to create positive attitudes towards the school environment. Encouraging, volunteering, and fostering connections between schools and the workplace can help reduce dropout rates. At the upper-secondary level, ensuring completion becomes more challenging if these school years are not made mandatory. However, motivating underperforming students through engaging tasks and vocational training opportunities can improve graduation rates. (Lyche, 2010[17])
Following secondary education, tertiary education plays a crucial role in enhancing individuals' living conditions and acts as a powerful driver of social mobility. Additionally, the financial benefits of obtaining a tertiary degree remain robust in Latin America). Graduates enjoy enhanced job prospects and higher salaries, while at a broader level, increased educational attainment drives growth and productivity through a more skilled workforce (OECD, 2016[18]; OECD/CAF/ECLAC, 2016[19]).
Note: The bars represent the estimated child’s average probability of achieving a higher level of education than his/her parents’ educational attainment, except for “complete tertiary” where it represents the probability of achieving the same level.
Latinobarómetro is a public opinion survey conducted annually, with approximately 20,000 interviews across 18 Latin American countries, representing over 600 million inhabitants.
The sample are people aged between 25 and 44 years at the time of the survey.
Source: Latinobarómetro 2023.
25-64 year-olds with income from employment (full-time full-year workers); upper secondary education = 100
Note: The statistical data for Israel are supplied by, and under the responsibility of, the relevant Israeli authorities. The use of such data by the OECD is without prejudice to the status of the Golan Heights, East Jerusalem, and Israeli settlements in the West Bank under the terms of international law.
There are cross-country differences in the inclusion/exclusion of zero and negative earners. See Definitions and Methodology sections for more information.
1. Index 100 refers to the combined ISCED levels 3 and 4 in the ISCED 2011 classification. See the Reader’s Guide for the list of ISCED levels.
2. Year of reference differs from 2022. Refer to the source table for more details.
3. Earnings net of income tax for Türkiye and a combination of gross (self-employed) and net (employees) earnings for Argentina.
Source:OECD (2024), Table A4.1. For more information see Source section and Education at a Glance 2024 Sources, Methodologies and Technical Notes (https://doi.org/10.1787/e7d20315-en).
In Latin America and the Caribbean, the returns on higher education are notably high, largely due to a lower proportion of tertiary-educated adults compared to OECD countries. For instance, individuals with tertiary attainment earn 56% more than those with only upper secondary education in OECD nations. In countries such as Chile, Costa Rica, and Mexico, the wage disparities are even greater, with tertiary degree holders in Brazil, Chile, Colombia, and Costa Rica earning over twice the average salary of high school graduates (OECD, 2024[20]).
This high return on investment in tertiary education can be attributed to the relative scarcity of graduates in the region. However, significant expansion has occurred over the past few decades, with gross enrolment rates - defined as the total number of students enrolled at this level relative to those in the five-year range following secondary education – more than doubling from 23% in 2000 to 52% in 2018. Nonetheless, this figure remains below the OECD average of 76% by over 20 percentage points (Valenzuela, 2022[21]) (OECD, 2023[22]).
Moreover, despite progress in gross enrolment, the measure should be interpreted with care as the metric includes both underage and overage students. Additionally, in Latin America, completion rates at this level are 25.1% compared to the OECD average of 40%, meaning that a considerable share of students enrolled in tertiary education will not complete it (Arias Ortiz et al., 2024[14]). The share of adults (25-64-year-olds) with at least short-cycle tertiary education is 40% in OECD countries, while in countries like Chile (29%), Costa Rica (22%) and Mexico (21%), these proportions do not surpass 30% (Figure 3.5) (OECD, 2022[23]). Nevertheless, younger generations are more likely to have tertiary attainment (OECD, 2024[24]).
2022 vs. 2011
Note: LAC countries in light blue. LAC average in orange and refers to Argentina, Brazil, Chile, and Costa Rica.
Source: Figure 4.4 OECD economic Surveys, Mexico
The significant wage premium in the region has the potential to enhance social mobility; however, access to tertiary education remains highly inequitable, particularly for low-income individuals, who are 20 percentage points less likely to enrol than their higher-income counterparts (Arias Ortiz et al., 2024[14]). Moreover, when underprivileged students make it to higher education, they are less likely to complete the program and, if they do, they are more likely to have lower wages than their peers with highly educated parents (OECD, 2015[25]) (OECD, 2010[26]).
