5. Well-being inequalities across social groups and territories
It is impossible to fully evaluate the well-being situation of a society without considering inequalities, and previous chapters have included various indicators of “vertical” inequality (such as the Gini coefficient) and deprivation (such as poverty rates). This chapter focuses on “horizontal” inequalities, or inequalities across different population groups, by gender, age, territory (urban vs rural), ethnicity and race, and educational attainment. Overall, while the focal countries have made progress in reducing inequalities in a number of well-being areas, such as education and employment, important gaps still remain. The consequences of the pandemic are likely to deepen the disadvantages experienced by vulnerable groups, widening inequalities in well-being outcomes. Overall, women, children and youth, those living in rural areas, Indigenous and Afro-descendant peoples, and those with lower education tend to experience worse material conditions, although the picture is more mixed when looking at indicators of quality of life and human and social capital.
It is impossible to fully evaluate the well-being situation of a country without considering inequalities. This is especially the case in Latin America and the Caribbean (LAC), where inequality has been a historical and structural feature of society for centuries, persisting even in periods of substantial economic growth and social development (Sánchez-Ancochea, 2021[1]). Combatting inequalities in opportunity and outcomes is at the heart of the UN 2030 Agenda, and its objective to “leave no one behind” recognises that development which serves only a privileged few cannot be sustainable. The 2030 Agenda also recognises that inequalities are multidimensional and interlinked, going far beyond income inequality. It is also important to recognise that tackling inequalities is about addressing the situation not only of those at the lowest end of the distribution but also of those in the vulnerable middle classes (OECD, 2019[2]). This is especially important in the Latin American context where rising dissatisfaction with inequalities and living standards was one important driver of the wave of social protests in late 2019 (ECLAC, 2021[3]; Ferreira, 2020[4]; Langman, 2019[5]).
Well-being inequalities can be conceptualised and measured in different ways. The OECD Well-being Framework, for example, looks at inequalities from three perspectives: vertical and horizontal inequalities and deprivation (OECD, 2017[6]). Measures of “vertical” inequalities address how unequally outcomes are spread across all people in society – for example, by looking at the size of the gap between people at the bottom and at the top of the distribution for all dimensions of people’s lives. By contrast, measures of “horizontal” inequalities focus on the gap between population groups defined by specific characteristics (such as men and women, or young and old). “Deprivation” measures focus on people who live below a certain level of well-being (such as those who live in an overcrowded household, or with insufficient income to meet basic needs). The previous chapters have already included a number of indicators of both vertical inequality (such as the Gini indicator of income inequality) and deprivation (such as poverty and overcrowding). Indeed, any high-level description of well-being outcomes that focuses on average outcomes alone will be incomplete, as inequality and deprivation are integral parts of the picture. The integration of these measures of vertical inequalities and deprivation in earlier chapters underlines that they are not a side issue: they not only affect those who are excluded or deprived in some way, but undermine overall development within a society.
This chapter focuses on the remaining type of inequality, i.e. horizontal inequality between social groups and territories. These horizontal inequalities matter both intrinsically and instrumentally, as the shared characteristics of various groups can provide a strong basis for their identity and be a source of political mobilisation.1 Understanding differences in well-being across different groups is fundamental for the design of effective policies to leave no one behind and to raise the overall well-being of a country’s population. Getting a clearer picture of the disadvantages of specific groups is particularly important in the context of the COVID pandemic, which has exacerbated pre-existing vulnerabilities for several population groups.
Horizontal inequalities and deprivation shed light on the issue of inequality of opportunities that are in large part established at birth, based on characteristics that are an in-built feature of people’s lives. Inequalities of opportunity, in all life dimensions, can be understood as the share of inequalities of outcomes due to circumstances that are beyond an individual’s control. While not all of these circumstances can be observed, some of them, such as gender, ethnicity and race, age or place of living, can be. A useful analogy put forward by Francois Bourguignon (Stiglitz, Fitoussi and Durand, 2018[7]) is that of a marathon where runners don’t start from the same starting line; in this setting, “ex post inequality (i.e. inequality of outcomes) would essentially be the distribution of the finishing times”, while “ex ante inequality would refer to the distance competitors have to run to reach the finish line”. The two concepts of ex post (i.e. vertical inequalities and deprivations) and ex ante inequalities are distinct but closely inter-related: an increase in ex ante inequality will, other things being equal, increase ex post inequality. In the same way, inequality of outcome at a point of time or within a generation may affect inequality of opportunity in the future or in the next generation (Stiglitz, Fitoussi and Durand, 2018[7]). Understanding differences in well-being across different groups is fundamental for the design of effective policies to leave no one behind and to raise the overall well-being of a country’s population. Getting a clearer picture of the disadvantages of specific groups is particularly important in the context of the COVID pandemic, which has exacerbated inequalities of both outcome and opportunity, as well as the negative feedbacks between the two types of inequality.
Following ECLAC’s social inequality matrix (ECLAC, 2016[8]), the chapter explores inequality among groups from the perspectives of gender, ethnicity and race, age (focusing on the particularly vulnerable age groups of children, young people and the elderly) and territory (focusing on urban-rural inequalities). In addition, it looks at inequalities by education level, an important aspect of socio-economic status. This is not an exhaustive exploration of horizontal inequalities, as there are many other personal and social characteristics that can exacerbate the disadvantage of certain individuals or groups, such as migrant status, disability or sexual orientation. However, the data needed to explore outcomes along these other dimensions are simply not available,2 implying that improving data collection to assess them remains a priority for the statistical agenda ahead (not least in the context of the UN 2030 agenda). It is especially important to improve the availability of data that can show the intersection of multiple sources of disadvantage (e.g. gender, ethnicity or race, and socio-economic status) in order to identify the most vulnerable. On occasion, this chapter highlights examples of intersecting inequalities, but it has not been possible to do this systematically.
Significant progress has been made in improving well-being outcomes for women in Latin America over recent decades, including reducing maternal mortality (as shown in Chapter 3) and increasing labour force participation and political representation (see later in this section). However, persistent gender inequalities remain in every country in the region, holding back wider social and economic development. In order to achieve gender equality, four structural barriers have been identified as priorities to overcome in the LAC region: socioeconomic inequality and poverty; discriminatory, violent and patriarchal cultural patterns and the predominance of a culture of privilege; the sexual division of labour and unfair social organisation of care; and the concentration of power and hierarchical relations in the public sphere (ECLAC, 2017[9]).
Figure 5.1 shows performance ratios for selected well-being outcomes for women in comparison to men, on average across the 11 focal LAC countries.3 To ease interpretation, all indicators are coded in the same direction, so that 1 implies parity between men and women, ratios above 1 denote better well-being outcomes for women in comparison with men, and ratios below 1 denote worse outcomes for women.
On average in the focal countries, women perform worse than men across almost all selected indicators of material conditions (Figure 5.1, Panel A). Women are much less likely to be employed, nearly one-third more likely to be unemployed, and more likely to work in informal employment. Only regarding perceived job insecurity and overtime is the opposite true, with men more likely to do more than 60 hours per week of paid work, and more likely to be worried about losing their job in the next 12 months. However, even these “positive” indicators for women have to be understood as part of the bigger picture. For example, the disproportionate burden of unpaid care work (as is explained later) that women bear constitutes an important barrier to engaging in the labour market and increasing paid working hours, thus explaining their lower propensity for working long hours.
Women’s marginalisation in the labour market is reflected, at least partly, in lower earnings – especially when looking at monthly earnings (with a gender pay gap of 14%). Overall, women are slightly more likely to live in poverty, and extreme poverty (with even starker differences when looking at the population aged 20-59 years, see Figure 5.2), are less likely to feel that their income is insufficient to meet their needs, and are more than twice as likely as men to have no income of their own.
Note: Each performance ratio is the simple average by gender calculated across the 11 focal countries for which data are available for all the inequality groups taken into consideration. As a result, performance ratios can cover a subset of the 11 focal countries. Performance ratios above 1 indicate better outcomes (i.e. higher well-being) for women, whereas performance ratios below 1 indicate better outcomes (i.e. higher well-being) for men – including negative indicators, which have been reverse-scored. Grey bubbles denote no clear difference between men and women, defined as gender ratios within 0.03 points distance to parity.
Source: OECD calculations. For a complete list of sources please refer to the "Sources and Methods" tab in the Statlink file
In the case of the selected indicators of quality of life (Figure 5.1, Panel B), on average across the focal countries, the largest gender gap relates to homicide. Men are over eight times more likely to die by homicide. Men are also 13% more likely to report that they or their families had been victims of a crime than women. Taken together, these indicators may seem to suggest that women are less vulnerable to violent outcomes than men overall, but the reality is more complex, as is explained in the later section on “violence against women”. Figure 5.1 also shows that in terms of perceived safety women fare worse than men and are less likely to feel safe walking alone at night in their neighbourhood. Women tend to be more physically vulnerable than men, and while less likely to be involved in risky activities such as crime and gang activities that may lead to violent death, they nonetheless face pervasive threats in terms of sexual assault and domestic or intimate-partner violence (IPV) that are less well-measured through comparable official statistics (see later in the section).
Overall, women live almost 6 years longer than men on average in the focal countries, with an average life expectancy of 79.8 years, compared with 74 years for men. In terms of mental and emotional well-being, the indicators are mixed. Men are over three times more likely to die by suicide than women.4 However, women are more likely than men to experience negative affect balance, where negative emotions (such as worry, sadness, stress or anger) outweigh positive emotions (such as enjoyment or laughter) on a typical day. In terms of overall life satisfaction, there is no clear difference, with women having only marginally higher levels.
Women in the focal group are more likely to have completed secondary and tertiary education, and girls show marginally better performance in reading cognitive tests at age 15 than boys (with an average mean PISA score of 419.5 for girls, compared with 401.5 for boys). On the other hand, boys at age 15 tend to score slightly higher in cognitive tests in mathematics and science than girls. While the differences are very small, gender gaps in these fields have tended to widen over time. The pattern of boys displaying a relative strength in science has been observed across almost all countries globally that participate in the Programme for International Student Assessment (PISA), and is associated with lower graduation and employment rates of women in STEM fields later in life (Mostafa, 2019[10]).5
Finally, while there is little clear gender difference in perceived elite State capture, with women only slightly more likely to believe that their country is governed by powerful groups for their own benefit, women are much less likely than men to voice their opinion to an official.
A selection of indicators of social and human capital are also available by gender (Figure 5.1, Panel C). These show that, on average across the focal countries, men are more likely than women to trust others and to trust in government, as well as being more likely to volunteer and more likely to believe that democracy is preferable over other kind of governments. On the other hand, there is little gender difference in the likelihood of men and women believing that the government is corrupt or saying that they belong to a discriminated group. This latter result is counter-intuitive, given the many manifestations of gender discrimination against women.6 Regarding human capital indicators, young men are around half as likely as women to be in neither employment nor education or training (NEET),7 and less likely to be obese, although there is little difference in the prevalence of overweight between the sexes. On the other hand, young women are more likely to have completed upper secondary education and men are almost twice as likely to consume tobacco and over 2.5 times more likely to consume alcohol.
The remainder of this section looks at a selection of indicators in more detail, including indicators that do not appear elsewhere in the report but are especially significant for understanding gender inequalities (such as violence against women).
Poverty
Not only are women in Latin America more likely to live in poverty8 than men, but the gender gap has widened even further over the last two decades. Gender differences are even starker for the working-age population than the total population. Figure 5.2, Panels A and B, shows data for the Feminity Index of Poverty and Extreme Poverty, as calculated by ECLAC, which focuses on the population aged 20 to 59. According to this measure, in 2019, for every 100 men living in (absolutely) poor households in the region there were at least 112 women in a similar situation (see Figure 5.2, Panel A), up from a regional average of 105 women in 2002. The feminisation of extreme poverty was even higher, at 115.3 in 2019, compared with 106.6 in 2002. In Chile, the Dominican Republic and Uruguay, women aged 20 to 59 were over 30% more likely than similarly aged men to live in poor households.9
Note: Panels A and B show the share of women aged 20-59 living in absolute poverty (or extreme poverty) as a ratio of the shares for men aged 20-59. Values above 100 mean poverty is more prevalent for women, while values below 100 mean poverty is worse for men. LAC is the regional average as calculated by ECLAC. For Panels A and B, the latest available year is 2018 for the Dominican Republic and 2017 for Chile. The earliest available year is 2001 for Brazil, Ecuador, Paraguay and Peru; 2002 for Colombia; and 2007 for Uruguay. For Panels C and D, the latest available year is 2018 for Mexico and 2017 for Chile. The earliest available year is 2001 for Brazil, Paraguay and Peru; 2002 for Colombia; and 2007 for Uruguay. In Panels A, B and C, data for Argentina refer to the urban population only.
Source: ECLAC Statistics, CEPALSTAT database, https://estadisticas.cepal.org/cepalstat/WEB_CEPALSTAT/estadisticasIndicadores.asp
The income poverty measures showed here are calculated based on the assumption of the equal sharing of household income amongst all household members. One way of capturing within-household inequalities is to look at the share of people who do not have their own income. Women are much more likely than men to have no income of their own10 (Figure 5.2, Panel C). On average, across the focal countries, almost a quarter of women (24%) had no own income, compared with 10% of men. The autonomy of women with no income is severely compromised, and their survival depends on belonging to a household where resources accessed by other household members are pooled between all members (Amarante, Colacce and Scalese, forthcoming[11])
While the drivers of gender inequalities in income poverty and economic autonomy are complex, reducing gender differences depends largely on two interlinked factors: improving women’s access to quality paid work on the one hand, and introducing policies to reduce the disproportionate female burden of unpaid work on the other (ECLAC, 2014[12])These issues are explored further below.
Work and job quality
In 2019, the female employment rate was 54%, well below the male employment rate of 79% (see Statlink for Figure 5.1). Female employment rates in the region increased considerably in the late 1990s and early 2000s (by 5.3 percentage points between 1997 and 2007), but there has been little change in the level of female participation or the size of the gender participation gap since 2007 (ECLAC, 2018[13]). This deceleration in female labour force participation has affected all groups of women, but especially married women and those from more vulnerable households (Gasparini et al., 2015[14]). Overall, Latin Americans tend to have favourable attitudes towards women’s right to work, with 89% of men and 92% of women in the region agreeing that any woman should have a paid job outside home if she wants one (Gallup Inc. and ILO, 2017[15]). Out of 11 world regions, only North America and Europe (excluding Eastern Europe) have higher favourable attitudes towards female employment. However, the acceptability of a woman’s right or desire to work is strongly conditioned by her role and bargaining power within the household and by the circumstances of other household members. In 2015, a third of respondents from the 11 focal countries (33.7%) agreed or strongly agreed with the notion that women should work only if their partner does not earn a sufficient income.11 This likely reflects expectations that women take on more traditional gender roles within a household, including a greater responsibility for childcare and other forms of unpaid work (see below).
Women in the LAC region face both horizontal and vertical segregation in the labour market. Horizontal segregation refers to the concentration of women in low-productivity jobs in certain sectors or occupations that tend to pay lower wages, provide weak or no social protection and have low job security (ECLAC, 2021[3]). For example, across 17 LAC countries for which data are available, around four-fifths of female workers in 2018 (79.2%) were employed in low-productivity sectors such as agriculture, commerce and services, compared with 58.3% of male workers (Gender Equality Observatory for Latin America and the Caribbean, 2021[16]). Women are also disproportionately employed as domestic workers, with 14.3% of female workers in the region in the domestic work sector in 2018, compared with only 1% of men (ILO, 2019[17]). The concentration of women in commerce, domestic service and accommodation and food service activities has been associated with a high incidence of female part-time work and relatively low wages (ILO, 2016[18]). An ILO analysis of 10 world regions showed that 37.7% of employed women in Latin America and the Caribbean worked short weekly hours (35 hours or less), a higher share than the global average of 34.2% (ILO, 2016[18]). Gender inequality in weekly working hours was also much higher than the global average, with a gender gap of 19.6 percentage points in the LAC region (with only 18.1% of men working 35 weekly hours or less), almost twice as high as the global gap of 11 percentage points (ILO, 2016[18]).
Vertical segregation, on the other hand, refers to the difficulties women experience in developing professionally and gaining access to positions with greater decision-making power and better pay. Due to interacting factors such as gender stereotypes and prejudices, unsupportive employer policies, and lack of opportunities for gaining managerial experience, women tend to be employed at the lower levels of the hierarchical structure, and once in this position they usually remain trapped in the lowest-paying, lowest-ranking or least responsible jobs. This leads to a vicious cycle where a large proportion of women are excluded from economic decision-making and influence, further hindering progress towards gender equality (ECLAC, 2018[13]).
These, and other factors, imply that overall women in Latin America tend to earn less and are more likely to work in informal jobs. On average, across the countries considered, a gender pay gap exists in both hourly earnings (Figure 5.3, Panel A) and monthly earnings (Figure 5.3, Panel B) of employees. The difference is more striking and more consistent across individual countries for monthly earnings, a pattern that is consistent with the fact that women are more likely to work fewer hours overall. The gender pay gap is lower for the regional average (LAC) than for the focal group average (LAC 10 in Figure 5.3, Panel A, and LAC 9 in Panel B), which in turn is lower than for the OECD average. Trends over time are mixed: out of the six countries for which comparable time series on monthly earnings are available, half (Uruguay, Brazil and Paraguay) saw a marked reduction in the gender gap between 2010 and 2019, and half (Argentina, Costa Rica and Mexico) saw little change or even a slight increase (Figure 5.3, Panel B).
It should be noted that these data are based on earnings of employees only, and levels of pay are lower and gender differences larger when looking at the labour earnings of the self-employed. On average across the LAC region in 2017, the relative incomes of self-employed women and men were indexed at 81.6 and 87.6 respectively, when compared with a baseline of 100 for women’s total average labour earnings (ILO Regional Office for Latin America and the Caribbean, 2019[19]). The gap with the baseline for employed women and men was smaller (at 104.7 and 107.3, respectively). In general, women with significant unpaid work and domestic care responsibilities are more likely to be self-employed than those without (ILO Regional Office for Latin America and the Caribbean, 2019[19]).
Note: The gender pay gap is defined as the difference between mean hourly (monthly) earnings of men and women relative to mean hourly (monthly) earnings of men. Earnings of employees relate to the gross remuneration in cash and in kind paid to employees, as a rule at regular intervals, for time worked or work done together with remuneration for time not worked, such as annual vacation, other type of paid leave or holidays. For monthly earnings, data reported as weekly and yearly are converted to monthly in the local currency series, using data on average weekly hours if available. Data for earnings have been converted to constant 2017 international dollars using purchasing power parity rates for private consumption. In Panel A, the latest available year is 2017 for Chile and 2016 for Peru. The earliest available year is 2012 for Brazil, 2014 for Ecuador and 2015 for Chile and the Dominican Republic. LAC 9 excludes Argentina and Colombia, as data are not available. LAC regional average comprises 11 Latin American and Caribbean countries, including the 8 focal countries with available data for both the earliest and latest available years. OECD 30 excludes Australia, Canada, Colombia, Iceland, Japan, New Zealand and the United States, due to breaks in the time series or incomplete time series. In Panel B, the latest available year is 2017 for Chile. The earliest available year is 2011 for Costa Rica, 2012 for Brazil, 2013 for Mexico and 2014 for Ecuador. LAC 8 excludes Argentina, Colombia and Peru, as data are not available. LAC regional average comprises 9 Latin American and Caribbean countries, including the 6 focal countries with available data for both the earliest and latest available years. OECD 28 average excludes Australia, Canada, Chile, Colombia, Hungary, Iceland, Japan, Lithuania and New Zealand, due to breaks in the time series or incomplete time series. In Panel C, the latest available year is 2015 for Brazil. LAC 10 excludes Mexico, due to incomplete time series. LAC regional average comprises 20 Latin American and Caribbean countries, including the 10 focal countries with available data.
Source: OECD calculations based on ILOSTAT, https://ilostat.ilo.org/data/
While, globally, men are more likely than women to work in informal employment, in most lower to middle-income countries, including in the majority of LAC countries, the opposite is true (ILO, 2018[20]). On average, across the focal countries, 51.6% of female total employment was informal in 2019, compared with 49.2% of men (LAC 11, Figure 5.3, Panel C).12 These averages mask large differences in informality rates across countries, which were noted in Chapter 2. While informal workers of both sexes face a greater range of general and occupational risks than formal workers, women and men tend to face different types of vulnerabilities when working informally (OECD/ILO, 2019[21]). For example, men are more likely to suffer from the physical hazards of working in the unsafe, unregulated conditions associated with informal work, thereby experiencing much higher rates of occupational injury (both fatal and non-fatal) than women (ILO, 2021[22]). The risk of work-related injury or illness is further compounded by the low rates of health and social protection coverage among informal workers. However, men are more likely to work in top-tier informal employment (e.g. as employers), while women are more likely to be at the bottom of the hierarchy (Jutting and de Laiglesia, 2009[23]). Women are also more likely to work in low-status jobs that afford them little control over their working conditions or treatment, such as domestic work, home-based work or contributing family work, than their male counterparts (ILO, 2018[20]). These women may face specific issues associated with working in private homes, i.e. often in situations that are less protected by State regulations and off-limits to labour inspectors (ILO, 2016[24]). The power imbalance faced by women working in vulnerable informal conditions means that, in addition to the usual disadvantages of informal work (low pay, unsafe working environments, labour precarity, etc.), they are also more likely to experience sexual harassment and other forms of violence and gender-based discrimination (UN Women, 2020[25]).
Unpaid care and domestic work
Women’s relatively low participation in paid employment stands in contrast to their high participation in unpaid work in their own households. In Latin America, women take on over three-quarters (77%) of all unpaid work in the home, with care and home maintenance tasks being the most prevalent (ECLAC, 2018[13]). Overall, in the focal countries, women spend over twice as much time as men on unpaid work, with an average of 36.5 hours per week compared with 16.2 hours for men (LAC 11, Figure 5.4, Panel A). The gender gap in unpaid working time among the 11 focal countries, at 20.3 hours, is larger than both the LAC average (18.7 hours) and OECD average (14.8).13
The economic value of unpaid work is substantial: it is estimated at being equivalent to an average of 20% of GDP across 10 Latin American countries, with women accounting for 70% of this contribution (ECLAC, 2021[26]). This work provides a fundamental contribution to individual and social well-being, especially in terms of supporting the needs of vulnerable household members (children, the elderly, disabled people) in the absence of adequate public childcare and care structures. However, it remains a largely invisible and unrecognised aspect of work, the burden of which falls disproportionately on women, and it stands as a barrier to greater female participation in paid employment. The drivers of gender imbalances in unpaid work are various but are mainly linked to cultural factors (social norms that reinforce traditional gender stereotypes) and weaker labour market incentives for women (given the relative lack of well-paid, secure and rewarding job opportunities). The burden of unpaid care and domestic work increases for women at the lower end of the income distribution. Recent time-use data for 11 LAC countries14 show that women in the poorest quintile allocate approximately 6 hours to unpaid care and domestic work per day, compared with 2.5 hours for women in the richest quintile (UN Women, 2019[27]).
Female workers face a double burden, as they are faced with a larger share of unpaid work in addition to their paid employment (Figure 5.4, Panel B). On average in the focal countries, working women spend almost 10 hours longer on total work time (including both paid and unpaid work) than men, at 71.3 weekly working hours, compared with 61.9 for men. This gender gap is broadly similar to the regional LAC average gap, although the regional LAC average total working hours are slightly lower (67.9 total weekly hours for women and 57.9 for men).
Note: Data refer to 2019 for Mexico, 2017 for Brazil, Colombia and Costa Rica, 2016 for the Dominican Republic and Paraguay, 2015 for Chile, 2013 for Argentina and Uruguay and 2012 for Ecuador. LAC regional average comprises 11 Latin American and Caribbean countries (in Panel A) and 10 Latin American and Caribbean countries (in Panel B), in addition to the focal countries. OECD 30 excludes Chile, Colombia, the Czech Republic, Iceland, Israel, the Slovak Republic and Switzerland, as data are not available. Data for OECD countries are collected in the format of minutes/day dedicated to each activity and refer to the age group 15-64. Data are harmonised ex post by the OECD. The OECD average for weekly hours of unpaid work was calculated by taking the available time-use data on daily hours and multiplying them by 7. In Panel B, data for both women and men represent stacked totals.
Source: ECLAC Statistics, CEPALSTAT database, https://estadisticas.cepal.org/cepalstat/WEB_CEPALSTAT/estadisticasIndicadores.asp and OECD Time use (database), https://stats.oecd.org/Index.aspx?DataSetCode=TIME_USE
Violence against women and sexual autonomy
Latin America is one of the most unsafe regions in the world when it comes to violent crime, with men in the focal group 8.5 times more likely than women to die by homicide (Figure 5.1). However, other types of violence are missed by homicide statistics. While, globally, women are less likely to suffer violence in the context of armed conflict or criminal activity, they are more likely to experience violence and injury from intimate partners and other people close to them (Heise L and Garcia Moreno C, 2002[28]). Girls and women are also more likely to experience sexual violence and harassment overall (Jewkes, Sen and Garcia Moreno, 2002[29]), including outside of the home – at work, school and other public places (Gherardi, 2016[30]). While timely and internationally comparable data on the full range of violence and harassment experienced by women are lacking, there is widespread acknowledgement that gender-based violence is an urgent problem in Latin America (ECLAC, 2020[31]). It is becoming even more of a pressing issue since the pandemic and associated lockdown measures have exacerbated women’s exposure and risk in this domain (see below).
The consequences of violence against women differ in important ways from those applying to men. Physical and sexual violence against women brings a range of reproductive health consequences, such as sexually transmitted infection, premature birth, pregnancy loss and adolescent pregnancy15 (WHO, 2013[32]; Bott et al., 2012[33]). In order to avoid dangerous situations outside the home, women may restrict their behaviour, such as being more frequently absent from school or the workplace, which directly affects their academic and labour market outcomes, and their well-being overall (Gherardi, 2016[30]). Even the threat of potential violence is enough to reduce women’s freedoms, economic opportunities and quality of life. The trauma of experiencing violence can also lead to increased incidence of mental health problems such as depression and alcohol or substance abuse (WHO, 2013[32]). Finally, there is also an important family and intergenerational aspect to domestic violence: in homes where women experience violence from their partner, children are also more likely to experience violence themselves, both in childhood and later in life16 (Bott et al., 2012[33]).
Overall, across the 11 focal countries, one in four women aged 15-49 (25.6%) have experienced some form of intimate partner violence (either sexual, physical or both) in their lifetime (Figure 5.5, Panel A). While this is only slightly higher than the OECD average (23.1%), estimates are not directly comparable, as the OECD average refers to a larger population (women aged 18-74). Some focal countries also have data on the incidence of intimate partner violence in the previous year (SDG indicator 5.2.1); in both Colombia and the Dominican Republic, over half of those who reported some lifetime experience of partner violence also reported experience in the past 12 months. These numbers certainly underestimate the true prevalence of domestic violence, as evidence shows that the majority of cases go unreported (Gracia, 2004[34]).
Note: In Panel A, the chart shows the incidence of intimate partner violence (IPV) committed by any lifetime partner, except for Colombia and Peru, which shows IPV committed by the current or most recent partner. The asterisk denotes that data for Chile are not directly comparable as they refer to sexual assault only. Data refer to 2017 for Brazil and Peru, 2016-2017 for Chile, 2016 for Mexico, 2015 for Argentina and Colombia, 2013 for the Dominican Republic and Uruguay, 2011 for Ecuador, 2008 for Paraguay and 2003 for Costa Rica. Data pertaining to the past year are not available for Ecuador, Mexico, Chile and the OECD average. The OECD average refers to the population aged 18-74. The LAC regional average comprises 22 Latin American and Caribbean countries, in addition to the focal countries. In Panel B, the chart shows data for deaths classed as femicide. The left axis shows rates per 100 000 women, while the right axis shows absolute numbers. The LAC regional average comprises 19 Latin American and Caribbean countries, in addition to the focal countries.
Source: National surveys as harmonised in (Bott et al., 2012[33]) and OECD Family Database, http://www.oecd.org/social/family/database.htm (Panel A); CEPALSTAT, https://estadisticas.cepal.org/cepalstat/WEB_CEPALSTAT/estadisticasIndicadores.asp (Panel B)
Femicide is the most extreme form of violence against women. It strengthens gender divisions, upholds male dominance and disempowers women by rendering them chronically and profoundly unsafe (GHRC - USA, n.d.[35]). While there is no international definition of femicide, a shared understanding is that it does not simply refer to the murder of women, but the murder of women by men because they are women (Russell, 1976[36]). Femicides can be motivated by hatred, contempt, pleasure or a sense of ownership over women (Caputi and Russell, 1990[37]). There is also evidence to suggest that although guns are the most widespread means of intentional killing in Latin America, women are more likely than men to die because of suffocation, strangulation or beating (INEGI, 2019[38]).
In 2019, at least 4 676 women were victims of femicide across 18 Latin American countries, according to available data, and there were at least 3 821 femicides in the 11 countries of the focal group (ECLAC, 2019[39]). This corresponds to an average femicide rate of 1.3 per 100 000 women in the LAC 11 focal group and 2.6 per 100 000 women across the LAC region. The higher LAC average rate reflects the exceptionally high levels of femicide observed in recent years in a number of Central American and Caribbean countries such as El Salvador, Honduras and Santa Lucia. Comparing rates of femicide between regions is not straightforward, as definitions and data sources can differ. However, to give some context, on average across 16 European countries for which data were available, 0.53 women per 100 000 were killed by an intimate partner or family member in 2018 (Eurostat, 2021[40]) (although this is based on a narrower definition of femicide that excludes gender-related deaths outside of the home or family).
Violence against women is a global phenomenon, with complex causes. It is not a private, personal issue shaped only by individual factors, but a deep-seated and urgent social problem. Social realities that drive gender-based violence include structural aspects (such as conflict, poverty or lack of economic opportunities for women and girls), cultural factors (such as harmful gender norms) and discriminatory formal and informal institutions (such as racism, inadequate legal frameworks, lack of access to justice, and property ownership rules) (Michaeljon, Bell and Holden, 2016[41]). The OECD Development Centre’s Social Institutions and Gender Index provides comparative evidence on the role of formal and informal social institutions in shaping gender inequality (Box 5.1) in Latin America.
The Social Institutions and Gender Index (SIGI), developed by the OECD Development Centre, measures discrimination against women in social institutions across 180 countries. By taking into account laws, social norms and practices, the SIGI captures the underlying drivers of gender inequality, with the aim of promoting gender-transformative policies that are built on data and evidence. The SIGI is also one of the official data sources for monitoring Sustainable Development Goal (SDG) indicator 5.1.1.
The SIGI and its dimensions look at the gaps that legislation, attitudes and practices create between women’s and men’s rights and opportunities. The SIGI covers four dimensions, spanning major socio-economic areas that affect women’s and girls’ entire lifetimes:
The “Discrimination in the family” dimension captures social institutions that limit women’s decision-making power and undervalue their status in the household and the family.
The “Restricted physical integrity” dimension captures social institutions that increase women’s and girls’ vulnerability to multiple forms of violence and limit their control over their bodies and reproductive autonomy.
The “Restricted access to productive and financial resources” dimension captures women’s restricted access to and control over critical productive and economic resources and assets.
The “Restricted civil liberties” dimension captures discriminatory laws and practices restricting women’s access to, and participation and voice in, the public and social spheres.
Note: Scores range from 0 to 100, with 0 indicating no discrimination and 100 indicating absolute discrimination.
Source: Social Institutions and Gender Index, http://stats.oecd.org
The SIGI is a composite index, with scores for the overall index, its dimensions and indicators ranging from 0 (in the case of no discrimination) to 100 (in the case of full discrimination against women). Figure 5.6, which presents a summary of SIGI dimension scores in the LAC region and its sub-regions, shows that “Discrimination in the family” is the dimension featuring the worse performance, underscoring the presence of deep discrimination in intra-household dynamics between men and women. All three sub-regions tend to score similarly in the other dimensions, with the exceptions of the Caribbean in the “Restricted civil liberties” dimension and of South America in the “Restricted access to productive and financial resources” dimension. The Caribbean’s poor performance in “Restricted civil liberties” is primarily the result of weak legal frameworks governing women’s ability to confer their nationality to their husband or children; low incentives for women’s political participation and representation; and weak guarantees of equal rights for women who apply for passports and travel documents. South America’s poor performance in the “Restricted access to productive and financial resources” dimension stems from weak legal frameworks governing women’s rights in the workplace, and the prohibition on women entering certain professions in some countries.
In terms of the region’s performance in a global context, the overall SIGI score for the LAC region is 25.4, which is higher (i.e. implying poorer performance) than in Europe (17) and North America (18.1, when excluding Mexico), but lower than in Africa (40.3) and Asia (35.9).
Source: OECD (2020[42]), SIGI 2020 Regional Report for Latin America and the Caribbean, Social Institutions and Gender Index.
While not an indicator of violence per se, high rates of adolescent fertility in the LAC region affect women’s well-being in a variety of ways. While there has been a dramatic reduction in fertility rates in Latin America and the Caribbean, they remain high among adolescent women (Ullman, 2018[43]). Adolescent motherhood has consequences across a range of dimensions of young women’s well-being in Latin America, exacerbating the intergenerational transmission of poverty and deprivations in educational attainment (ECLAC, 2014[44]; ECLAC/UNICEF, 2007[45]), and implying an infringement of young people’s access to sexual and reproductive health information and services. Another related violation of human rights is child marriage, which appears under SDG 5.3 and disproportionately affects girls. The situation in Latin America and the Caribbean varies widely from one sub-region to another: 15% of girls aged 15-19 are or have been married or are in informal unions in the Caribbean, against 20% in Central America. Across the region as a whole, child marriage rates have remained stable over the last 30 years, with the Dominican Republic featuring among the top 20 countries internationally with the highest prevalence of child marriage (OECD, 2019[46]). Child marriage and adolescent fertility rates are highly correlated in the LAC region and worldwide: where child marriage is more pervasive, adolescent fertility rates are also higher (OECD, 2020[42]).
Women’s political participation
Representation in political decision-making is central to achieving an inclusive and gender-equal society. Countries in the focal group have made substantial progress in this regard, with the average share of women in parliament almost doubling since 2000, from 14.8% up to 29.2% in 2019 (Figure 5.7). Mexico and Costa Rica came close to achieving full gender parity by 2019 (with female parliamentary representation of 48.2% in Mexico and 45.6% in Costa Rica). The increase in female parliamentary representation was greater over the reference period across the focal group than for the OECD average, meaning that although female representation was higher in the OECD at the start of the 2000s (at 19.6%), the average OECD level of female parliamentary representation was similar to that in the focal group by 2019 (at 30.2%).
Legislation is an effective way to increase the participation of women in the political sphere, and a growing number of countries in Latin America (less so in the Caribbean) have established political-electoral gender parity laws. Currently, three groups can be identified in terms of progress with gender quotas: in the first group, a total of 10 countries (Bolivia, Costa Rica, Ecuador, Nicaragua, Mexico, Honduras, Panama, Argentina, Peru and Colombia) have enacted regulations to stipulate complete gender parity in popularly elected positions; the second group (Brazil, Chile, El Salvador, Haiti, Paraguay, the Dominican Republic, Uruguay and Guyana) have implemented various gender quotas with percentages for positions ranging from 20% to 40%; and the third group have no parity or quota stipulations for popularly elected positions (UN Women, 2021[47]). While, depending on how these measures have been implemented and enforced, this has helped to normalise the participation of women in the public sphere and facilitated women’s access to political representation, this progress cannot be taken for granted. Indeed the very fact that legal mechanisms are necessary shows that improvements in gender equity are not automatic in this area, and where laws have been instated, efforts to resist their application or limit their effectiveness generally follow (UN Women, 2021[47]). For example, at the local level, where quotas are less applied and enforced, women obtained only 15.2% of mayoral positions in the 2018-2019 elections across the LAC region, compared to 5% in the 1990s (UN Women, 2021[47]). In addition, improvements in women’s access to public or popularly elected positions have not translated into a presence that reflects their diversity in terms of Indigenous or Afro-descendant status, sexual orientation, or other marginalised identities or statuses, and more efforts are needed to improve this situation (UN Women, 2021[47]). Finally, as elsewhere, women in the public political sphere in the LAC region continue to face threats in terms of physical violence and online intimidation, risks that have been exacerbated through the rise of openly discriminatory rhetoric in ultra-conservative discourse (UN Women, 2021[47]).