For tertiary education to effectively drive economic progress and social mobility, it must be high quality and accessible. Financial barriers must be addressed to increase participation from low- and middle-income sectors. Grants and student loans can play a vital role in this regard, as evidence suggests that universal financing improves success rates for these groups (OECD, 2010[26]).
Unfortunately, the rapid expansion of tertiary education in Latin America has often compromised quality. Increased demand, particularly from less affluent sectors, has led to the proliferation of teaching-oriented institutions at the expense of research-focused ones, diminishing overall educational standards (Brunner, 2013[27]). In an increasingly knowledge-based and digital economy, this poses an important restriction in terms of innovation, productivity, and inclusive economic growth.
Overall, while LAC countries have made important strides in educational attainment and coverage since the 2000s - especially achieving near-universal coverage in primary education - many challenges remain. Secondary education completion rates still lag behind those of OECD countries, with high dropout rates persisting, especially among students from disadvantaged backgrounds. This trend often leads to long-term negative impacts on employment opportunities, earnings, and the risk of poverty, as those who do not complete secondary education are less likely to pursue higher education. Despite the high private returns associated with higher education, particularly for bachelor’s degree holders who can earn more than double their peers with just secondary education, access remains inequitable, and earnings are still heavily influenced by family background.
However, educational attainment is only a necessary and but not a sufficient condition for achieving sustained and inclusive economic growth - as the level of skills acquired at the school and tertiary level will determine a worker’s productivity and contribution capacity. In this regard, international assessments that measure students’ cognitive skills across educational systems, such as the OECD’s Programme for International Student Assessment, are essential to evaluate this aspect effectively.
Unlike physical units that have a substantive and universal meaning, PISA scores are established based on the range of results observed across all participants in each assessment. Results are standardised to approximate a normal distribution, meaning there are no strict theoretical minimums or maximums. As a result of the standardisation, the average score is around 500 points, with a standard deviation of approximately 100.
Proficiency levels
The score scale can be categorised into proficiency levels, which for PISA 2022 spanned eight. An individual at a particular proficiency level can generally handle items at that level and below, but struggle with higher-level tasks. Each mathematics proficiency level spans approximately 62 score points, while reading and science proficiency levels differ by about 73 and 75 score points, respectively. Differences of these magnitudes will indicate different levels of skills and knowledge.
More importantly, Level 2 will be considered the minimum proficiency level across reading, mathematics, and science. In mathematics, students will accomplish high levels of proficiency if they understand mathematical problems and can formulate models to solve them, whereas reading proficiency is defined as “(…) understanding, using, evaluating, reflecting on, and engaging with texts in order to achieve one’s goals, to develop one’s knowledge and potential, and to participate in society.” Finally, proficiency in science is the capability of engaging with the discipline as a reflective citizen (OECD, 2019[28]).
Score differences
Differences of smaller magnitudes may or may not reflect variations in terms skills and knowledge. To compare these smaller differences, their statistical significance must be assessed, because PISA results are estimates derived from samples. This is why interpreting results across different assessment years must be done using “link errors” as the same score may not represent the same skills and knowledge across assessments.
Source: Box 1, (OECD, 2023[5])
The Programme for International Student Assessment (PISA) is a triennial evaluation for 15-year-olds who are near the end of their compulsory education. It evaluates them on the areas of reading, mathematics, and science, with each assessment focusing on a particular domain (OECD, 2019[29]). PISA does not just focus on whether students can reproduce knowledge, but whether they can apply it in unknown settings. The assessment also provides schools and policymakers with useful information that allows them to look at what others are doing around the world to improve their educational systems (OECD, 2023[5]).
The most recent evaluation took place in 2022, with the core module being mathematics. It was the first assessment since the COVID-19 pandemic, during which Latin America experienced the world's longest school closures, with an average of 72 weeks (ECLAC, 2023[6]). Despite the difficulties posed by the pandemic, PISA 2022 included data for 81 countries and economies, and almost 700,000 students, who represented 29 million students across all participating countries. Out of the total number of countries that participated, only 13 did not meet one or more sampling standards, but were still included considering the exceptional circumstances.