Note: LAC is the regional average for Latin America and the Caribbean as calculated by ECLAC.
Source: ECLAC Statistics, CEPALSTAT database, https://estadisticas.cepal.org/cepalstat/WEB_CEPALSTAT/estadisticasIndicadores.asp?idioma=I, and IPU Inter-Parliamentary Union, Women in Parliaments, for OECD average, http://archive.ipu.org/wmn-e/world-arc.htm
Impact of COVID on gender inequalities
The economic, social and health impacts of the pandemic have been very different for men and women. Integrating a gender perspective into policy responses will therefore be fundamental to the efficacy of mitigation and recovery efforts (UN Women, 2020[48]).
Regarding the health consequences, clear gender disparities have emerged through the course of the COVID-19 pandemic. As of February 2021, more women were being tested (57%) than men throughout the world, and they accounted for slightly more than half of all confirmed cases (51%). However, men made up a higher share of reported hospitalisations (53%), intensive care admissions (68%) and deaths (57%) globally (Global Health 50/50, APHRC and ICRW, 2021[49]), reflecting higher incidence of chronic diseases (i.e. hypertension) and of risky and or health-reducing behaviours (i.e. smoking), as well as immunological differences (World Bank, 2020[50]). However, there are still many unknowns, and while the availability of data by gender has improved during the course of the pandemic, as of February 2021 only 51% of countries reported sex-disaggregated case data and only 41% reported sex-disaggregated death data (Global Health 50/50, APHRC and ICRW, 2021[49]).17
While women experience lower fatality rates overall, they are more likely to work in paid and unpaid roles with high levels of exposure to the virus, such as frontline healthcare roles and jobs in sectors that require women to interact with other people during the confinement phase (such as agriculture or domestic work) (World Bank, 2020[51]). This is especially true in Latin America, which has the highest share of female healthcare workers in the world (half of doctors and more than 80% of nurses) (Inter-American Development Bank, 2018[52]), in addition to the very high share of women working in agriculture and domestic services.
Beyond the direct health impacts of the pandemic, the economic and social consequences are differentiated by gender in a number of key areas. As shown above, women in the region already faced vulnerabilities on a number of fronts before the onset of the pandemic, hence the danger that the subsequent economic and social crises will further undermine women’s autonomy and deepen structural inequalities (see Figure 5.8).
Source: (ECLAC, 2021[53])
Overall across the region, women have experienced disproportionately negative outcomes in labour market indicators, due to their over-representation in sectors that have been more affected by pandemic control measures (such as restaurants and hotels, commercial activities and domestic services) (ECLAC and ILO, 2020[54]). The female unemployment rate is expected to reach 22.2% for 2020, a 12.6 percentage point increase year-on-year (ECLAC, 2021[53]). Latin American women have experienced a greater proportional fall in employment (by 18.1%, compared with 15.1% for men), as well as greater exits from the labour market (15.4%, compared with 11.8% for men) (ECLAC and ILO, 2020[54]). In total, the negative impact of the pandemic is expected to wipe out a decade’s progress in increasing women’s labour market participation in Latin America (ECLAC, 2021[53]).
The high rate of women withdrawing from the labour market was likely due to them taking on an even greater unpaid work burden related to increased care responsibilities, home schooling and other tasks during the pandemic (ECLAC and ILO, 2020[54]; OECD, 2020[42]). In addition to deepening gender inequalities in unpaid and paid work time, the increased unpaid workload is undoubtedly having a mental health impact, exposing women to higher levels of stress and anxiety. A survey conducted during the quarantine period in Chile revealed that women experienced higher prevalence of symptoms of mental health problems than men, and that they felt more overwhelmed and under stress (63.3%, compared to 46.3% among men (ECLAC, 2021[3]). Higher rates of poverty amongst women before the pandemic may also deepen gender inequalities in income and poverty. It is estimated that 118 million women in the region will be living in absolute poverty following the crisis (compared with a total poor population of 187 million in 2019) (ECLAC, 2021[53]; ECLAC, 2021[3]).
Confinement measures taken to limit the spread of the virus have likely increased the risk of violence, exploitation and harassment faced by women. The frustration and uncertainty caused by lockdown situations can lead to anger amongst men that manifests itself through increased violence against women, both within and outside the home (OECD, 2021[55]; OECD, 2020[56]). Further, travel restrictions, increased economic dependency and disruptions to support services mean that abused women may be trapped in dangerous situations (OECD, 2020[57]). There is a widespread perception that the scale of violence against women in Latin America has become a “shadow pandemic”, although timely, high-quality data are lacking to fully understand the scope of the problem (UN Women, 2020[58]). Available data show mixed outcomes across countries. For example, amongst the countries in the region that have released data on calls to help centres for March-June 2020, calls increased year-on-year compared with 2019 in Mexico, Paraguay and Peru, while they fell in other countries, such as Ecuador and the Dominican Republic (ECLAC, 2021[3]). However, these trends need to be interpreted with caution, as a drop in calls may not correspond to lower rates of violence, as women are likely to face greater limitations on the use of hotlines during confinement periods. Available femicide data are also mixed but show a decrease in the number of reported cases in eight of the ten countries for which data are available (Chile, Colombia, Costa Rica, Dominican Republic, Ecuador, Honduras, Paraguay and Peru), with data for Panama remaining stable, and those for Mexico pointing to an increase in March-June 2020, compared with the same period in 2019 (ECLAC, 2021[3]).
At the time of writing, data from the Gallup World Poll (referring to 2020) shed some light on the impact of the first months of the pandemic on people’s well-being across a number of dimensions (see Box 5.2). Between 2019 and 2020, the share of women saying they were satisfied with their living standards or that they had someone to count on for support fell more than for men; women’s life satisfaction also declined more than for men (Figure 5.9).
At the time of writing, very little internationally comparable data were available to compare well-being outcomes before and after the onset of the pandemic, and even less data allowing for comparisons between different population groups. However, 2020 data from the Gallup World Poll were available for all focal countries and for a majority of OECD countries, providing some indication of the impact of the first year of the pandemic on various aspects of self-reported well-being. Given the way the pandemic unfolded over the year, with peaks and troughs of the virus infection rate occurring at different times in different countries, the timing of the survey fieldwork is critical (OECD, forthcoming[59]). In the focal countries, Gallup data collection for 2020 ranged from 21 August 2020 to 6 Jan 2021.1 Although this encompasses a wide timeframe, the pandemic was already well-advanced even at the time of the earliest surveys. While it is likely that understanding the full impact of the pandemic on the different aspects of people’s well-being will not be possible for several years, the data available from the Gallup World Poll can give some initial insights into a number of subjective dimensions of people’s lives based on a range of self-reported indicators. Figure 5.9 shows the differentiated impact of the first months of the pandemic for three variables: satisfaction with living standards (Panel A), life satisfaction (Panel B) and social network support (Panel C).
Across the three indicators shown, the magnitude of the impact and the level of differentiation between groups was larger for the focal countries than for the OECD average. Across OECD countries, satisfaction with living standards increased for all groups with available data between 2019 and 2020 (educational breakdowns for the OECD average were not available with sufficient data coverage to include in the analysis), a reflection of the buffering role that government policies played in protecting people’s material conditions (through job retention schemes, more generous unemployment and social assistance benefits, etc.) during the first year of the crisis (OECD, forthcoming[59]).
However, across the focal group countries, the average share of people satisfied with their living standards decreased very slightly (from 72% in 2019 to 71% in 2020). While changes in satisfaction with living standards were small for all groups, Figure 5.9 highlights some differences in both the magnitude and direction of the various indicators. Overall, among the focal countries, the share of people satisfied with their living conditions fell slightly more for women than for men; it fell in rural areas while increasing slightly in urban areas; it fell among young people while increasing slightly among middle-aged and older adults; finally, it increased by five percentage points (from 75% in 2019 to 80% in 2020) among people with tertiary education, while remaining broadly stable for those with primary education (up by 1 percentage point) and for those with secondary education.
The differences in life satisfaction and social network support are starker. While in OECD countries, both indicators changed only marginally between 2019 and 2020, changes were much larger across the focal countries: the decreases in life satisfaction were greater for women (-0.5 points) than for men (-0.3), for those living in rural areas (-0.6) than for those in urban centres (-02), for young people (-0.6) than for prime age (-0.5) and older people (-0.2), and for the low-educated (-0.5) than for those with tertiary education (-0.04). Women and those living in rural areas also experienced the largest decreases in social network support, while slightly different patterns can be observed when considering age (with people aged 30-49 showing the largest reduction) and education (with those with secondary education showing a larger decrease than others).
Note: Selected indicators from the Gallup World Poll.
Source: OECD calculations based on the Gallup World Poll (database), https://gallup.com/analytics/232838/world-poll.aspx
StatLink https://stat.link/xuktbv
1. Data collection dates for 2020 were as follows: Sep 7 – Nov 20, 2020 in Argentina; Sep 10 – Nov 11, 2020 in Brazil; Sep 11 – Nov 16, 2020 in Chile; Aug 21 – Oct 27, 2020 in Colombia; Sept 15, 2020 – Jan 4, 2021 in Costa Rica; Sep 24 – Oct 23, 2020 in the Dominican Republic; Aug 26 – Oct 23, 2020 in Ecuador; Sep 08 – Nov 18, 2020 in Mexico; Nov 28 - Dec 28, 2020 in Paraguay; Oct 29, 2020 – Jan 6, 2021 in Peru; and Sep 24 – Nov 30, 2020 in Uruguay.
Issues for statistical development in gender statistics
Improving the availability of high-quality and comparable gender statistics is central to achieving a better understanding of the realities of women and girls and for designing policies that effectively address their needs. The importance of gender statistics for monitoring well-being and sustainable development has been recognised by both governments and statistical offices in the LAC region, especially in the context of the UN 2030 Agenda.18 The pandemic has further underlined the need for gender-specific information to inform effective policy responses and recovery strategies. Many Latin American statistical offices have prioritised the collection of sex-disaggregated data (such as on labour market outcomes) despite the additional pressures and limitations they have faced due to Covid-19, often through innovative approaches such as adapting existing operations, generating new statistical operations or improving alternative sources and administrative records (ECLAC and UN Women, 2021[60]).
Beyond improving the availability of sex-disaggregated data wherever possible, better data are needed on a range of specific and under-measured issues that affect women and girls uniquely or disproportionately, such as discrimination in the workplace, sexual harassment, unpaid work, reproductive health and autonomy, economic autonomy and different forms of gender-based violence.
Time-use surveys are a particularly rich source of information on activities performed by men and women and on the distribution of time spent on these activities by gender. Time-use measurement has a long history in the region, with gender-focused work on time-use issues gradually developed over the last four decades through the Regional Gender Agenda in Latin America and the Caribbean (ECLAC, 2019[61]). In 2015, the member states of the Statistical Conference of the Americas of the Economic Commission for Latin America and the Caribbean adopted the Classification of Time-Use Activities for Latin America and the Caribbean (CAUTAL) in order to harmonise time-use surveys in the region (ECLAC/INEGI/INMUJERES/UN-Women, 2016[62]). As of 2019, 19 countries in the region had run at least one-time-use survey (ECLAC, 2019[61]).
However, not all of these surveys are fully incorporated into the system of official statistics as a regular data collection tool, and the CAUTAL classification system is not yet universally applied by countries (ECLAC, 2016[63]). In addition, vulnerable groups such as the rural population and ethnic and racial minorities are often under-represented in survey samples (ECLAC, 2016[63]). There are also issues to consider related to the most effective mode of time-use data collection. In recent years, two main approaches have been used, i.e. either including a short list of questions on time use as a module within existing household surveys or conducting a stand-alone survey collecting information on the breadth of time-use activities in more detail. The advantage of the former is that it is more cost-effective and allows for the joint analysis of time use with other modules of the survey. The latter provides much richer information, but at a higher cost. Ideally, both modes could be employed, with a repeated inclusion of a limited number of questions in regular household surveys supplemented by lower frequency surveys to provide more context. For this to happen, time-use measurement should be integrated as a core aspect of national statistical planning and budgeting (Villatoro, 2017[64]). Further, a harmonised approach to time-use measurement should be applied in a consistent manner so as to ensure the comparability of results across countries and over time. Finally, as far as possible, including a representative sample of the most vulnerable populations would shed light on the links between deprivations in time use and other forms of disadvantage experienced by vulnerable women.
The life cycle can be categorised into four basic stages: childhood, youth, adulthood and old age. While in terms of well-being each stage presents its own opportunities, risks and challenges, childhood, youth and old age are times of particular vulnerability. The well-being of children is highly dependent on their family and surroundings, and experiences in early life can be fundamental to determining outcomes across the life course (OECD, 2021[65]). As children grow into young adults, they gain independence, but their ability to thrive with more autonomy often depends on a successful transition to working life and on the skills and opportunities that support this. They also have to navigate the demands of moving away from a dependent role within their families to starting families of their own (with these roles themselves highly dependent on gender). Eventually, as individuals move through adulthood into old age, they once again enter a stage of greater dependency, with increased needs for health care and other support (OECD, 2017[66]; Cecchini et al., 2015[67]).
This section will take a closer look at these life cycle differences, focusing on childhood, youth and old age, compared to adulthood. A number of indicators referring to the well-being of children and young adults have already been covered in previous chapters due to their relevance to broader societal outcomes.19 Those indicators, pertaining to child mortality, child malnutrition, youth employment and educational attainment, will not be covered in detail here, but references to figures featuring in other sections of this report will be provided as needed.
The COVID-19 pandemic has the potential to exacerbate intergenerational differences in well-being outcomes in Latin America. Children are among the “hidden victims” of the pandemic; despite being spared from high rates of mortality due to the virus, they have been heavily impacted by disruptions at all levels, particularly children in households where pre-existing stressors have been accentuated by the crisis. The pandemic has also exposed vulnerable teenagers and young adults to higher risks of disengagement and dropout from education and training, in a region where youth unemployment is already high. Finally, the COVID-19 outbreak has posed severe challenges for older people, who are not only at higher risk of serious health complications in case of infection, but also disproportionately affected by confinement measures restricting their access to care and support.
For youth and the elderly, this section shows figures summarising outcomes relative to the “middle-aged” adult population. Generally speaking, the youth group covers the population aged around 15 to 29 (thus having some overlap with children), while the middle-aged group covers the population aged around 30 to 55, and the elderly population those aged over 55. However, the exact age range used differs for each indicator, depending on the available information, with more information provided in the Statlink for each figure.
Childhood is a critical period for determining factors involved in individual development that will continue to shape well-being throughout the life course. In this sense, experiences in childhood matter for both the well-being that children enjoy today, and for the resources that will help to sustain societal well-being over time. In 2019, children aged 0-14 made up just below a quarter (24%) of the population in Latin America and the Caribbean (World Bank, 2020[68]). Research highlighting links between well-being in childhood and in adulthood is extensive, particularly with regards to the influence that family conditions and children’s early experiences have on educational outcomes in later life (OECD, 2021[65]; OECD, 2015[69]). Because children are dependent members of society, their well-being largely depends on the well-being of their own families and communities.
Child poverty
Growing up in poverty is harmful to children’s well-being and development, both in the short term and in the long run as adults (Thévenon et al., 2018[70]). Childhood poverty has certain specificities that heighten children’s vulnerability. Given the dependence of children on their families, poverty may also be cumulative for children and adolescents, and there is a strong intergenerational component to child poverty. There is extensive evidence that those who live in poor conditions at an early age are more likely to experience poverty as adults (Kendig, Mattingly and Bianchi, 2014[71]). Finally, the effects of childhood poverty may be irreversible, as in the case of malnutrition or recovery from preventable disabilities (UNICEF/CEPAL, 2019[72]). As a rule in Latin America and the Caribbean, the lower the age group, the higher the incidence of poverty (ECLAC, 2018[13]). In 2019, 31% of children aged 0-14 were living in absolute income poverty in the focal group of countries, compared with 17% for those aged 25-54 (Figure 5.10, Panel A.). Extreme poverty rates followed a similar pattern, affecting 9% of children aged 0-14, compared with 4% of the 25-54-year-old population (Figure 5.10, Panel B.). Across the focal group, results vary greatly but are generally in line with the national levels described in Chapter 2. Thus, in Mexico, the share of children aged 0-14 living in absolute poverty is almost nine times higher than in Uruguay (Figure 5.10, Panel A.).
Note: Data on absolute and extreme poverty refer to the regional poverty and extreme poverty rates as calculated by ECLAC. Data refer to 2019 for all Latin American countries except for Mexico (2018) and Chile (2017). LAC is the regional average for Latin America and the Caribbean as calculated by ECLAC. LAC 10 average excludes Argentina due to a lack of available data.
Source: ECLAC Statistics, CEPALSTAT database, https://estadisticas.cepal.org/cepalstat/WEB_CEPALSTAT/estadisticasIndicadores.asp
Child labour
One of the main consequences of child poverty is driving children into the workplace (Thévenon et al., 2018[70]). Generally, children tend to work because their own material conditions and those of their families depend on it, as child labour is part of how families, especially poor ones, buffer negative shocks to income (Thévenon et al., 2018[70]). Child labour appears to be more sensitive to changes in permanent household income and adult wages than to changes in children's wages. To make matters worse, children are naturally vulnerable, and adults may take advantage of this. The consequences of child labour affect almost all dimensions of life. Beyond impacts on their physical health and psychological and social development, child labourers tend to have limited access to school, reduced safety and less time for leisure and interactions with friends and family (Santana, Kiss and Andermann, 2019[73]).
A number of Latin American countries need to make further progress in order to reach the target set by SDG 8.7 of ending child labour in all its forms by 2025 (UNDESA, 2020[74]). On average across the focal group, 5% of children aged 10-14 are employed according to the latest data. In Mexico, more than one in ten children aged 10-14 years are employed, compared with one in five hundred in Chile (Figure 5.11, Panel A.). The prevalence of paid child labour is twice as high for boys (11%) as for girls in the focal group (Figure 5.11, Panel B).20
Child labour is also more common in rural areas (10%) than in urban areas (3%) in the focal group, and over half of all child labour (52%) is in agriculture (ILO, 2017[75]). Child labour is concentrated in the lowest income quintile across the focal group (7%). Nonetheless, child labour is also present when looking at the higher income quintiles (3.7% in Quintile 4 and 2.8% in Quintile 5) in the focal group, indicating that poverty is not the only factor that determines child labour (Figure 5.11, Panel B.). Finally, while data are available for only a limited selection of countries, child labour is much more prevalent amongst Indigenous communities, especially for those aged in their mid-teens (Figure 5.11, Panels C and D). In Ecuador, Peru, Brazil and Mexico, between 30.4% and 43.5% of Indigenous children aged 14-17 work, shares that are much higher than among their non-Indigenous peers.21
Note: In Panel A and B, the latest available year is 2019 for Argentina and 2015 for Chile and Peru. In Panel C and D, the latest available year is 2015 for Brazil and Chile.
Source: SEDLAC database (CEDLAS and The World Bank), https://www.cedlas.econo.unlp.edu.ar/wp/en/estadisticas/sedlac/, and ECLAC (2020), "Los pueblos indígenas de América Latina – Abya Yala y la Agenda 2030 para el Desarrollo Sostenible: desafíos desde una perspectiva territoria"
Child malnutrition
Although child health has improved in many respects, many children in Latin America remain vulnerable and face severe risks – some of which are specific to their age group. International development initiatives such as the SDGs have contributed towards improving child health and monitoring the impact of specific actions in Latin America (Arnesen et al., 2016[76]; Grove et al., 2015[77]), and the region has made progress in reducing child mortality in the past two decades. This is reflected in a decrease not only in the number of children who die before reaching the age of five (see Chapter 3) but also in the number of children affected by diarrheal diseases and pneumonia (PAHO, 2017[78]). A core component of human capital is for people to be well-nourished throughout their lives, yet many children in Latin America are unable to access sufficient nutritious food nor attain a balanced diet that meets their needs for optimal development and growth, ultimately enabling a healthy, active life (OECD/The World Bank, 2020[79]). Malnutrition at an early age has consequences in other areas of well-being such as cognitive and educational outcomes later in life, shaping an individual’s long-term socio-economic status (OECD/The World Bank, 2020[79]). As part of the UN 2030 Agenda, SDG target 2.2 aims to end all forms of malnutrition by 2030 (UNDESA, 2020[74]).
Chapter 4 showed that in the five focal countries where data are available (Argentina, Colombia, Mexico, Peru and Paraguay), one in ten children below five years old are stunted on average (Figure 4.18, Panel A), with this share ranging from below 2% in Chile to almost 13% in Colombia. On average, stunting rates have almost halved since 2000, with the largest drop in Paraguay and Peru (by more than 10 percentage points) and the smallest in Argentina and Chile (by 1 percentage point or less), countries where stunting rates were already below the regional average.
Overweight and obesity are another consequence of malnutrition. Chapter 4 showed that on average in the focal countries, almost 60% of the adult population is overweight and 25% is obese, up from 50% and 21% respectively in 2000. While the prevalence of overweight tends to increase with age, overweight in childhood is nonetheless significant in the LAC region. While the share of children under age 5 who are overweight changed little between 2000 and 2020 across the focal countries (from 7.8% to 8.8%), a much greater increase occurred among children aged 5-19 (Figure 5.12, Panel A), rising from 22% in 2000 to 31% in 2016, a level that exceeds the LAC regional average (29.5%) by 1.5 percentage points, and the OECD average (29%) by 3 percentage points.
Note: In Panel A, LAC regional average comprises 27 Latin American and Caribbean countries, including the focal countries. OECD 16 includes Australia, Belgium, Canada, Chile, Colombia, the Czech Republic, Estonia, Germany, Greece, Japan, Mexico, the Netherlands, Poland, Portugal, the Slovak Republic and the United States, due to missing data for other OECD countries. In Panel B, LAC regional average comprises 33 Latin American and Caribbean countries, including the focal countries.
Source: WHO, https://apps.who.int/gho/data/view.main.CHILDOVERWEIGHTv (Panel A) and https://apps.who.int/gho/data/node.imr.NUTOVERWEIGHTPREV?lang=en (Panel B)
One-quarter of the Latin American population are aged between 15 and 29, and two-thirds of this age group (over 100 million young people) live in poor or vulnerable households (OECD/CAF/ECLAC, 2016[80]). Further, most youth, especially those from households in the bottom of the income distribution, have access only to poor quality services and precarious jobs, while having low savings and experiencing little social mobility. This sharp disconnect between society’s expectations and demands on the one hand and actual socio-economic outcomes on the other has fuelled social dissatisfaction and weakened trust in democratic institutions (OECD/CAF/ECLAC, 2016[80]). Figure 5.13 summarises some of these intergenerational disparities in the focal group of countries. As mentioned above, the youth category focuses on people aged around 15-29 and the middle-aged category on adults aged around 30-55, although the exact age range differs for each indicator (see Statlink for Figure 5.13 for more details).
In the selected indicators of material conditions (Figure 5.13, Panel A), on average across the focal countries, in 2019 young people are only half as likely to be employed as middle-aged adults (with an employment rate of 39% for 15-24 year-olds, compared with 77% for 25-54 year-olds. While this may reflect the fact that younger people are more likely to be in education or other activities, their unemployment rate is three times higher (at 18.8%, compared with 6.1% for the comparison group), suggesting that young people actively seeking employment have a harder time to enter the labour market than their older peers. Young people are also more likely to work in informal employment than the middle-aged comparison group (with an informal employment rate of 64% for 15-24 year-olds, compared to 48% for 25-54 year-olds in 2019). Lack of decent employment opportunities is one of the most significant factors affecting the inclusion of youth in countries of the focal group and in the region more widely, and there are strong links between informal employment, poverty and social exclusion (ILO, 2015[81]). Indeed, young people are more likely to be in absolute poverty and extreme poverty compared with middle-aged adults. Nonetheless, young people are 13% more likely to be satisfied with their living standards than middle-aged adults.
Regarding quality of life, the picture is more balanced. As health deteriorates with age, young people have much better health than the middle-aged across the focal countries. For example, they are half as likely to say that they have health limitations that prevent them from doing usual activities and 73% less likely to have a negative balance of emotions (i.e. to experience more negative than positive emotions in a given day), and they report higher levels of life satisfaction, social network support and satisfaction with education and health services. However, although there is no difference in levels of perceived safety reported by young people and the middle-aged, young people are 31% more likely to be the victim of homicide, particularly among young men (see below). Young people are also 17% more likely to commit suicide than the middle-aged across the focal countries.
Finally, across the selected indicators of social capital, young people are less likely to voice their opinion to an official, less likely to trust police, less likely to say that tax avoidance is completely unjustifiable and less likely to volunteer. However, they are slightly (8%) more likely to trust their national government. Finally, there is little clear difference between youth and the middle-aged for trust in others, perceived corruption, perceived inequality (the share of people thinking the income distribution is unfair) and support for democracy over all other forms of governance.
Note: Each performance ratio is the simple average by age group calculated across the 11 focal countries for which data are available for all the inequality groups taken into consideration. As a result, performance ratios can cover a subset of the 11 focal countries. Performance ratios above 1 indicate better outcomes (i.e. higher well-being) for youth, whereas performance ratios below 1 indicate better outcomes (i.e. higher well-being) for the middle-aged – including negative indicators, which have been reverse-scored. Grey bubbles denote no clear difference between youth and middle-aged, defined as age ratios within 0.03 points distance to parity.
Source: OECD calculations. For a complete list of sources please refer to the "Sources and Methods" tab in the Statlink file
Homicide and violence
Homicide is by far the most important cause of death among young people in Latin America, with young men accounting for the large majority of both victims and perpetrators (UNODC, 2019[82]). The average rate of homicide for young people was 23 per 100 000 population in 2017 in the focal countries, much lower than the LAC regional average (44 per 100 000) but still over five times higher than the OECD average (4.3 per 100 000) (Figure 5.14, Panel A). Across the focal countries, young men are over nine times more likely to die from homicide than young women, with a male youth homicide rate of 42 per 100 000 compared with 4.5 per 100 00 for female youth.
Note: Youth are defined as people aged 15-29. Middle aged are defined as people aged 30-59. Data refer to 2016 for all countries, except for Argentina and Mexico (2017) and Costa Rica, the Dominican Republic and Paraguay (2014). LAC regional average comprises 25 Latin American and Caribbean countries, including the focal countries. OECD 31 excludes Belgium, Ireland, Korea, Lithuania, the Slovak Republic and Turkey, due to missing data.
Source: UNODC, https://dataunodc.un.org/data/homicide/Homicide%20by%20sex%20and%20age%20group
The drivers of the rise of violence, and particularly of violence linked to organised crime, in the LAC region are complex. However, poverty tends to exacerbate the likelihood of young people becoming involved in criminal activities with a heightened risk of violence. Criminal organisations such as gangs provide young Latin Americans with a sense of identity and belonging: when poverty is widespread, employment options are limited and the State is absent, many young people turn to gangs in the barrio to acquire power, cash income, space and a feeling of belonging that no other social institution gives them (OECD/CAF/ECLAC, 2016[80]; Escotto, 2015[83]; Soto and Trucco, 2015[84]).
Violence tends to occur unevenly throughout the territories of Latin American countries, with high levels in deprived urban areas. Slums and shanty towns are both violent and poor, a scenario that reproduces and exacerbates social exclusion. Youth in these areas bear the burden of stigmatisation for a way of life seen as violent, and they are hence often denied dignity and solidarity. As a result, many of them are marginalised and fall victims to exploitation in adult-led criminal practices, in part because individuals under 18 years of age cannot be held criminally responsible (ECLAC, 2014[12]). Young people may also be victims or perpetrators of collective violence in school or community environments, directed either by youth groups towards specific individuals (young or not) or by neighbourhood groups or authorities towards young individuals or groups. Two cases of this type of violence have become significant in the youth setting: violent confrontation between groups of young people, which can have serious social impacts — in the case of gangs, for example; and school bullying perpetrated through social networks — including cyberbullying, to which girls are more likely to be subject (UNESCO, 2017[85]; OECD/CAF/ECLAC, 2016[80]).
Latin America is undergoing a deep demographic transformation. As life expectancy rises, the proportion of elderly people in the population increases, as does their age. Better understanding their needs and leveraging their active contribution to society become critical challenges (ECLAC, 2016[8]; Huenchan, 2013[86]). Figure 5.15 provides an overview of selected well-being outcomes for the older population (aged around 55 and over), compared with the middle-aged population (aged around 29-54). In terms of material conditions, older people are 35% less likely to live in extreme poverty than middle-aged adults and 40% less likely to live in absolute poverty. Their earnings are higher than the comparison group, whether they work in formal (+21%) or informal employment (+5%). However, there is no difference in satisfaction with living standards between the two groups, and older people are 15% less likely to work longer hours than the middle-aged comparison group, but they are much more likely to be in informal employment, as discussed later.
In terms of quality of life, however, most indicators show worse outcomes for the elderly across the focal countries, with the exception of satisfaction with services (health and education) and homicides. Elderly people are almost two-thirds more likely than the middle-aged comparison group to report physical limitations due to health reasons, less likely to voice their opinion to an official, more likely to experience more negative than positive emotions on a given day, less likely to feel safe walking in their area, less likely to have someone to count on in a time of need, more likely to commit suicide, and report marginally lower life satisfaction (with an average score on a 11-point scale of 5.8, compared with 6.1 for the middle-aged comparison group). These outcomes stand in contrast to the experience of OECD countries, where elderly people generally report better outcomes than their prime-age peers, in particular for life satisfaction, which is strongly associated with mental health problems, social ties and social network support (Gigantesco et al., 2019[87]; Costa and Ludermir, 2005[88]; Kawachi, 2001[89]).
Note: Each performance ratio is the simple average by age group calculated across the 11 focal countries for which data are available for all the inequality groups taken into consideration. As a result, performance ratios can cover a subset of the 11 focal countries. Performance ratios above 1 indicate better outcomes (i.e. higher well-being) for older people, whereas performance ratios below 1 indicate better outcomes (i.e. higher well-being) for middle-aged – including negative indicators, which have been reverse-scored. Grey bubbles denote no clear difference between older and middle-aged, defined as age ratios within 0.03 points distance to parity.
Source: OECD calculations. For a complete list of sources please refer to the "Sources and Methods" tab in the Statlink file
Despite this mixed picture, elderly people tend to have stronger confidence in the capacity of collective action to address their own needs as well as the broader societal problems faced by Latin America. The older age group is 25% more likely to trust in government, 15% more likely to trust in the police, 9% less likely to think that government is corrupt, 7% more likely to volunteer, 6% more likely to believe that tax avoidance is never justified and 4% less likely to believe that the income distribution is unfair. There is no clear difference in trust in others and support for democracy over other forms of government between the elderly and middle-aged populations.
Informal employment
As Figure 5.15 shows, elderly people are much more likely to be employed in informal work than their middle-aged peers. The share of older people in informal employment is particularly high in Peru, Paraguay and Colombia, where it reaches over 80% of total employment among persons aged 55 or above (Figure 5.16). Despite the progress with employment formalisation throughout Latin America over the past decade, a high proportion of older people still lack social security coverage (ECLAC, 2015[90]; ECLAC, 2015[91]), contributing to higher levels of vulnerability and inequality. For example, old age poverty in Colombia is high, as low-skilled workers spend much of their working lives in informal employment, without paying pension contributions (OECD, 2019[92]). In Brazil and Argentina, informal workers retire later than others for the same reason, until they reach the age to benefit from a non-contributory pension (OECD, 2019[93]; OECD, 2018[94]).
Source: ILO, https://www.ilo.org/shinyapps/bulkexplorer23/?lang=en&segment=indicator&id=EMP_NIFL_SEX_ECO_RT_A
The demographic transition in Latin America is likely to have an impact on pension systems – placing their sustainability at jeopardy. This is the case for both individual saving plans (due to the imbalance between the years of contributions and those over which benefits are drawn) and pay-as-you-go systems (due to a higher ratio of retirees to people of working age). Both developments may lead to measures to encourage working for longer (e.g. by raising the legal retirement age) (ECLAC-ILO, 2018[95]). In a context of fewer multigenerational households, many older people may therefore be left with little choice but to keep working until a later retirement age to meet their own needs. The older persons who reach this stage of life with the least protection are those who suffered deprivations in earlier stages (ECLAC, 2016[8]).
Pension coverage
The low pension coverage is a major policy challenge faced by most Latin American and Caribbean countries, both in terms of the proportion of workers participating in pension schemes and the proportion of the elderly receiving some kind of pension income. Efforts to close the coverage gap through non-contributory (or “social”) pensions are at the heart of the policy debate in the region. However, these policies may pose significant fiscal challenges (OECD/IDB/The World Bank, 2014[96]). A key determinant of pension coverage in the region is the type of employment people have. Frequent transitions between formality, informality and inactivity generate significant contribution gaps in workers’ careers, which put the adequacy of future retirement incomes at risk. In almost all systems, incomplete contribution histories result in lower pension entitlements, or even ineligibility (OECD/IDB/The World Bank, 2014[96]). As a result, a large share of older people in Latin America have to rely on sources of income other than contributory pensions, including income from informal work (Figure 5.16) and social pensions.
Figure 5.17 shows that huge progress has been made in pension coverage across the focal countries over the last two decades, with average coverage rates almost doubling from 35% in 2000 to 67% in 2020. Mexico shows a particularly impressive improvement, from only 10% coverage in 2000 to universal coverage in 2020. However, coverage rates vary substantially across focal countries - from only 11% in the Dominican Republic in 2020, to 100% in Mexico and Uruguay – and on average, almost one-third of the eligible population above statutory pensionable age do not receive a pension.
Women tend to have lower pension coverage than men, and the value of their pensions tends to be less, thus exacerbating the socio-economic disadvantage faced by older women and reflecting the discriminations women face on the labour market and other areas throughout their working life (ECLAC, 2018[97]). Across the focal countries in 2014-2015, only in Ecuador and Uruguay was pension coverage (marginally) higher for women than men, and the value of pension income was 20-42% lower for women than men in the majority of focal countries (only in Argentina, Brazil and Colombia was the gap less than 20%, and only in the Dominican Republic was there no substantial difference between men and women) (ECLAC, 2018[97]).
Note: This indicator shows the proportion of older persons receiving a pension, measured by the ratio of persons above statutory retirement age receiving an old-age pension to persons above statutory retirement age (including contributory and non-contributory). The earliest year available is 2003 for Chile and 2004 for the Dominican Republic. LAC is the regional average for Latin America and the Caribbean as calculated by UN DESA. OECD 35 excludes Ireland and Korea, due to incomplete time series.
Source: UN DESA Global SDG Indicator Database, indicator 1.3.1, https://unstats.un.org/sdgs/indicators/database/
The impact of COVID-19
Children
The COVID-19 pandemic may have devastating impacts on child well-being in the short, medium and long term, with repercussions at a physical, mental or socio-economic level, even though children have been relatively spared from the direct mortality impacts of the pandemic (UNICEF, 2021[98]; OECD, 2020[99]). Children have been less affected from an epidemiological perspective, although at the time of writing there is still uncertainty around precisely how the disease infects and spreads among children (Hobbs et al., 2020[100]). At the time that Latin America became the epicentre of COVID-19 cases in the second half of 2020 (PAHO, 2020[101]), millions of children in the region were living in poor households with no or little access to healthcare, whilst no longer receiving an education and being continuously exposed to violence and conflict (UNICEF, 2020[102]).
The especially intense strains on children’s lives during the extended periods of lockdown may follow them into the medium and long term. School closures may have severe effects especially on vulnerable families and children beyond the stress endured during lockdowns. During the first wave of the pandemic in Latin America, it is estimated that approximately 95% of children enrolled in education were out of school (UNICEF, 2020[103]). First, the success of provisional educational measures implemented during school closures, for example remote learning, largely depends on the quality of home learning environments (OECD, 2020[104]). In Latin America, this meant that the consequences for child learning were particularly serious, with certain students set to never return to school (UNICEF, 2020[102]). Second, closures entailed the interruption of various parallel services, including school meals, infirmaries, drinking water and even the psychosocial support external to their household. Since the beginning of the pandemic, 80 million Latin American children have been denied hot meals in the region (WFP, 2020[105]). Children with disabilities have also been disproportionately affected (ECLAC, 2020[106]), as they are even more likely to miss out on their special education needs whilst compromising parents’ abilities to meet new demands of home schooling for other children. Thus, interrupted educational services have severe consequences on children’s current well-being and may leave scores of children ill-equipped for their pursuit of a brighter future (OECD, 2020[104]).