This version of PISA recorded participation from 14 Latin American and Caribbean (LAC) countries - of which 10 also participated in 2018 - totalling 95,000 students; representing more than 6 million students in the region. First-time participants include El Salvador and Jamaica, while Guatemala and Paraguay joined after participating in the “PISA for Development” initiative (OECD, 2023[30]). Only Jamaica and Panama did not meet all sampling requirements, so results from these countries should be interpreted with caution.
Overall, PISA 2022 revealed a significant decline in the performance of OECD countries in mathematics and reading, with scores dropping by 15 points and 10 points respectively, while science scores remained stable. Historically, the OECD average has fluctuated by no more than four points in mathematics or five points in reading, between consecutive assessments, making these declines unprecedented. This drop suggests a widespread negative impact of school closures; however, it cannot be solely attributed to the pandemic, as declines in reading and science performance had been observed prior to it (OECD, 2023[5]).
Mean scores in Latin America and the Caribbean
The results in Latin America and the Caribbean are notably lower compared to those of OECD countries. (Figure 3.6). The regional mean performance in mathematics is below OECD levels, with 373 points compared to the OECD average of 472. Moreover, all Latin American OECD member countries - Chile, Colombia, Costa Rica, and Mexico - exhibit the lowest average performance within the OECD. The top performer among the OECD is Japan, with 536 points. The highest scores in Latin America were obtained by Chile (412), followed by Uruguay (409) and Mexico (395); while the countries with the lowest average scores were Paraguay (338), the Dominican Republic (339), and El Salvador (343).
For reading, the participating Latin American and the Caribbean countries averaged a score of 404 points, compared to 476 in the OECD (Figure 3.7). The top performer among OECD countries was Ireland, with 516 points.1 The top LAC performers were Chile (448), Uruguay (430), and Mexico (415), with Chile scoring higher than Greece (438) and Iceland (436). Conversely, the LAC countries with the lowest mean scores were the Dominican Republic (351), El Salvador (365), and Paraguay (373).
In science, the LAC average was 400 points, whereas OECD countries’ average was 485 (Figure 3.8). For the OECD, the top performer was Japan, with a mean score of 547 points. In LAC countries, the top performers were Chile (444), followed by Uruguay (435) and Colombia (411). The Chilean results, while low, were similar to those of Iceland (447) and Greece (441). The LAC countries with the lowest average scores are the Dominican Republic (360), Paraguay (368), and Guatemala (373).
The differences in performance observed between the top-performing OECD countries and those in the LAC region highlight the significant challenges faced by Latin America and the Caribbean in improving the quality of their educational systems. Given the strong correlation between PISA performance and both tertiary education enrolment and employment likelihood, the results from PISA 2022 serve as a crucial call to action for LAC countries.
Note: OECD average considers Chile, Colombia, Costa Rica, and Mexico. LAC countries highlighted in orange.
Source: Table I.B1.2.1 PISA Results 2022 Vol. I
Note: OECD average considers Chile, Colombia, Costa Rica, and Mexico. LAC countries highlighted in orange.
Source: PISA 2022 Results (Volume I): The State of Learning and Equity in Education
Note: OECD average considers Chile, Colombia, Costa Rica, and Mexico. LAC countries highlighted in orange.
Source: Table I.B1.2.3 PISA 2022 Results (Volume I): The State of Learning and Equity in Education
Proficiency levels in Latin America and the Caribbean
In terms of proficiency levels, Level 2 is considered the minimum proficiency level. In mathematics, three out of four LAC students underperform - meaning that they perform below Level 2 - compared to less than one in three (32%) at the OECD (Figure 3.9). The OECD country with the highest proportion of students in level 2 or above in mathematics is Japan, where almost 9 out of 10 students (88%) obtain scores above Level 2. In LAC, the countries with the highest proportion of students achieving basic proficiency (2 and above) are Chile (44%), Uruguay (43%), and Mexico (34%). On the contrary, the countries with the lowest share of students performing at baseline proficient level in the region are the Dominican Republic (8%), El Salvador (11%), and Guatemala (13%).
Across the fourteen Latin American countries that participated in PISA 2022, more than half (55%) of students do not reach basic proficiency in reading (or a score that situates them at Level 2 or higher), compared to only 1 in 4 (25%) in OECD countries (Figure 3.10). The OECD country with the highest share of students achieving baseline proficiency (level 2 and above) in reading is Ireland with 89%; whereas, in LAC, Chile leads with 66%, followed by Uruguay with 59%, Mexico with 53% and, finally, Costa Rica with 52%. Among the countries with the lowest shares of students scoring at level 2 or higher, the Dominican Republic has only 24%, followed by El Salvador at 28% and Guatemala at 32%. Notably, reading is the area in which the region demonstrates its strongest performance.
In science, less than 1 in 5 (17%) students in Latin America score at, or above, Level 2 compared to 1 in 2 in OECD countries (Figure 3.11). Japan has the highest share of students performing at Level 2 or higher (75%), whereas in LAC, Chile (30%), Uruguay (30%), and Colombia (20%) have the highest shares. In contrast, the Dominican Republic (5%), Guatemala (5%), and Paraguay (7%) have the lowest shares of students scoring at Level 2 or above.
Note: OECD average considers Chile, Colombia, Costa Rica, and Mexico. LAC countries highlighted in orange.
Countries are ranked in descending order of proficient students (Level 2 and up).
Source: Table I.B1.3.1 PISA 2022 Results, Vol I
Note: OECD average considers Chile, Colombia, Costa Rica, and Mexico. LAC countries highlighted in orange.
Countries are ranked in descending order of proficient students (Level 2 and up).
Source: Table I.B1.3.2 PISA 2022 Results (Volume I): The State of Learning and Equity in Education
Note: OECD average considers Chile, Colombia, Costa Rica, and Mexico. LAC countries highlighted in orange.
Countries are ranked in descending order of proficient students (Level 2 and up).
Source: Table I.B1.3.3 PISA 2022 Results (Volume I): The State of Learning and Equity in Education
Socio-economic status encompasses a wide range of factors that reflect students' access to familial resources such as economic, social, and cultural capital, as well as their family's social standing. In the PISA context, this status is quantified using the PISA index of economic, social, and cultural status (ESCS).
A higher ESCS score indicates a higher socio-economic status, with the index centred around a mean of 0 and a standard deviation of 1 among OECD countries. The ESCS is a composite measure derived from three indicators: the highest educational attainment of the parents, the highest occupational status of the parents, and household possessions. These indicators are collected from each student via a questionnaire completed after the PISA cognitive tests.
Source: Box I.4.1, (OECD, 2023[5])
In PISA, a student’s socio-economic status is estimated by the index of economic, social, and cultural status - the ESCS index - a composite measure that combines into a single score the financial, social, cultural, and human capital resources available to students. Information about these three components for each student was collected through the student questionnaire, a survey that students answered after completing the PISA cognitive assessment.
The socio-economic status of students varies between countries/economies but in the vast majority of cases, differences in socio-economic status, which may be seen as a proxy of the socio-economic inequalities in the countries, are larger within than between countries/economies. Comparing the ESCS value at the top and bottom 10% of the index distribution shows wider socio-economic differences in Latin America and the Caribbean as compared to OECD countries. The most pronounced differences are found in Guatemala, Paraguay, and Panama, while Jamaica, Chile, and the Dominican Republic exhibit the smallest gaps. Nonetheless, all participating LAC countries demonstrate a more significant disparity between the top and bottom 10% than the OECD average (Figure 3.12).
LAC and OECD countries
Note: Only countries with available data are shown. LAC countries highlighted in orange.
All differences between the 90th and the 10th percentiles are statistically significant (see PISA Results Volume I Annex A3).
Countries ranked in ascending order of the ESCS difference between the top and bottom 10%.
Source: OECD, PISA 2022 Database, Table I.B1.4.2.
While the ESCS index offers a proxy for students' access to family resources within their country or economy, the socio-economic gradient in PISA assesses the relationship between students' socio-economic status and academic performance. The strength of the gradient is measured by the proportion of the variation in student performance that is accounted for by differences in student socio-economic status. When this relationship is strong, socio-economic status predicts performance well. In other words, a strong relationship between socio-economic status and performance in PISA may indicate low social mobility or less fairness within the country/economy (Willms, 2006[31]).