The measures imposed by lockdowns throughout 2020 resulted in increased household tensions, economic uncertainty, social isolation as well as added stress on caregivers (UNICEF, 2020[107]; OECD, 2020[99]). 21% of adolescents aged 13-17 in Latin America reported more arguments with their parents and other household members during quarantine, increasing the risk of domestic violence (UNICEF, 2020[108]). Child protection services were already relatively weak in Latin America following a decade of gradual deterioration (ECLAC, 2020[109]). Recent research estimates that 55% of children in the region experience physical aggression and 48% suffer psychological aggression (Cuartas et al., 2019[110]). Potential impacts on victims include lifelong impairments in emotional and cognitive capacities, along with antisocial and/or high-risk behaviour (Cuartas et al., 2019[110]). The COVID-19 crisis may also lead to the first increase in child labour in the region after almost 20 years of progress (ECLAC, 2020[109]). As seen in Chapter 2, one of the main impacts of COVID-19 has been a rise in poverty levels, which is pushing vulnerable families to use every resource available to them in order to increase household income and ensure survival, including sending children to work.
Young adults
COVID-19 exposes youth in the region to higher risks of disengagement and dropout from education and training and may increase the overall number not in education, employment or training (NEET). Although the reasons for disengagement and dropout are complex and change over time (Aarkrog et al., 2018[111]), COVID-19 may act as a potent multiplier through various vectors. These include breaks in education and training that lead to declines in performance and loss of motivation, the loss of connections with supportive adults and positive peer interactions, and increases in household poverty and higher household stress (OECD, 2020[104]). In addition, the practical or workplace learning components of vocational education and training are less well-suited to remote learning. Many youths are likely to have been the first to lose their jobs in 2020 – particularly those in the informal economy and in sectors such as tourism, non-electronic commerce, transport and other services in which teleworking is not an option (ILO, 2020[112]). These prolonged periods of inactivity or unemployment may lead to further discouragement and exclusion.
The unprecedented impact of COVID-19 in the region has the potential for long-term effects on youth unemployment. As seen in Figure 5.9, employment opportunities for young people in the focal group were already poor before the crisis hit, especially for young women (whose unemployment rate in 2020, at 22%, is almost 7 percentage points higher than that of men), and whose share not in employment, education or training, at 29%, is twice as high). Together, these elements paint a negative picture for youth well-being in Latin America, which is characterised by a dangerous pattern of self-reinforcing aspiration gaps.
Elderly
The COVID-19 outbreak poses significant challenges for older people. First, older people (and older men in particular) have higher risks for developing serious complications in case of infection. Second, the development of illness in old age has a larger potential to significantly deteriorate older people’s general health status. Third, stronger confinement measures tend to affect older people disproportionately, significantly changing their day-to-day lives and restricting their independence. These challenges will be heightened for those who are in poor health, living alone or in long-term care, and for those caring for a family member (OECD, 2020[104]).
COVID-19 will also have considerable impacts on older people’s social connections. Limiting their exposure to COVID-19 requires older people to self-isolate and rely on support networks and local care services for necessities, such as grocery shopping and cooked meals. In times of need, older people are more likely than middle-aged people to report not having a family member or friend that they can rely on. Furthermore, many older people live alone. For instance, in Argentina and Uruguay, almost one-third of the population aged 80 or above live on their own (34% and 32%, respectively) (IDB, 2017[113]). Moreover, changes in family structures and women’s increasing participation in the labour market during the past decades in Latin America have lowered families’ capacity to care for people with dependencies.
Moreover, COVID-19 is disrupting routine health care for the many older people with chronic health conditions – although in many countries taking care of elderly and sick relatives was still allowed under confinement. COVID-19 poses particular risks for elderly residents in long-term care facilities, in terms of increased mortality and low subjective well-being (OECD, 2020[104]). A sizeable proportion of older people across Latin America are care-dependent (12% of people over 60, 27% of those over the age of 80), and by 2050 more than 27 million people over age 60 may need long-term care (Cafagna et al., 2019[114]). Moreover, the communal living environment of long‐term care facilities and the vulnerability of residents are conducive to the rapid spread of influenza virus and other respiratory pathogens (OECD, 2019[115]; Lansbury, Brown and Nguyen-Van-Tam, 2017[116]). To protect residents, some long-term care facilities were shut off from visitors. The absence of contact with family members has, however, negative effects on psychological well-being, especially in the case of a prolonged outbreak (OECD, 2020[104]).
Issues for statistical development
Children
Measuring child well-being is a challenge, as children are not generally the main target of common data collection instruments such as household surveys, unless specifically designed for them. Age-disaggregated data covering the child population aged under 15 is therefore scarce, and there is little information on child-specific issues, such as access to initial and early childhood education programmes, learning outcomes and cognitive skills, social and emotional well-being, malnutrition and other aspects of health status, and violence against children (both in the household and in schools). Measuring the well-being of children carries additional difficulties and considerations compared to other population groups, such as taking into account the strong consequences of child development on later life outcomes and the close connection between children’s well-being and the opportunities and resources found within their families, schools and communities. This is a concern in the context of the UN 2030 Agenda, as in order to achieve the SDG targets related to children (e.g. the eradication of child poverty in target 1.2, or ending violence against children in target 16.2), countries must have accurate, timely and disaggregated data.
In addition, even where children are covered in household surveys, these can fail to measure their situation in the most marginalised positions, such as children with disabilities, children experiencing maltreatment and children living outside the home. Surveys are therefore not fully representative of all children, and more specific surveys could help provide a clearer picture. Countries of the focal group have made progress in this regard, and for instance Chile, Costa Rica, Mexico and Peru have all developed specific survey tools for measuring child disability (INEC, 2018[117]; INSP, 2016[118]; SENADIS, 2015[119]; INEI, 2014[120]). Administrative data can provide important information on the situation of institutionalised children and the provision of child protection services. Finally, experts increasingly see value in listening to children’s thoughts and views on aspects of their own lives. While there are challenges to collecting self-reported data for children, especially at a young age, techniques have been established to do this,22 and just as with adults, subjective measures can act as a valuable complement to (rather than replacement for) other measures of child well-being (OECD, 2021[65]).
One important measurement initiative is the Multiple Indicator Cluster Survey (MICS) programme, instigated by UNICEF, which aims to support governments in carrying out surveys focused on children through technical assistance, material support and standardised methodologies. To date, 34 MICS surveys have been completed across 18 countries in the region (UNICEF, 2021[121]). Examples of topics covered in the surveys include access to education; experiences of child labour; child discipline; access to water, sanitation and handwashing facilities; and exposure to insecticide.
Young adults
Over one-third of the SDG targets refer to young people either implicitly or explicitly, with major focuses on empowerment, participation and well-being. Youth-specific targets (under goals on hunger, education, gender equality, decent work, inequality and climate change) call for better information on inequalities in an intergenerational setting. The current, limited scope of analysis highlights the importance of further developing longitudinal studies, for instance, which include those that follow people from birth. An important (and much less expensive) option is to include retrospective questions on parents’ conditions (and on the well-being outcomes of respondents at previous stages of their life) in cross-sectional surveys: while cognitively demanding and liable to memory biases, these questions have the potential to significantly enhance research and policy design (OECD, 2017[6]).
There are specific measurement gaps in this study relating to young people’s health. For example, relatively few epidemiological studies of mental health among young people exist in the region – and those that do exist are difficult to compare due to differences in measurement instruments, the range of subject ages as well the periods covered (ECLAC, 2014[12]).
This lack of comparable data is also problematic for addressing challenges that have not been mentioned in this section, such as alcohol and substance abuse. National youth surveys may include the issue in detail, yet methodological differences impede comparability. In this regard, international studies such as the Global School-based Student Health Survey (GSHS) developed by the WHO are of particular relevance for shedding light on regional trends, but fail to capture adolescents who do not attend school - and for whom substance abuse is often prevalent.
Finally, in relation to gender and sexual identity, the limited available data on LGBT (Lesbian, Gay, Bisexual, Transgender) adolescents and youth stands in stark contrast to their disproportionate vulnerability and exposure to risks (CDC, 2020[122]; Coker, Austin and Schuster, 2010[123]). According to data from the 2015 national Youth Risk Behavior Survey (YRBS), lesbian, gay and bisexual (LGB) students in the United States were 140% more likely (12% vs. 5%) to not go to school at least one day during the 30 days prior to the survey because of safety concerns, as compared with heterosexual students (Kann et al., 2016[124]). LGBT youth were also at greater risk for depression, suicide, substance use and risky sexual behaviours. Nearly one-third (29%) of LGB youth had attempted suicide at least once in the prior year, as compared to 6% of heterosexual youth (Kann et al., 2016[124]).
Elderly
As the LAC region faces a demographic transition characterised by an ageing population, it will become increasingly necessary to better monitor and understand issues of specific importance for the well-being of older people. This has long been recognised in the region, and as far back as 2006, following the establishment of the 2002 Madrid International Plan of Action on Ageing, ECLAC produced a Manual for Indicators of Quality of Life in Old Age (ECLAC, 2006[125]). This manual covered measurement of a number of topics included in the well-being framework, including economic security (labour force participation, social protection, poverty), health and well-being (health status, lifestyle risks) and the social environment (social network support, social participation, violence and maltreatment of the elderly). However, at the current time, a number of data gaps exist for compiling regular and harmonised statistics on the well-being of older people.
In terms of the health of the elderly, there is relatively little information available regarding chronic conditions, functional capacity, self-perceived health status, depression, lifestyle habits, out-of-pocket expenses, surgeries and the use of medication or assistive devices (NASEM, 2015[126]). Although the proportion of people with disabilities tends to increase with age, few statistical offices compile comparative statistics in this field (ECLAC-ILO, 2018[95]). There are no comparable data on the share of people in long-term care.
Time-use surveys could be a useful tool to improve the evaluation of the care services that older people in Latin America receive and request. Other, more specific surveys on the population aged 60 or above should be a priority for countries of the region in order to keep track of the rapidly ageing population and better understand causalities in different areas at the end of the life cycle.
The Latin American and Caribbean region is characterised by a high spatial concentration of population and economic activity, with 80% of the population living in urban areas (55% in cities and 25% in towns) (UNDP, 2020[127]; OECD/European Commission, 2020[128]), the highest share among world regions and much higher than the world average of 56%. Large inequalities in living conditions also mark different locations within a country (ECLAC, 2020[129]). Figure 5.18 shows performance ratios for selected well-being outcomes and resources for future well-being for people living in rural areas in comparison to those living in urban areas, on average across the 11 focal LAC countries. To ease understanding, all indicators are coded in the same direction so that higher ratios always correspond to better performance for people living in rural areas.
While satisfaction with living standards and employment do not much differ, on average, across rural and urban areas in the focal countries (Figure 5.18, Panel A), informal employment is around one-third higher in rural areas than in urban areas, while rural monthly earnings in the formal sector are around one-third lower than in urban areas. People living in rural areas are two-thirds more likely to live in poverty than those in urban areas (with rural poverty rates of 29% compared with urban poverty rates of 17.4%), and over three times more likely to live in extreme poverty (with respective rural and urban extreme poverty rates of 11.2% and 3.6%). People in rural areas are also more likely to live in poor housing conditions: they are three times more likely to live in dwellings built with low-quality materials and over one-third more likely to live in overcrowded dwellings, compared to their urban counterparts.23 Availability of infrastructure is also more limited in rural areas; just below 70% of the population have access to water and sanitation facilities, compared with almost universal coverage in urban areas, and less than one-third of households in rural areas have access to the Internet, as compared to more than half in urban areas (56%).
On the other hand, people in rural areas are less likely to be unemployed (5% of the rural population was unemployed, compared to 8% of urban dwellers in 2019), but informal earnings are almost one-third lower than in urban areas. Income inequality is also lower in rural areas, when considering both the Gini coefficient and the gap between the income share of the top and bottom 20% of the population.
When looking at quality of life (Figure 5.18, Panel B), people living in rural areas feel safer and civic engagement is higher. They are 55% more likely to report feeling safe when walking alone at night in the area where they live and 8% more likely to voice their opinion to an official than their respective urban counterparts in the focal countries. People living in rural areas are also 13% more likely to be satisfied with the education system, possibly reflecting their lower educational attainment, less awareness about the limitations of the education system, and lower standards when evaluating it (Cárdenas et al., 2008[130]). On the other hand, people living in rural areas are more likely to report health problems that prevent them from doing the things that people of their age normally do, reflecting higher poverty and informality and limited availability and access to healthcare, which can discourage people to seek treatment. People in rural areas are also slightly less likely to report satisfaction with the availability of quality healthcare. Additionally, people in rural areas are slightly more likely to report more negative than positive emotions on a given day (negative affect balance) and to report slightly lower life satisfaction than people in urban areas.
Social capital is generally higher in rural areas (Figure 5.18, Panel C): across the focus countries, people in rural areas are almost 20% more likely to have volunteered their time than people in urban areas, 20% more likely to trust the national government, 4% less likely to believe that government is corrupt and 18% more likely to trust the police. On the other hand, human capital is lower in rural areas than in urban areas (Figure 5.18, Panel C): the share of young adults (aged 20-24) with upper secondary educational attainment is 25% lower in rural areas, while the proportion of youth (aged 15-24) not in education, employment or training, and not working exclusively in the home (NEET), is 16% higher.
Note: Each performance ratio is the simple average by territorial area calculated across the 11 focal countries for which data are available for all the inequality groups taken into consideration. As a result, performance ratios can cover a subset of the 11 focal countries. Performance ratios above 1 indicate better outcomes (i.e. higher well-being) for rural, whereas performance ratios below 1 indicate better outcomes (i.e. higher well-being) for urban – including negative indicators, which have been reverse-scored. Grey bubbles denote no clear difference between urban and rural, defined as territorial ratios within 0.03 points distance to parity.
Source: OECD calculations. For a complete list of sources please refer to the "Sources and Methods" tab in the Statlink file
Income inequality and poverty
Absolute and extreme poverty are generally higher in rural areas (Figure 5.19). The shares of people living in households with income insufficient to buy a basic food basket (ECLAC’s definition of extreme poverty) as well as other necessary goods and services (ECLAC’s definition of absolute poverty) are, respectively, 8 and 11 percentage points higher in rural areas in the focal countries, on average. On these definitions, extreme and absolute poverty are highest in Colombia and Mexico (above 20% for extreme poverty and above 45% for absolute poverty). Rural/urban gaps are largest in Paraguay and Peru (where the shares of people living in extreme and absolute poverty in rural areas are more than four times and more than two times larger than in urban areas, respectively) and lowest in Chile (with gaps limited to 0.2 and 0.4 percentage points) and Uruguay (where more people live in poverty in urban areas than in rural areas).
Note: Regional extreme poverty is defined by ECLAC as the share of people living in households with income insufficient to buy a basic food basket. ECLAC absolute poverty refers to households with an income insufficient to buy a basic food basket and other non-food necessities. The latest available year is 2018 for Mexico and 2017 for Chile. LAC 10 excludes Argentina, as data for rural areas are not available. LAC is the regional average for Latin America and the Caribbean calculated by ECLAC.
Source: ECLAC Statistics, CEPALSTAT database, https://cepalstat-prod.cepal.org/cepalstat/tabulador/ ConsultaIntegrada.asp?idIndicador=3328&idioma=i
When looking at the distribution of income across the population (Figure 5.20), income inequality is higher in urban areas than in rural ones, except in Paraguay (where it is higher in rural areas) and in Peru (where there is almost no difference between the two). The two income inequality measures presented (the Gini coefficient, which focuses on the middle of the income distribution, in Panel A; and the S80/S20 income ratio, which informs about the gap between the income of the top 20% and bottom 20%, presented in Panel B) convey, with few exceptions, a consistent picture.
Note: The latest available year is 2018 for Mexico and 2017 for Chile. LAC 10 excludes Argentina, as data for rural areas are not available. LAC is the regional average for Latin America and the Caribbean calculated by ECLAC.
Source: ECLAC Statistics, CEPALSTAT database, https://cepalstat-prod.cepal.org/cepalstat/tabulador/ ConsultaIntegrada.asp?IdAplicacion=1&idTema=935&idIndicador=3289&idioma=i (Panel A) and https://cepalstat-prod.cepal.org/cepalstat/tabulador/ConsultaIntegrada.asp?idIndicador=3328&idioma=i (Panel B)
Housing infrastructure
Infrastructure coverage is more limited in rural areas, where just below 70% of the rural population have access to water and sanitation, while the coverage is almost complete in urban areas. Coverage in rural areas is the lowest (below 40%, Figure 5.21) in Brazil (for both water and sanitation), Mexico (for sanitation only) and Uruguay (for water only) and highest (almost 90% and above) in Costa Rica (for both water and sanitation), Paraguay (for water only) and Uruguay (for sanitation only).
Note: The latest available year is 2017 for Chile (Panel A and B), Paraguay (Panel A) and 2015 for Brazil (Panel A and B). LAC 10 excludes Argentina, as data for rural areas are not available. LAC regional average comprises 16 Latin American and Caribbean countries, including the focal countries.
Source: Socio-Economic Database for Latin America and the Caribbean (CEDLAS and The World Bank), https://www.cedlas.econo.unlp.edu.ar/wp/en/estadisticas/sedlac/estadisticas/ (accessed in June 2021)
Differences in access to the Internet in the focal countries are also wide: only 27% of households in rural areas have Internet access, but close to half of those in urban areas (Figure 5.22). The Internet access of rural households ranges from less than 10% in Paraguay and Peru to about half in Chile, Costa Rica and Uruguay, countries where rural access is also greater.
Note: The latest available year is 2019 for Argentina and Brazil, 2017 for Chile and Ecuador and 2018 for all the other countries. LAC 10 excludes Argentina, as data for rural areas are not available. LAC regional average comprises 15 Latin American and Caribbean countries, including the 10 focal countries with available data.