Given the high-income inequality in the region, and the stark differences between advantaged and disadvantaged students based on the ESCS index, one might expect a strong socio-economic gradient in Latin America and the Caribbean (LAC). However, data from Figure 3.13 shows that LAC countries demonstrate levels of fairness comparable to, or better than, OECD countries. Chile, the Dominican Republic, Jamaica, Mexico, and Paraguay exhibit significantly higher fairness levels than the OECD average, while other LAC countries with available socio-economic data are not statistically different from the OECD average. However, all LAC countries perform below the OECD average in mathematics.
This is further reflected in the share of resilient students - those in the bottom quartile of the ESCS index but performing in the top quartile academically. The proportion of resilient students varies across Latin America, ranging from 7.4% in Peru to 15.2% in Jamaica, with the latter exhibiting one of the highest shares of resilient students in PISA 2022. Higher levels of resilience correlate with greater educational equity, inclusive systems and, therefore, promote social mobility (OECD, 2016[32]).
Unlike other metrics, the region shows diversity in terms of the rate of resilient students, with some Latin American and Caribbean countries even outperforming the OECD average in terms of resilient students, particularly in mathematics (Figure 3.14). However, when interpreting these results related to a lower socio-economic gradient and a higher share of resilient students in some of the LAC countries, a key piece of information is necessary: the average differences in performance between the most advantaged and disadvantaged groups.
Note: Only countries with available data are shown. LAC countries highlighted in orange.
Source: PISA 2022 Database, Tables I.B1.2.1 and I.B1.4.3.
Note: Only countries with available data are shown. LAC countries highlighted in orange.
Percentage of socio-economically disadvantaged students who scored in the top quarter of mathematics performance in their own country.
Countries ranked in descending order of share of resilient students.
LAC average calculated as a simple average.
Source: PISA 2022 Database, Table I.B1.4.3.
All assessed countries demonstrate a significant performance gap in terms of PISA scores based on socio-economic status, as measured by the ESCS index. This suggests intergenerational transmission of privilege, as the ESCS combines parents’ education level, occupational status, and household assets (a proxy for wealth) into a single measure. In other words, differences in educational outcomes of advantaged and disadvantaged students may reflect the influence of parental education and financial resources. Students from advantaged backgrounds tend to perform better academically due to greater access to resources, unlike those from families with lower educational attainment, unemployment, or poverty.
On average, OECD countries exhibit a 93-point gap between students in the top and bottom ESCS quartiles, whereas in Latin America, the difference is 68 points (Figure 3.15). Those in the top 25% in OECD countries score on average 525 points in the mathematics assessment, compared to the mean score of 432 for the bottom quartile. In Latin America, the top quartile achieves an average score of 412 points, with the bottom quartile averaging just 344 points. Worryingly, those in the top 25% in Latin America and the Caribbean have lower scores in terms of performance, on average, than those in the bottom 25% in OECD countries. Rather than indicating greater fairness, this narrower performance gap in the LAC may reflect underachievement among the more advantaged students.
Note: Only countries and economies with available data are shown. LAC countries highlighted in orange.
Countries and economies are ranked in descending order of mathematics performance for students in the second quarter of national socio-economic status.
LAC average calculated as a simple average.
Source: OECD, PISA 2022 Database, Table I.B1.4.3.
To further explore this, it is essential to compare the top and bottom 10% of students by socio-economic status at the international level. In OECD countries, the performance gap between these groups is 144 points, compared to 107 points in Latin America (Figure 3.16).
In LAC, the top 10% of students in terms of the ESCS index displays an average of 444 points in mathematics, while the bottom 10% reaches a mean score of 337 points. Differences between both groups are significant for all countries and range between 77 points in Jamaica2 to 137 points in Brazil.
Note: Only countries with available data (Costa Rica excluded in LAC).
Countries are ranked in descending order by tenth decile performance.
Source: OECD PISA 2022 Database, Tables I.B1.4.7 and I.B1.4.11
The share of underperforming disadvantaged students is also concerning. On average, 88% of disadvantaged students - defined as those in the national bottom quartile of the ESCS index - in the LAC region underperform in mathematics, compared to 47% in OECD countries. In four countries in the region, i.e., the Dominican Republic, El Salvador, Guatemala, and Panama, 90% or more of disadvantaged students lack basic math skills (Figure 3.17). At the lower end, Chile and Uruguay have the smallest share of underperformers in the bottom socio-economic quartile, though the rates are still high at 71% and 75%, respectively. In contrast, Japan leads the OECD with only 21% of disadvantaged students underperforming.
Shifting the focus to the top quartile, in OECD countries, only 13.5% of socio-economically advantaged students underperform in mathematics, compared to 57% in LAC. For the former, the country with the highest share of underperformers in the top ESCS quartile is Greece with 26%; while, in LAC, El Salvador, Guatemala, and the Dominican Republic have the highest shares of advantaged underperformers, with 71%, 74% and 81%, respectively. Uruguay, with 30%, fares best in Latin America and the Caribbean but still lags far behind the OECD's top performers such as Switzerland, Japan, and Estonia, which have the lowest shares of advantaged underperformers at just around 5%.
Note: Only countries and economies with available data are shown. LAC countries highlighted in orange.
Countries and economies are ranked in ascending order of the share of low performers in mathematics among students in the bottom quarter of the ESCS index.
Source: OECD, PISA 2022 Database, Table I.B1.4.14.
Most of the students in the region underperform in mathematics, regardless of socio-economic status. Furthermore, even those at the top of the income distribution do not show high levels of proficiency, indicating that the region's educational quality is lagging behind OECD standards. Measures of fairness in the education system, such as the share of resilient students, offer a glimpse of hope for the region. But while a weak association between student socio-economic status and performance within countries/economies is necessary for achieving fairness in education, it is not, of itself, a sufficient condition. It is also important to consider fairness in terms of education systems’ overall levels of performance. The challenge in LAC countries is the coexistence of high levels of fairness in terms of student socio-economic status with low mean performance – indicating poor achievement across the board regardless of students’ socio-economic status. This should not be taken as a desirable outcome. Therefore, many of the region's positive outcomes related to socio-economic disparities, such as the socio-economic gradient - which is the proportion of performance explained by socio-economic background - are largely explained by the underperformance of students at the top of the distribution, rather than strong performance by those at the bottom.
If even socio-economically advantaged students with greater access to resources perform at basic proficiency levels, the region will struggle to develop a highly skilled and productive workforce, which is crucial for long-term growth and well-being. This widespread low performance calls for universal policies to raise the educational outcomes of all students.
As spending per student on education increases, so does a country’s mean performance in PISA. When examining the cumulative spending per student, it is possible to establish a significant and positive correlation between spending and performance. However, this relationship holds only up to a cumulative spending threshold of USD 75,000, beyond which per-student spending and performance begin to diverge (OECD, 2023[5]).
This analysis is particularly relevant to Latin America and the Caribbean on two accounts (Figure 3.18). First, none of the countries in the region have surpassed this spending threshold, leaving room for improvement through increased investment - Panama, the highest spender, reaches USD 62,000 per student. Second, at similar levels of spending, some countries outside the region, such as Vietnam, Turkey, and Serbia, outperform LAC nations. For instance, Vietnam outperforms the region's countries despite comparable spending levels.
Top-performing economies in PISA 2022 also vary widely in their spending per student. Japan, the highest-performing OECD country in mathematics, spends more than the USD 75,000 threshold but below the OECD average. Korea, on the other hand, spends significantly more than Japan without achieving better results. This highlights that both the amount of spending, and how funds are allocated, are critical to improving student outcomes.
Given that once cumulative spending per student exceeds the threshold it is difficult to reduce, policymakers should carefully consider resource allocation. For example, in high-performing countries such as Korea and Hong Kong-China, significant investments are made in teachers’ working conditions. Lower secondary teachers in these countries earn more than double the average salary to attract top talent to the profession. This emphasis on quality teaching may be a key factor in their success (OECD, 2012[33]).
Note: Only countries and economies with available data are shown. LAC countries highlighted in orange.
Source: OECD, PISA 2022 Database, Tables I.B1.2.1 and I.B3.2.2.
Additionally, PISA 2022 data confirms that higher mean country performance in mathematics is associated with elevated GDP per capita levels (Figure 3.19). As noted earlier, there is evidence linking PISA results to adulthood well-being indicators, such as employment probability and the likelihood of attaining a tertiary degree by age 25.