Source: ECLAC Statistics, ECLAC Household Survey Data Bank (Banco de Datos de Encuestas de Hogares (BADEHOG)) for Argentina, Brazil, Chile, Colombia, Costa Rica, Ecuador, Paraguay, Peru, and Uruguay; ITU World Telecommunication / ICT Indicators Database 2020, https://www.itu.int/en/ITU-D/Statistics/Pages/publications/wtid.aspx, for all the other countries
The impact of COVID-19 on urban-rural inequalities
The marked spatial concentration and high population density in Latin America, together with large territorial inequalities, are high-risk factors that accelerate the spread of COVID-19, particularly in population segments that experience significant vulnerabilities and material deprivations (ECLAC, 2020[129]). The people at greatest epidemiological risk, as well as the most vulnerable to the pandemic’s socio-economic impacts, are those living in overcrowded dwellings, with limited access to water or sanitation, in particular those living in slums or informal settlements in urban areas who also frequently have pre-existing health conditions. These are largely informal workers, with limited or no assets, nor social security and often no Internet access. Among the urban poor, family dysfunctions are common, which, under lockdown measures, can lead to domestic violence and child abuse. Many of these conditions apply also to poor people living in rural areas (Lustig and Tommasi, 2020[131]). In these conditions, staying at home is unhealthy, unsafe and very hard for people who cannot work from home and need to go out to earn a living. The economic and social impacts will be the highest in disadvantaged neighbourhoods in large urban areas and will exacerbate pre-existing problems (ECLAC, 2020[129]).
Access to water and handwashing facilities, and to sanitation more generally, are essential to contain the spread of COVID-19. Access to the Internet and to digital services has become necessary to continue regular activities (education and work, when possible), to gain access to health care and, more generally, for living (to keep social connections, for leisure, etc.). Information technologies will therefore be crucial to limit the consequences of future crises of this type.24
Issues for statistical development
Harmonised well-being data are not always available by urban and rural areas, and are very limited in some well-being domains (e.g. health status, knowledge and skills, civic engagement and empowerment, and human capital). When looking at the individual indicators used in this section, the scope for improvement is broad. For example, the indicator measuring overcrowding, defined as the share of households with more than two people per bedroom, could be better refined: bedrooms can have different surfaces and could be bigger in rural areas. Moreover, the indicator does not account for urban slums or informal settlements. A more precise measure would consider the square metres available per person in the dwelling. However, this information is not widely available for Latin American countries, nor for OECD countries more widely.
As this section shows, geography matters for well-being, and the binomial classification urban/rural can hide a more nuanced reality: urban areas differ from each other, as cities have different sizes, while rural areas can have different characteristics and geographies (from well-serviced communities near urban areas to remote and sparsely-populated places with limited access to basic services). The collection of harmonised indicators for cities, urban and rural areas requires harmonised definitions for the delineation of these areas. National definitions vary considerably across countries and thus limit international comparability. A new method, called the Degree of Urbanisation, has been endorsed by the 51st session of the United Nations’ Statistical Commission as the recommended method for international comparisons. The Degree of Urbanisation classifies the entire territory of a country into three classes: 1) cities, 2) towns and semi-dense areas, and 3) rural areas. The Degree of Urbanisation has two extensions. The first extensions identifies cities, towns, suburban or peri-urban areas, villages, dispersed rural areas and mostly uninhabited areas. The second extension adds a commuting zone around each city to create a functional urban area (FUA) or metropolitan area (European Commission et al., 2020[132]).
Another important spatial level to understand inequalities in Latin America and the Caribbean is the region. Regions are of different forms and size depending on the country (e.g. northeast Brazil, southwest Mexico and Norte Grande in Argentina) and have specific sociocultural identities and shared problems.25 Comparable well-being data at regional level are very limited for Latin America (Box 5.3). A regional scale would allow for a more holistic approach to the various socio-spatial and geographical aspects of development and the interactions between them, such as urban and peri-urban dynamics, rural development, river basins, natural resource management and governance, clean energy conversion and connectivity infrastructure. At the regional level, the realities of the different areas and the differences between them can be better identified, investments can be better focused and human settlements can be better recognised and sustainably managed as part of ecosystems (ECLAC, 2020[129]). Measuring well-being outcomes at the level of different regions would, however, require larger sample sizes than those currently available in the LAC region, or the mobilisation of administrative records. The OECD has also been conducting work to develop typologies to classify regions, for example based on a region’s accessibility to Metropolitan areas (Fadic et al., 2019[133]).
Finally, while (as this chapter has shown) rural areas tend to be more deprived in terms of access to basic services (such as water, sanitation, electricity), these indicators are capturing very extreme manifestations of deprivation, which may not be the most meaningful measures for relatively more developed countries and for urban areas. Different thresholds may be required to measure relative deprivation in urban areas (such as, for example, the number of hours a day the service in question is available, or the quality of water) (Santos, 2019[134]). Better and more comparable information about access to waste retrieval services and access to public transport services along with their frequency would also be highly relevant.
The OECD Regional Well-being web tool allows to measure well-being at regional level and compare 403 OECD regions based on eleven well-being dimensions (income, jobs, housing, access to services, health, education, community, civic engagement, environment, safety and life satisfaction). Regions are classified on two territorial levels, reflecting the administrative organisation of countries: large regions (TL2) and small regions (TL3). Small regions are classified according to their access to metropolitan areas (Fadic et al., 2019[133])Data are available for three OECD Latin American countries: Chile, Colombia and Mexico.
According to the report OECD Regions and Cities at a Glance 2020 (OECD, 2020[135]), which describes how OECD regions and cities are progressing in their efforts to build stronger, more sustainable and more resilient economies and societies, Chile, Colombia and Mexico face stark regional disparities. In each of the three countries, there are two regions that are in the top 20% of the OECD regions in some well-being dimensions: in Chile, Ñuble performs in the top for environment and health, and Aysén leads in life satisfaction and jobs; in Colombia, Vichada performs in the top for environment and Arauca in community; and in Mexico, Baia California Sur is among the top 20% of OECD regions in community and Tamaulipas in life satisfaction. In contrast, most Chilean regions are in the bottom 20% of OECD regions in income, housing and civic engagement; all Colombian regions are in the bottom OECD 25% of regions in terms of safety and all Mexican states are in the bottom 20% of OECD regions in the dimensions of income and health.
Additionally to the OECD Regional Well-being web tool (https://www.oecdregionalwellbeing.org/), the OECD web tool Measuring the distance to the SDGs in regions and cities measures the distance towards the SDGs in more than 600 regions and 600 cities of OECD and partner countries. For Latin America, information is available for Argentina, Brazil, Chile, Colombia, Costa Rica and Mexico (https://www.oecd-local-sdgs.org/).
1. This work was produced before Costa Rica became an OECD member.
In Latin America, the concept of ethnicity is most commonly used with reference to Indigenous people and the concept of race primarily for Afro-descendants (ECLAC, 2016[8]). Across the LAC region as a whole, around 10% of the population self-identify as Indigenous and 21% as Afro-descendant (Figure 5.23). In the 11 focal countries, the proportions are a little lower but still substantial, with around 8% identifying as Indigenous and a similar proportion identifying as Afro-descendant. The size of these groups varies substantially across countries (Figure 5.23): 26% of the population in Peru and 21.5% of the population in Mexico self-identify as Indigenous, compared with 0.5% in Brazil; on the other hand, over half of the Brazilian population (50.9%) identify as Afro-descendant, compared with less than 0.5% in Argentina, Chile and Paraguay. Aside from these differences in size, these groups also display significant social and linguistic diversity within and across countries. It is estimated that there are 800 different Indigenous Peoples across the LAC region (ECLAC et al., 2020[136]), and while the Afro-descendent population in the region has a common history rooted in slavery, today it is highly varied culturally, socio-economically and racially, both within and across countries (World Bank Group, 2018[137]).
Note: Data refer to 2018 for Colombia, 2017 for Chile and Peru, 2015 for Mexico, 2014 for Dominican Republic, 2012 for Paraguay, 2011 for Costa Rica, and Uruguay, and 2010 for Argentina, Brazil and Ecuador. The LAC regional average includes Bolivia, Guatemala, Honduras, El Salvador, Nicaragua, Panama and Venezuela in addition to the focal countries and covers data collection years ranging from 2007 to 2018.
Source: “The Indigenous people of Latin America - Abya Yala and the SDG Agenda: Tensions and challenges from a territorial perspective” (ECLAC and FILAC, 2020[138]) for Panel A, and “Afro-descendants and the social inequality matrix: Inclusion challenges” (ECLAC, 2020[139]) for Panel B
However, both Indigenous and Afro-descendant populations in the region face shared challenges in terms of exclusion, deprivation and discrimination. Figure 5.10 shows that across almost all the available indicators for material conditions, quality of life, and social and human capital, Indigenous people tend to have lower well-being outcomes than non-Indigenous people, and Afro-descendant people tend to have lower well-being outcomes than non-Afro-descendant people.26
Note: Each performance ratio is the simple average, calculated across the focal countries for which data are available (see the “Sources and Methods” tab in the Statlink file for details on the countries covered per indicator). All indicators have been coded so that performance ratios above 1 indicate better outcomes (i.e. higher well-being) for Indigenous and Afro-descendant people, whereas performance ratios below 1 indicate better outcomes (i.e. higher well-being) for Non-Indigenous and Non-Afro-descendant people. Grey (for Indigenous) and black (for Afro-descendant) bubbles denote no clear difference with the reference group, defined as performance ratios within 0.03 points distance to parity. The default reference group for the comparisons is the population who are non-Indigenous and Non-Afro-descendant, however there are some exceptions. For the Indigenous performance ratios for employment, unemployment, hourly earnings, overcrowding, illiteracy, mean years of schooling, youth secondary attainment, youth tertiary attainment, and infant mortality, the comparison group is non-Indigenous only (i.e. the comparison group includes Afro-descendant people). For the Afro-descendant performance ratios for infant mortality and maternal mortality, the comparison group is Non-Afro-descendant only (i.e. the comparison group includes Indigenous people).
Source: OECD calculations. For a complete list of sources please refer to the "Sources and Methods" tab in the Statlink file
When looking at the available indicators of material conditions (Figure 5.24, Panel A), Indigenous people are twice as likely to live in absolute poverty and over three times as likely to live in extreme poverty as non-Indigenous people. They also earn lower hourly earnings. Afro-descendent people are twice as likely to live in poverty and over twice as likely to live in extreme poverty as non-Afro-descendant people. Both Indigenous and Afro-descendant people are less likely to think that their income is sufficient to meet their needs or to have a greater fear of losing their job than their respective comparison groups. However, the picture is more mixed when looking at employment and unemployment. There is no substantial difference in employment rates for Indigenous and Afro-descendant people with reference to the comparison group; and while Afro-descendant people are more likely to be unemployed than non-Afro-descendant (with unemployment rates of 9.8% and 7.1% respectively), Indigenous people are 13% less likely to be unemployed than the comparison group. These “positive” labour market outcomes for Indigenous employment and unemployment and Afro-descendant employment need to be interpreted with caution, as they mask the fact that the type of jobs available to workers in both groups tend to be of low quality. Globally, Indigenous people are more likely to work in informal jobs than non-Indigenous, and the gap is even higher in Latin America, where on average, the informality rate is 87% for Indigenous workers compared with 51% for non-Indigenous (ECLAC and FILAC, 2020[138]). Afro-descendant workers are more likely to work in the informal sector than non-Afro-descendant workers in most focal countries with available data (World Bank Group, 2018[137]), although the gaps are smaller than for Indigenous workers.27 Informal jobs entail higher vulnerability, such as employment in intensive agriculture, which has led to an increase of rural Indigenous workers migrating away from their communities to work under precarious conditions in degraded living situations (ECLAC and FILAC, 2020[138]). One of the principal characteristics of informal work is a lack of social protection, including pension coverage, an issue that is explored in more detail later in the section. Finally, Indigenous people are over twice as likely to live in overcrowded conditions.28
For the available indicators of quality of life (Figure 5.24, Panel B), Indigenous and Afro-descendants tend to perform worse in areas related to health and education. Infant mortality is higher for both Afro-descendant and Indigenous infants than the comparison group, and maternal mortality is over 2.5 times higher for the Afro-descendant population than for non-Afro-descendants. Young people in both groups are less likely to complete secondary education and less likely to access tertiary education than the comparison group. In addition, illiteracy is almost three times higher for Indigenous people than for non-Indigenous, and mean years of schooling are also lower.
Across other selected indicators of quality of life, however, the differences are smaller or more ambiguous. Both groups report slightly lower levels of life satisfaction, and slightly higher rates of reported victimisation. However, both groups show slightly lower fear of crime, slightly lower perceptions of elite State capture (the belief that their country is governed by the powerful for their own interests) than their comparison groups. While Afro-descendants were slightly less likely to have voted in the last election compared with non-Afro-Descendants, Indigenous people were marginally more likely to have voted than non-Indigenous. These results are sometimes counter-intuitive and underline the need for better data and more research on these issues. For example, the slightly lower fear of crime goes against what is known about the increased exposure to State and paramilitary violence experienced by Indigenous peoples, indicating that it may not be the best measure to capture the types of risks these groups face. The differences are also very small and may not be statistically significant (in the summary charts, any difference of 3% or below is presented as showing no clear difference).
Finally, regarding the available indicators of social and human capital, the biggest gap is observed in perceived discrimination, with Afro-descendant and Indigenous people significantly more likely than non-Afro-descendant and non-Indigenous people to believe they belong to a discriminated group. Both groups are less likely to trust the police, to support democracy over other forms of governance, and to believe that tax avoidance is always unjustifiable (i.e. lower tax morale). When considering trust in government, trust in others and perceived inequality (i.e. the share of people believing that the income distribution is unfair), there is very little difference between the Indigenous and Afro-descendant groups and their comparison groups, while Afro-descendants aged 15-29 are more likely to be neither in employment nor in education (NEET) than non-Afro-descendants (with NEET rates of 26% for Afro-descendants and 21% for non-Afro-descendants).
Housing conditions and basic services
Inadequate housing and insufficient access to basic services heighten the vulnerability of those affected and are more likely to impact those who experience other forms of material deprivation as well, such as income poverty. Across a range of indicators related to housing conditions, Indigenous and Afro-descendant people experience worse outcomes than their comparison group (Figure 5.25). Afro-descendant people are less likely to have access to water, toilets, the Internet and sewerage than non-Afro-descendants. Gaps in housing and service outcomes are even larger for the Indigenous population, and they are less likely than non-Indigenous to have access to sanitation services, twice as likely to be living in overcrowded dwellings, and around three times less likely to have access to electricity.
Note: Performance ratios have been calculated by dividing the average results for Afro-descendant (Panel A) and Indigenous (Panel B) people across the focal countries for which data are available by the average results for Non-Afro-descendant (Panel A) or Non-Indigenous (Panel B). Performance ratios above 1 indicate better outcomes (i.e. higher well-being) for Afro-descendant (Panel A) or Indigenous (Panel B), whereas performance ratios below 1 indicate better outcomes (i.e. higher well-being) for Non-Afro-descendant (Panel A) or Non-Indigenous (Panel B) – including negative indicators, which have been reverse-scored.
Source: OECD calculations. For a complete list of sources please refer to the "Sources and Methods" tab in the Statlink file
Pension coverage
Social protection, as discussed in Chapter 2, can take a variety of forms, encompassing basic welfare guarantees, insurance against risks arising from the context or the life cycle, and the moderation or repair of social harm that occurs when social risks materialise (Cecchini et al., 2015[140]). It provides an essential safety net in times of increased vulnerability, such as unemployment or old age, although many types of protection are linked to formal employment. Informal workers are therefore less likely to access social benefits for health care, old-age pensions, insurance against unemployment, injury or maternity.
While comparable social protection data are not widely available for Indigenous or Afro-descendant populations, data on pension coverage give an indication of gaps in social protection by ethnicity and race. Figure 5.26 shows that across the countries and age groups with available data, Indigenous and Afro-descendant people have lower pension coverage than others. Around four out of five Indigenous workers are not affiliated to a pension system in Mexico (80%), Ecuador (79%) and Peru (78%), representing a gap of between 25 percentage points (in Ecuador) and 14 (in Mexico) with respect to non-Indigenous workers (Figure 5.26, Panel A). In the four countries with available data (Figure 5.26, Panel B), the Afro-descendant working-age population is consistently less likely to be affiliated to a pension system than the non-Afro-descendant comparison group.
Note: For Panel A, data refer to 2016 for Ecuador, Mexico, Peru and Uruguay, and to 2015 for Brazil and Chile. For Panel B, the data refer to 2018 for all countries.
Source: “The Indigenous people of Latin America - Abya Yala and the SDG Agenda: Tensions and challenges from a territorial perspective” (ECLAC and FILAC, 2020[138]) for Panel A, and “Afro-descendants and the social inequality matrix: Inclusion challenges” (ECLAC, 2020[139]) for Panel B
Discrimination
Discrimination and racism are both cause and effect of the existing inequalities in well-being outcomes by ethnicity and race in Latin America. They have been constant presences in the region for centuries, having their roots in the process of colonisation and slavery. Starting at the beginning of the 20th century, the concept of mestizaje – the notion that most people were of mixed race and discrimination was non-existent – gained widespread acceptance in the region (Sánchez-Ancochea, 2021[1]). However, the existence of ethnic and racial discrimination has become increasingly recognised by governments in recent decades, leading to improved data by ethnicity and race. On average across the focal countries, 29% of Indigenous people and 25% of Afro-descendant people say they belong to a discriminated group, compared with 17% of people who are neither Indigenous nor Afro-descendant (Figure 5.27). Large differences exist across countries, with over half (52%) of Indigenous people in Peru and almost four-fifths (39%) of Afro-descendant people in Brazil reporting they belong to a discriminated group. Experimental surveys in four Latin American countries (Brazil, Colombia, Mexico and Peru) using a spectrum of skin colour (from darkest to lightest) as an identifying category show that inequalities in social and economic status, and in experience of discrimination, are as much a function of skin colour as of ethno-racial grouping (Telles, 2014[141]).