At the aggregate level, it is estimated that increasing PISA scores by 25 points over two decades could result in an economic gain of USD 115 trillion in OECD countries, attributed to enhanced lifetime earnings (OECD, 2019[7]). Given that all countries in Latin America and the Caribbean have significant proportions of low-performing students regardless of socio-economic status, there is substantial potential for these nations to benefit from further and more efficient investments in education.
Note: LAC countries highlighted in orange.
Source: OECD, PISA 2022 Database, Tables I.B1.2.1 and I.B3.2.1.
For education systems in Latin America and the Caribbean to produce a skilled and productive workforce that propels social mobility, they must prioritise both high quality and equitable access. This means ensuring that success in the education system is not determined by a student's socio-economic background. While the region has made significant strides in improving educational attainment - achieving near-universal primary coverage and increasing secondary enrolment - issues such as dropout rates remain a pressing concern, particularly as they tend to rise with age. Addressing dropout levels, especially among at-risk students, is crucial, as those without a secondary education are at a heightened risk of poverty and unemployment. Policymakers should implement targeted measures, including therapeutic programs and strategies that ensure smooth transitions between educational levels, to combat this issue effectively.
Completing upper secondary education is vital for accessing tertiary programs, which have also seen considerable improvement in enrolment rates over the past two decades. However, completion rates still lag behind OECD averages, and the financial barriers faced by students from lower socio-economic backgrounds hinder the potential of higher education to serve as a true engine for social mobility. Financing mechanisms such as student loans and grants must be made accessible to level the playing field.
Despite advancements in access and attainment, the quality of education in the region remains inadequate. The latest PISA data reveals a significant performance gap between LAC countries and their OECD counterparts, with underperformance in mathematics, science, and reading. Interestingly, while LAC exhibits higher socio-economic fairness, indicated by a weak socio-economic gradient that allows students to reach their potential regardless of background, overall proficiency levels remain low across the spectrum. Consequently, it appears that higher socio-economic fairness is not a result of strong performances among disadvantaged students; instead, it highlights the inadequate outcomes of advantaged students.
To counteract these challenges, countries in the region should focus on increasing their education spending and, more critically, on investing those resources wisely. Evidence suggests a positive correlation between cumulative spending and performance, up to a certain threshold that no LAC country has yet reached. Furthermore, countries with similar levels of expenditure outperform LAC nations, highlighting the need for improved resource allocation. Learning from the experiences of top-performing countries, such as Korea and Hong Kong, which have implemented strategies to attract the best candidates into teaching careers, could provide valuable insights for reform.
As the region moves forward, it is essential to recognise the connection between education performance and long-term well-being indicators, such as employment probabilities and tertiary attainment. By committing to more effective and equitable education funding, Latin America and the Caribbean can not only enhance their educational systems, but also foster inclusive growth, reduce inequalities, and build a productive workforce that benefits all citizens. The time to act is now; the future of the region depends on the choices made in education today.
References
[14] Arias Ortiz, E. et al. (2024), El estado de la educación en América Latina y el Caribe 2023, Banco Interamericano de Desarrollo, https://doi.org/10.18235/0005515.
[27] Brunner, J. (2013), “The Rationale for Higher Education Investment in Ibero-America”, OECD Development Centre Working Papers, No. 319, OECD Publishing, Paris, https://doi.org/10.1787/5k40d67l7l8x-en.
[3] ECLAC (2024), Preventing and reducing school dropout in Latin America and the Caribbean.
[6] ECLAC (2023), Panorama Regional en Educación, https://www.cepal.org/sites/default/files/presentations/presentacion_d_trucco_cepal.pdf.
[4] ECLAC, UNESCO and UNICEF (2022), Education in Latin America and the Caribbean at a crossroads.
[15] ECLAC, UNESCO, UNICEF (2024), “:<i>Education in Latin America and the Caribbean at a Crossroads: Regional Monitoring Report SDG4—Education 2030</i>”, Comparative Education Review, Vol. 68/2, pp. 312-314, https://doi.org/10.1086/730227.
[17] Lyche, C. (2010), “Taking on the Completion Challenge: A Literature Review on Policies to Prevent Dropout and Early School Leaving”, OECD Education Working Papers, No. 53, OECD Publishing, Paris, https://doi.org/10.1787/5km4m2t59cmr-en.
[20] OECD (2024), Education at a Glance, OECD Publishing.