Note: Data shows pooled results for 2010, 2011, and 2015 to ensure adequate sample size. Indigenous refers to people who self-identified as Indigenous and Afro-descendant refer to those who self-identified as Black or Mulato in the Latinobarómetro survey.
Source: OECD calculations based on Latinobarometro (database), https://www.latinobarometro.org/latOnline.jsp
The impact of COVID-19 on ethnic and racial inequalities
The deprivation of both Indigenous and Afro-descendant populations implied high vulnerability to the consequences of the pandemic. The common challenges faced by the two groups in terms of poverty, informality, lack of social protection, inadequate housing and other areas increase the risks that they have experienced during the pandemic, in terms both of the direct health impact as well as of broader socio-economic outcomes (ECLAC, 2021[142]; ECLAC et al., 2020[136]). These disadvantages are reinforced by spatial inequalities, and Indigenous and Afro-descendant people will face different risks depending on whether they live in urban or rural areas and depending on the specific territories in which they are concentrated due to historical patterns of settlement.
The Indigenous population is no longer predominantly rural in all Latin American countries, and already according to the 2010 round of censuses most Indigenous people lived in cities in four out of the 12 countries for which information was available (ECLAC et al., 2020[136]). Further, those who live in cities tend to live in deprived conditions: 36% of Indigenous urban dwellers in the region live in slums, nearly twice the proportion of non-Indigenous urban dwellers (World Bank, 2015[143]). This has important implications for responses to the pandemic focused on Indigenous peoples, as the concentration of Indigenous environmental migrants and displaced persons living in very precarious conditions in large cities exposes them disproportionately to the risk of illness and death from COVID (ECLAC et al., 2020[136]).
Nonetheless, many Indigenous people continue to live in rural areas, and at the regional level the Indigenous population accounts for 24% of the total rural population of Latin America (ECLAC et al., 2020[136]). As described in the previous section, rural areas face greater deprivation in terms of access to water and sanitation (necessary to prevent the spread of the virus), as well as in access to the Internet (which is needed to participate in remote schooling or economic activities during periods of social distancing). However, Indigenous peoples in rural areas tend to be especially marginalised, due to their remoteness from public services (including health care services), the continued encroachment and appropriation of Indigenous territories, and other factors linked to the systematic erosion of their political, economic, social and cultural rights (ECLAC et al., 2020[136]). The emphasis on communal life and practices in traditional Indigenous communities, while being a source of cultural resilience, also implies increased risk of spreading the disease during the pandemic (ECLAC et al., 2020[136]).
Approximately three to seven million Indigenous people live in forest areas, maintaining traditional languages, knowledge and cultural practices (ECLAC et al., 2020[136]). The relationship between forests and the Indigenous peoples who inhabit them is profound and reciprocal: the forests provide subsistence and cultural continuity for Indigenous communities, who in turn practice traditional and sustainable techniques of forest management and use that contribute to the restoration and adaptation of forests and their biodiversity. These areas are increasingly exposed to large-scale industrial activity such as mining and agriculture, which not only destroy forest habitats but also bring large numbers of external workers to the areas, thus promoting the spread of the virus. Indigenous peoples in voluntary isolation and in a phase of initial contact29 (covering an estimated 200 Indigenous groups, mainly in the Amazon and the Gran Chaco of Paraguay) are especially vulnerable in this respect, as monitoring activities to ensure their protection have been reduced during the pandemic (ECLAC et al., 2020[136]).
Indigenous peoples have implemented a number of collective efforts to address the pandemic, where official State responses have been lacking or deficient. For example, measures such as closing the territorial boundaries of communities have been implemented in almost all countries in the region, and without them it is likely that the health impact among Indigenous peoples would be even greater. Strategies of reciprocity and inter-community cooperation have made up for shortfalls in the coverage of humanitarian aid provided by governments, and traditional medicine techniques have been used to complement or replace formal healthcare, where access to formal health systems has been inadequate. Similarly, in the face of insufficient data to track the progress of the disease and mortality rates amongst Indigenous peoples, some communities have created their own epidemiological monitoring systems (ECLAC et al., 2020[136]).
The Afro-descendant population, on the other hand, is predominantly urban, with a level of urbanisation exceeding 70% in most countries in the region, and reaching 97% in Uruguay (ECLAC, 2021[142]). Due to the higher levels of poverty experienced by Afro-descendant people, they tend to live in more overcrowded dwellings, often in slums or informal settlements, making social distancing almost impossible (ECLAC, 2021[142]). The crisis has also put a spotlight on the vulnerabilities inherent in certain previously less visible occupations, especially in informal employment. For example, Afro-descendent women are more likely to work in the domestic sector than non-Afro-descendant women in most LAC countries with available data.30 Domestic work, being largely indoors and in close contact with employers or other clients, entails a higher exposure to the virus, whether it be in a private home or in medical or care environments. Further, the high rates of informality in the domestic care sector, and the essential nature of these service during the pandemic, has meant that domestic workers typically lacked the option to stay at home (ECLAC, 2021[142]).
Data from Brazil show clearly the disproportionate impact of the pandemic on the Afro-descendant population. Various studies and surveys from the first months of the pandemic (up to July 2020) showed that in Brazil, the second-highest risk factor for death from COVID among hospitalised people was being of African descent (the highest risk factor being age), and the Afro-descendant population was at 47% greater risk of death than the non-Afro-descendant population. Afro-descendant people were almost half as likely to work remotely as the non-Afro-descendant population (with 9% and 17.6% of the respective populations working from home). While education has played a role (with an illiterate Afro-descendent patient being 3.8 times more likely to die from COVID-19 than a non-Afro-descendant patient with higher education), even when comparing people with the same level of education, there were 37% more deaths amongst Afro-descendants, rising to 50% more when comparing people with higher education, suggesting that discrimination and racism have played a role (ECLAC, 2021[142]).
Issues for statistical development
Huge progress has been achieved in the measurement of ethno-racial inequalities in the Latin American region over recent decades. This has been thanks largely to the efforts of social movements advocating for better data to make the needs of Indigenous and Afro-descendant populations more visible, supported by the ongoing process of democratisation in the region (Telles and Paschel, 2014[144]). In the 1980s, only around half of Latin American countries identified Indigenous populations in their national censuses, and only two countries (Brazil and Cuba) included questions to differentiate Afro-descendants. By the 2010 round of censuses, almost every country in the region either included a question to identify Indigenous and Afro-descendant people or planned to do so (Loveman, 2021[145]; ECLAC, 2019[146]).
While censuses are powerful sources of information, they take place only once every decade. Further efforts are needed to improve the availability of disaggregated data by race and ethnicity (ideally identifying not only Indigenous status overall, but also the specific Indigenous groupings, where appropriate) across other data sources such as household surveys and administrative data. This includes the need for data that is better disaggregated by ethnicity and race for medical and death records so as to more accurately evaluate the differentiated health impact of the COVID crisis on Indigenous and Afro-descendant populations (ECLAC, 2021[142]; ECLAC et al., 2020[136]).
Many of the priorities for improving measures of well-being outcomes by ethnicity and race in the Latin American region are common with those in OECD countries. These include (Balestra and Fleischer, 2018[147]) :
Expanding all relevant data collection exercises to include ethnicity/race/Indigenous identity variables, while respecting the fundamental rights and privacy of individuals by ensuring appropriate measures for data protection and disclosure control.
Involving relevant communities in the processes of survey development (including the wording of question and response categories), validation of the accuracy of self-reported information, data collection efforts, and the dissemination of results. This will build trust and improve data quality.
Ensure the representation of hard-to-reach populations, such as Indigenous communities, through non-standard sampling techniques such as time-location sampling or respondent-driven sampling, and include these communities among pre-coded response options where applicable.
Gather information on diversity in both population censuses and surveys in order to provide robust demographic statistics and timely data that allow assessing multiple well-being outcomes and discriminatory experiences. Where possible, link census, sample survey data and administrative records pertaining to these populations.
When data are compared across two or more different collections, consideration needs to be given to how and when the data was collected. Also, assumptions about uncertainties in the resulting data need to be made explicit. Wherever possible, national statistical offices should invest in developing diversity statistical standards and provide clear guidance to improve consistency and comparability across all data sources (censuses, surveys, administrative data).
Allow respondents to declare more than one identity to allow for the fluidity of ethnic and racial classifications, and to better mirror the increasingly diverse make-up of societies. Statistical categories should reflect demographic changes as well as evolutions in the understanding of racial and ethnic identities.
On this last point, there is a need for more debate and reflection on multiple identities in the region and how to address this issue in national statistical systems, a discussion that should be held with organisations of Afro-descendant and Indigenous peoples. Ethnicity and race are social rather than biological constructs, meaning that the way people identify themselves (and are identified by others) is largely dependent on context and situation, allowing for multiple identities to co-exist (ECLAC, 2020[139]). However, official statistics in the region tend to employ self-identification with exclusionary categories, allowing for the capture only of the “main” category selected. Information on the size of different ethnic and racial populations has an undeniable political component, as it can impact the targeting of resources or of the population’s access to decision-making processes,31 so ensuring its accuracy and representativeness should be a priority.
The impact of ethnicity and race in shaping well-being outcomes is mediated by other intersecting variables such as gender, age, geographic location and socio-economic status. Indigenous and Afro-descendant women, people living in rural areas, the elderly, and people with lower education or other socio-economic markers tend to be more vulnerable, with those accumulating multiple risks being the most deprived. Better understanding the intersectionality of disadvantage requires including larger samples of ethnic and racial minorities in population surveys in order to allow for a more robust analysis of groups with multiple sources of vulnerability. It is also important that data on the situation of Indigenous and Afro-descendant peoples should be analysed with an awareness of the relevant social, territorial and cultural contexts.
From a more conceptual standpoint, the Indigenous notion of well-being, as encompassed in Ecuador’s Buen Vivir framework or Bolivia’s Vivir Bien framework, puts a greater emphasis on communal relations (including relations between the community and the natural environment) and collective practices than other Western societies (Garcia and Viteri, 2018[148]). Indigenous perspectives are seldom incorporated into well-being measurement exercises (although Ecuador and Bolivia are notable exceptions). This underscores the need to involve relevant communities in the process of survey development wherever possible. Incorporating Indigenous priorities would also entail a better measurement of important aspects specific to their communities such as territorial rights,32 the maintenance of language, cultural artefacts and representations, and the protection of sacred sites and traditional knowledge (OECD, 2019[149]).
Education allows individuals to acquire the skills needed to understand and master the world, opening opportunities and enhancing their control over their lives (OECD, 2011[150]). Despite improvements in educational attainment, still less than half of the population in the focal countries aged 25 or more has attained at least upper secondary education, compared to more than 70% on average in the OECD (see Chapter 3).
Figure 5.28 shows performance ratios for selected well-being outcomes and resources for future well-being for people with primary education (dark blue) and secondary education (light blue) in comparison to those with tertiary education, on average across the 11 LAC countries in the focal group. To ease understanding, all indicators are coded in the same direction so that the higher the ratio, the better the relative performance of people with primary and secondary education.
Education has a strong positive impact on people’s material living conditions (Figure 5.28, Panel A). In general, people with lower educational attainment experience lower material living conditions. People with primary education and those with secondary education are, respectively, 11 and 6 times more likely to be poor, are more likely to report having insufficient income to meet their needs (respectively twice and 50% more), and less likely to be employed, compared with people with tertiary education. When employed, they are more likely to be in informal employment and to earn less, and more likely to work long hours compared to those with tertiary education. Primary and secondary educated people are also more likely to fear losing their job and more likely to be unemployed. However, the likelihood to be unemployed is slightly higher for secondary educated than for primary educated people. This could be explained by the increasing labour market polarisation, mainly driven by digitalisation, which narrows the demand for middle-skill jobs in favour of low- and high-skill employment (OECD, 2020[151]; OECD, 2017[152]).
The relation between education and quality of life is less clear-cut (Figure 5.28, Panel B). On average, primary and secondary educated people, in comparison with the tertiary-educated, report lower life satisfaction, higher negative affect balance, and less social network support. Primary and secondary educated people are also less likely to voice their opinion to an official, while the perception that the country is governed by a few powerful groups for their own benefit is widespread and consistent across all education levels. People with primary education are three times more likely than people with tertiary education to report limitations in daily activities due to health problems, while people with secondary education are also slightly more likely to report these limitations. This pattern is partially explained by differences in the age distribution across education levels: due to the rise of educational attainment over time, the share of people aged 50 or more (who are also those who are more likely to report health limitations) is higher for those with primary education (41%) than for those with tertiary education (19%) on average across the focal countries. However, when comparing the share of those who report health limitations across the age groups, this is also consistently lower for those with tertiary education compared to those with primary education.
Satisfaction with services (health care and education) is much higher among people with primary and secondary education than among those with tertiary education. This can be partially explained by the fact that people with higher education have more awareness about the limitations of the education and health systems, as well as higher standards when evaluating them (Cárdenas et al., 2008[130]). Safety and the perception of safety are also higher among primary educated people: only 21% of them reported having been victim of a crime in the previous 12 months, compared to 27% and 32% for secondary and tertiary educated people, respectively. The education gap for perceived safety is smaller, with the share of those feeling safe when walking alone at night in their neighbourhood being only 4 percentage points higher for primary educated people relative to their tertiary-educated counterparts.
Educational inequalities in social capital are smaller (Figure 5.28, Panel C). While support for democracy, trust in others, volunteering and support for paying taxes are lower for people with primary or secondary education than for the tertiary educated, the perception of an unequal distribution of income is very similar across all levels of education (around 80% of people in all educational categories think that the distribution of income is unfair). Trust in government is higher among primary and secondary educated people (40% and 31% of them, respectively, trust the government, compared to 30% among tertiary educated). Perception of government corruption is high, but lower among primary educated people (69% of them think that corruption is widespread across the government) compared to secondary (75%) and tertiary (76%) educated. Compared to people with tertiary education, trust in the police is higher among people with primary education, but slightly lower among people with secondary education, compared with the tertiary-educated.
Note: Each performance ratio is the simple average by education calculated across the 11 focal countries for which data are available for all the inequality groups taken into consideration. As a result, performance ratios can cover a subset of the 11 focal countries. Performance ratios above 1 indicate better outcomes (i.e. higher well-being) for primary-educated and secondary-educated, whereas performance ratios below 1 indicate better outcomes (i.e. higher well-being) for tertiary-educated – including negative indicators, which have been reverse-scored. Grey bubbles denote no clear difference with the reference group (tertiary-educated), defined as education ratios within 0.03 points distance to parity.
Source: OECD calculations. For a complete list of sources please refer to the "Sources and Methods" tab in the Statlink file
The impact of COVID-19 on educational inequalities
The impact of COVID-19 on the lower-educated population is likely to have been more severe across a number of dimensions of well-being, given their vulnerability in terms of material conditions and some dimensions of quality of life. COVID-19 impacted the lower educated more severely, as they were more likely to lose their job or experience other forms of job disruption compared to higher educated workers (OECD, 2021[153]). When employed, lower educated people were more likely to be essential workers (e.g. working in transport, cleaning, essential retail), more at risk of being exposed to the virus and less likely to be able to telework, compared to higher educated workers (OECD, 2020[154]). As they are also more likely to be unemployed and to experience financial strain, lower educated people are also more likely to experience depression and anxiety. As noted in Chapter 3, school closures and the shift to remote learning in most countries is likely to have exacerbated gaps in learning outcomes, with a particularly negative impact on vulnerable students with poor Internet connections or weak digital skills or without enough space of their own to focus (OECD, forthcoming[59]).
The rapid spread of COVID-19 and the severity of its effects on human health have called for people to rapidly acquire and apply information on preventative measures and to adapt their behaviour to avoid getting or spreading the virus. Health literacy (i.e. the capacity to acquire, understand and use health information in a sound and ethical manner) has become critical during the pandemic to help people understand the reasons behind official recommendations and reflect on the outcomes of their actions (Paakkari and Okan, 2020[155]). People’s health literacy is influenced by their level of education, adding an additional dimension to the vulnerability of the low-educated.
Issues for statistical development
Information on well-being outcomes is often available by education, with the exception of housing conditions and infrastructures. However, for some indicators (such as insufficient income, fear of losing the job, perceived elite State capture) information is not available by educational attainment, but only by years of education, which do not necessarily inform on the level of education attained by individuals (because of the possibility of “repeat years”). Further harmonisation of education categories at the source level is needed to ensure a consistent approach to educational inequalities based on educational attainment across the well-being dimensions and indicators.
References
[111] Aarkrog, V. et al. (2018), “Decision-Making Processes Among Potential Dropouts in Vocational Education and Training and Adult Learning”, International Journal for Research in Vocational Education and Training, Vol. 5/2, pp. 112-129, https://doi.org/10.13152/ijrvet.5.2.2.
[11] Amarante, V., M. Colacce and F. Scalese (forthcoming), “Poverty and gender in Latin America: how far can income-based measures go?”.
[76] Arnesen, L. et al. (2016), “An analysis of three levels of scaled-up coverage for 28 interventions to avert stillbirths and maternal, newborn and child mortality in 27 countries in Latin America and the Caribbean with the Lives Saved Tool (LiST)”, BMC Public Health, Vol. 16/1, https://doi.org/10.1186/s12889-016-3238-z.
[147] Balestra, C. and L. Fleischer (2018), “Diversity statistics in the OECD: How do OECD countries collect data on ethnic, racial and indigenous identity?”, OECD Statistics Working Papers, No. 2018/09, OECD Publishing, Paris, https://dx.doi.org/10.1787/89bae654-en.
[33] Bott, S. et al. (2012), Violence Against Women in Latin America and the Caribbean: A comparative analysis of population-based data from 12 countries, Pan American Health Organization, Washington, DC, https://www.paho.org/hq/dmdocuments/2014/Violence1.24-WEB-25-febrero-2014.pdf.
[114] Cafagna, G. et al. (2019), Envejecer con cuidado: Atención a la dependencia en América Latina y el Caribe, Inter-American Development Bank, https://doi.org/10.18235/0001972.
[37] Caputi, J. and D. Russell (1990), “Femicide: speaking the unspeakable”, Ms, Vol. 1/2, pp. 34-37.
[130] Cárdenas, M. et al. (2008), Education and Life Satisfaction: Perception or Reality?, https://core.ac.uk/download/pdf/6783786.pdf.