[24] OECD (2024), OECD Economic Surveys: Mexico 2024, OECD Publishing, Paris, https://doi.org/10.1787/b8d974db-en.
[22] OECD (2023), Education at a Glance 2023: OECD Indicators, OECD Publishing, Paris, https://doi.org/10.1787/e13bef63-en.
[13] OECD (2023), Multi-dimensional Review of El Salvador: Strategic Priorities for Robust, Inclusive and Sustainable Development, OECD Development Pathways, OECD Publishing, Paris, https://doi.org/10.1787/2f3d5e1f-en.
[5] OECD (2023), PISA 2022 Results (Volume I): The State of Learning and Equity in Education, https://doi.org/10.1787/53f23881-en.
[30] OECD (2023), PISA 2022: Resultados para América Latina y el Caribe, https://www.cepal.org/sites/default/files/presentations/presentacion_d_salinas_ocde.pdf.
[23] OECD (2022), Education at a Glance 2022: OECD Indicators, OECD Publishing, Paris, https://doi.org/10.1787/3197152b-en.
[12] OECD (2022), Gender Equality in Peru: Towards a Better Sharing of Paid and Unpaid Work, Gender Equality at Work, OECD Publishing, Paris, https://doi.org/10.1787/e53901b5-en.
[2] OECD (2021), How’s Life in Latin America?: Measuring Well-being for Policy Making, OECD Publishing, Paris, https://doi.org/10.1787/2965f4fe-en.
[7] OECD (2019), How are PISA results related.
[28] OECD (2019), PISA 2018 Assessment and Analytical Framework, PISA, OECD Publishing, Paris, https://doi.org/10.1787/b25efab8-en.
[29] OECD (2019), PISA 2018 Assessment and Analytical Framework, https://doi.org/10.1787/b25efab8-en.
[10] OECD (2018), Equity in Education: Breaking Down Barriers to Social Mobility, PISA, OECD Publishing, Paris, https://doi.org/10.1787/9789264073234-en.
[9] OECD (2018), Opportunities for All: A Framework for Policy Action on Inclusive Growth, OECD Publishing, Paris, https://doi.org/10.1787/9789264301665-en.
[16] OECD (2017), OECD Economic Surveys: Argentina 2017: Multi-dimensional Economic Survey, OECD Publishing, Paris, https://doi.org/10.1787/eco_surveys-arg-2017-en.
[32] OECD (2016), PISA 2015 Results (Volume I): Excellence and Equity in Education, PISA, OECD Publishing, Paris, https://doi.org/10.1787/9789264266490-en.
[18] OECD (2016), Skills Matter: Further Results from the Survey of Adult Skills, OECD Skills Studies, OECD Publishing, Paris, https://doi.org/10.1787/9789264258051-en.
[25] OECD (2015), “Higher education in Latin America: Challenges and opportunities”, in E-Learning in Higher Education in Latin America, OECD Publishing, Paris, https://doi.org/10.1787/9789264209992-4-en.
[33] OECD (2012), “Does Money Buy Strong Performance in PISA?”, PISA in Focus, No. 13, OECD Publishing, Paris, https://doi.org/10.1787/5k9fhmfzc4xx-en.
[26] OECD (2010), Latin American Economic Outlook 2011: How Middle-Class Is Latin America?, OECD Publishing, Paris, https://doi.org/10.1787/leo-2011-en.
[8] OECD (2010), The high cost of low education performance.
[19] OECD/CAF/ECLAC (2016), Latin American Economic Outlook 2017: Youth, Skills and Entrepreneurship, OECD Publishing, Paris, https://doi.org/10.1787/leo-2017-en.
[1] Torche, F. (2014), “Intergenerational Mobility and Inequality: The Latin American Case”, Annual Review of Sociology, Vol. 40/1, pp. 619-642, https://doi.org/10.1146/annurev-soc-071811-145521.
[11] UNESCO (2024), The urgency of educational recovery in Latin America and the Caribbean.
[21] Valenzuela, J. (2022), Trajectory and policies for inclusion in higher education in Latin America and the Caribbean in the context of the pandemic.
[31] Willms, J. (2006), “Variation in socioeconomic gradients among cantons in French- and Italian-speaking Switzerland: Findings from the OECD PISA”, Educational Research and Evaluation, Vol. 12(2), pp. 129-154.