[122] CDC (2020), Lesbian, Gay, Bisexual, and Transgender Health, https://www.cdc.gov/lgbthealth/youth.htm.
[67] Cecchini, S. et al. (eds.) (2015), Towards universal social protection: Latin American pathways and policy tools, CEPAL, https://repositorio.cepal.org/bitstream/handle/11362/39484/S1500752_en.pdf.
[140] Cecchini, S. et al. (2015), Towards universal social protection: Latin American pathways and policy tools, ECLAC, Santiago.
[86] CEPAL (ed.) (2013), Ageing, solidarity and social protection in Latin America and the Caribbean. Time for progress towards equity, https://repositorio.cepal.org/handle/11362/2620.
[123] Coker, T., S. Austin and M. Schuster (2010), “The Health and Health Care of Lesbian, Gay, and Bisexual Adolescents”, Annual Review of Public Health, Vol. 31/1, pp. 457-477, https://doi.org/10.1146/annurev.publhealth.012809.103636.
[88] Costa, A. and A. Ludermir (2005), “Transtornos mentais comuns e apoio social: estudo em comunidade rural da Zona da Mata de Pernambuco, Brasil”, Cadernos de Saúde Pública, Vol. 21/1, pp. 73-79, https://doi.org/10.1590/s0102-311x2005000100009.
[110] Cuartas, J. et al. (2019), “Early childhood exposure to non-violent discipline and physical and psychological aggression in low- and middle-income countries: National, regional, and global prevalence estimates”, Child Abuse & Neglect, Vol. 92, pp. 93-105, https://doi.org/10.1016/j.chiabu.2019.03.021.
[142] ECLAC (2021), “COVID-19 Reports: People of African descent and COVID-19: unveiling structural inequalities in Latin America”, ECLAC, Santiago.
[53] ECLAC (2021), COVID-19 Special Report No. 9: The Economic Autonomy of Women in a Sustainable Recovery with Equality, ECLAC, https://www.cepal.org/sites/default/files/publication/files/46634/S2000739_en.pdf.
[26] ECLAC (2021), “Repository of information on time use in Latin America and the Caribbean”, https://oig.cepal.org/sites/default/files/c2100061_web.pdf.
[3] ECLAC (2021), Social Panorama of Latin America 2020, ECLAC, https://www.cepal.org/sites/default/files/publication/files/46688/S2100149_en.pdf.
[31] ECLAC (2020), Addressing violence against women and girls during and after the COVID-19 pandemic requires financing, responses, prevention and data compillation, ECLAC, https://www.cepal.org/sites/default/files/publication/files/46425/S2000874_en.pdf.
[139] ECLAC (2020), Afro-descendants and the social inequality matrix: Inclusion challenges [Afrodescendientes y la matriz de la desigualdad social en América Latina: Retos para la inclusión], ECLAC, Santiago.
[109] ECLAC (2020), Latin America and the Caribbean and the COVID-19 pandemic: economic and social effects, https://www.cepal.org/en/publications/45351-latin-america-and-caribbean-and-covid-19-pandemic-economic-and-social-effects.
[160] ECLAC (2020), Measurement and status of young women and men in paid and unpaid work, World’s Women 2020, https://undesa.maps.arcgis.com/apps/MapJournal/index.html?appid=17627ede6e6241bab21c21deaf483ab1.
[106] ECLAC (2020), Persons with disabilities and coronovirus disease (COVID-19) in Latin America and the Caribbean: status and guidelines, https://repositorio.cepal.org/bitstream/handle/11362/45492/S2000299_en.pdf?sequence=1&isAllowed=y.
[129] ECLAC (2020), Reconstruction and transformation with equality and sustainability in Latin America and the Caribbean, https://repositorio.cepal.org/bitstream/handle/11362/46130/1/2000652_en.pdf.
[146] ECLAC (2019), “Aspectos conceptuales de los censos de población y vivienda: Desafíos para la definición de contenidos incluyentes en la ronda 2020 [Conceptual aspects of population and housing censuses: Challenges for the definition of inclusive content in the 2020 round.]”, serie Seminarios y Conferencias, No. 94 (LC/TS.2019/67), https://www.cepal.org/es/publicaciones/44944-aspectos-conceptuales-censos-poblacion-vivienda-desafios-la-definicion.
[156] ECLAC (2019), “Follow-up of the SDGs from a gender perspective in Latin America and the Caribbean”, Note prepared for the Work Session on Gender Statistics, Conference of European Statisticians, 15-17 May 2019, ECLAC.
[39] ECLAC (2019), Indicator: “Number of femicides or feminicides”, https://cepalstat-prod.cepal.org/cepalstat/tabulador/ConsultaIntegrada.asp?idIndicador=2780&idioma=i.
[61] ECLAC (2019), Time-use measurements in Latin America and the Caribbean, https://oig.cepal.org/sites/default/files/time_use-measurement_in_lac_0.pdf.
[97] ECLAC (2018), Social Panorama of Latin America 2017, ECLAC, Santiago, https://repositorio.cepal.org/bitstream/handle/11362/42717/6/S1800001_en.pdf.
[13] ECLAC (2018), Social Panorama of Latin America 2018, http://www.cepal.org/en/suscripciones.
[9] ECLAC (2017), Estrategia de Montevideo para la Implementación de la Agenda Regional de Género en el Marco del Desarrollo Sostenible hacia 2030 [Montevideo Strategy for the Implementation of the Regional Gender Agenda in the context of Sustainable Development to 2030].
[63] ECLAC (2016), Social Panorama of Latin America 2016.
[8] ECLAC (2016), The social inequality matrix in Latina America, United Nations, https://www.cepal.org/sites/default/files/events/files/s1600945_en.pdf.
[90] ECLAC (2015), Inclusive social development. The next generation of policies for overcoming poverty and reducing inequality in Latin America and the Caribbean, https://repositorio.cepal.org/bitstream/handle/11362/39101/4/S1600098_en.pdf.
[91] ECLAC (2015), Social Panorama of Latin America 2015, United Nations, https://repositorio.cepal.org/bitstream/handle/11362/39964/S1600174_en.pdf?sequence=5&isAllowed=y.
[44] ECLAC (2014), La reproducción en la adolescencia y sus desigualdades en América Latina. Introducción al análisis demográfico, con énfasis en el uso de microdatos censales de la ronda de 2010, https://www.cepal.org/es/publicaciones/36853-la-reproduccion-la-adolescencia-sus-desigualdades-america-latina-introduccion-al.
[12] ECLAC (2014), Social Panorama of Latin America 2014, ECLAC, https://repositorio.cepal.org/bitstream/handle/11362/37627/4/S1420728_en.pdf.
[125] ECLAC (2006), Manual sobre indicadores de calidad de vida en la vejez, https://www.cepal.org/es/publicaciones/3539-manual-indicadores-calidad-vida-la-vejez.
[60] ECLAC and UN Women (2021), “Measures and actions promoted by the Governments of Latin America and the Caribbean against COVID-19 in key areas for the autonomy of women and gender equality (Preliminary working document)”, https://www.cepal.org/sites/default/files/events/files/220222_documento_mapeo_medidas_covid-19_rev_dag_eng.pdf.
[136] ECLAC et al. (2020), The impact of COVID-19 on indigenous peoples in Latin America (Abya Yala): between invisibility and collective resistance, ECLAC, Santiago.
[62] ECLAC/INEGI/INMUJERES/UN-Women (2016), Classification of Time-Use Activities for Latin America and the Caribbean (CAUTAL).
[45] ECLAC/UNICEF (2007), Teenage motherhood in Latin America and the Caribbean. Trends, problems and challenges, https://www.cepal.org/en/publications/36002-teenage-motherhood-latin-america-and-caribbean-trends-problems-and-challenges.
[138] ECLAC and FILAC (2020), The Indigenous people of Latin America - Abya Yala and the SDG Agenda: Tensions and challenges from a territorial perspective [Los pueblos inidígenos de América Latina - Abya Yala y la agenda 2030 para el Desarrollo Sostenible: Tensiones y desafíos desde una perspectiva territorial].
[54] ECLAC and ILO (2020), “Employment trends in an unprecedented crisis: policy challenges”, Employment Situation in Latin America and the Caribbean, No. 23, https://repositorio.cepal.org/bitstream/handle/11362/46309/4/S2000600_en.pdf.
[95] ECLAC-ILO (2018), Employment situation in Latin America and the Caribbean. Labour market participation of older persons: needs and options, ECLAC/ILO, https://www.ilo.org/wcmsp5/groups/public/---americas/---ro-lima/---sro-santiago/documents/publication/wcms_630074.pdf.
[83] Escotto, T. (2015), Las juventudes centroamericanas en contextos de inseguridad y violencia, CEPAL, https://repositorio.cepal.org/bitstream/handle/11362/39229/1/S1500621_es.pdf.
[132] European Commission et al. (2020), A recommendation on the method to delineate cities, urban and rural areas for international statistical comparisons, https://unstats.un.org/unsd/statcom/51st-session/documents/BG-Item3j-Recommendation-E.pdf.
[40] Eurostat (2021), Indicator: “Intentional homicide rate for women by intimate partner or family member/relative”, https://appsso.eurostat.ec.europa.eu/nui/show.do?dataset=crim_hom_vrel&lang=en.
[133] Fadic, M. et al. (2019), “Classifying small (TL3) regions based on metropolitan population, low density and remoteness”, OECD Regional Development Working Papers, No. 2019/06, OECD, Paris.
[4] Ferreira, F. (2020), Inequality and social unrest in Latin America: The Tocqueville Paradox revisited, World Bank blog.
[15] Gallup Inc. and ILO (2017), Towards a Better Future for Women and Work: Voices of women and men, Gallup Inc., https://www.ilo.org/wcmsp5/groups/public/---dgreports/---dcomm/---publ/documents/publication/wcms_546256.pdf.
[148] Garcia, N. and J. Viteri (2018), “Methodological proposal for the construction of well-being measures in Ecuador [Propuesta metodológica para la Construcción de Medidas de Bienestar en el Ecuador]”, INEC Instituto Nacional de Estadísticas y Censos, Quito, https://www.ecuadorencifras.gob.ec/documentos/web-inec/Bibliotecas/Libros/Documento_metodologico_Metricas_de_Bienestar_11122018.
[14] Gasparini, L. et al. (2015), “Female Labor Force Participation in Latin America: Evidence of Deceleration”, No. 181, Universidad Nacional de la Plata, Centro de Estudios Distributivos, Laborales y Sociales (CEDLAS), https://www.econstor.eu/bitstream/10419/127697/1/cedlas-wp-181.pdf.
[16] Gender Equality Observatory for Latin America and the Caribbean (2021), Distribution of total employed population by productivity level and sex, https://oig.cepal.org/en/indicators/distribution-total-employed-population-productivity-level-and-sex.
[30] Gherardi, N. (2016), Otras forma de violencia contra las mujeres que reconocer, nombrar y visibilizar, ECLAC, https://www.cepal.org/sites/default/files/publication/files/40754/S1601170_es.pdf.
[35] GHRC - USA (n.d.), Femicide and Feminicide: Fact Sheet, Guatemala Human Rights Commission, Washington, DC, http://www.ghrc-usa.org/Programs/ForWomensRighttoLive/factsheet_femicide.pdf.
[87] Gigantesco, A. et al. (2019), “The Relationship Between Satisfaction With Life and Depression Symptoms by Gender”, Frontiers in Psychiatry, Vol. 10, https://doi.org/10.3389/fpsyt.2019.00419.
[49] Global Health 50/50, APHRC and ICRW (2021), The Covid-19 Sex Disaggregated Data Tracker: February Update Report, https://globalhealth5050.org/the-sex-gender-and-covid-19-project/about-us/.
[34] Gracia, E. (2004), “Unreported cases of domestic violence against women: towards an epidemiology of social silence, tolerance, and inhibition The ’’iceberg’’ of domestic violence”, Journal of Epidemiology and Community Health, Vol. 58, pp. 536-537, https://doi.org/10.1136/jech.2003.019604.
[77] Grove, J. et al. (2015), “Maternal, newborn, and child health and the Sustainable Development Goals—a call for sustained and improved measurement”, The Lancet, Vol. 386/10003, pp. 1511-1514, https://doi.org/10.1016/s0140-6736(15)00517-6.
[28] Heise L and Garcia Moreno C (2002), “Violence by intimate partners”, in Krug E.G., L. et al. (eds.), World Report on Violence and Health, World Health Organization, Geneva.
[100] Hobbs, C. et al. (2020), “COVID-19 in Children: A Review and Parallels to Other Hyperinflammatory Syndromes”, Frontiers in Pediatrics, Vol. 8, https://doi.org/10.3389/fped.2020.593455.
[113] IDB (2017), ¿Cómo viven los adultos mayores en América Latina y el Caribe?, https://publications.iadb.org/publications/spanish/document/Panorama_-_Como_viven_los_adultos_mayores_en_ALC_es_es.pdf.
[164] ILO (2021), SDG Indicator 8.3.1: Indicator: 8.3.1: Proportion of informal employment in total employment, by sector and sex, https://unstats.un.org/sdgs/indicators/database/.
[22] ILO (2021), SDG Indicator 8.8.1: Fatal and non-fatal occupational injuries per 100,000 workers, by sex and migrant status, https://ilostat.ilo.org/topics/safety-and-health-at-work/.
[112] ILO (2020), Global Employment Trends for Youth 2020: Technology and the future of jobs, https://www.ilo.org/wcmsp5/groups/public/---dgreports/---dcomm/---publ/documents/publication/wcms_737648.pdf.
[17] ILO (2019), Panorama Laboral America Latina y el Caribe 2019 [Labour Overview for Latin America and the Caribbean 2019], ILO Regional Office for Latin America and the Caribbean, Peru, https://www.ilo.org/wcmsp5/groups/public/---americas/---ro-lima/documents/publication/wcms_732198.pdf.
[20] ILO (2018), Women and men in the informal economy: a statistical picture (Third edition).
[158] ILO (2017), Global Estimates of Child Labour: Results and trends 2012-2016, ILO, Geneva, https://www.ilo.org/wcmsp5/groups/public/---dgreports/---dcomm/documents/publication/wcms_575499.pdf.
[75] ILO (2017), Regional factsheet for the Americas: 2017 Global Estimates of Modern Slavery and Child Labour, https://www.ilo.org/wcmsp5/groups/public/@ed_norm/@ipec/documents/publication/wcms_597871.pdf.
[24] ILO (2016), Formalizing Domestic Work, ILO.
[18] ILO (2016), Women at Work Trends 2016.
[81] ILO (2015), Youth and Informality Promoting Formal Employment among Youth: Innovative Experiences in Latin America and The Caribbean, ILO Regional Office for Latin America and the Caribbean, Lima.
[19] ILO Regional Office for Latin America and the Caribbean (2019), Women in the World of Work: Pending Challenges for Achieving Effective Equality in Latin America and the Caribbean, ILO, Lima, https://www.ilo.org/wcmsp5/groups/public/---americas/---ro-lima/documents/publication/wcms_736930.pdf.
[117] INEC (2018), Encuesta Nacional sobre Discapacidad [National survey on disability], https://www.inec.cr/encuestas/encuesta-nacional-sobre-discapacidad.
[38] INEGI (2019), Statistics related to the International Day of the Eradication of Violence Against Women, 25 November [“Estadísticas a propósito del día internacional de la eliminación de la violencia contra la mujer, 25 noviembre], INEGI, https://www.inegi.org.mx/contenidos/saladeprensa/aproposito/2019/Violencia2019_Nal.pdf.
[120] INEI (2014), Primera Encuesta Nacional Especializada sobre Discapacidad 2012 [First National Specialised Survey on Disability 2012], https://www.inei.gob.pe/media/MenuRecursivo/publicaciones_digitales/Est/Lib1171/ENEDIS%202012%20-%20COMPLETO.pdf.
[118] INSP (2016), Encuesta Nacional de Percepción de la Discapacidad en Población Mexicana 2010 [National Survey of Perceptions of Disability in the Mexican Population 2010], https://encuestas.insp.mx/enpdis/index.php.
[163] Inter-American Commission on Human Rights /Rapporteurship on the Rights of Indigenous Peoples (2013), Indigenous Peoples in Voluntary Isolation and Initial Contact in the Americas: Recommendations for the full respect of their human rights, OAS Organization of American States, http://www.oas.org/en/iachr/indigenous/docs/pdf/report-indigenous-peoples-voluntary-isolation.pdf.
[52] Inter-American Development Bank (2018), The Future of Work in Latin America and the Caribbean: Education and Health, the Sectors of the Future?, https://publications.iadb.org/en/future-work-latin-america-and-caribbean-education-and-health-sectors-future-interactive-version.
[29] Jewkes, R., P. Sen and C. Garcia Moreno (2002), “Sexual violence”, in Krug, E. et al. (eds.), World Report on Violence and Health, World Health Organization, Geneva, https://apps.who.int/iris/bitstream/handle/10665/42495/9241545615_eng.pdf;jsession.
[23] Jutting, J. and J. de Laiglesia (2009), Is Informal Normal? Towards more and better jobs in developing countries, OECD Development Centre, https://www.oecd-ilibrary.org/docserver/9789264059245-en.pdf?expires=1615473078&id=id&accname=ocid84004878&checksum=31453504A7E68FF504E107A704E4295B.
[124] Kann, L. et al. (2016), “Sexual Identity, Sex of Sexual Contacts, and Health-Related Behaviors Among Students in Grades 9–12 — United States and Selected Sites, 2015”, MMWR. Surveillance Summaries, Vol. 65/9, pp. 1-202, https://doi.org/10.15585/mmwr.ss6509a1.
[89] Kawachi, I. (2001), “Social Ties and Mental Health”, Journal of Urban Health: Bulletin of the New York Academy of Medicine, Vol. 78/3, pp. 458-467, https://doi.org/10.1093/jurban/78.3.458.
[71] Kendig, S., M. Mattingly and S. Bianchi (2014), “Childhood Poverty and the Transition to Adulthood”, Family Relations, Vol. 63/2, pp. 271-286, https://doi.org/10.1111/fare.12061.
[5] Langman, J. (2019), From Model to Muddle: Chile’s Sad Slide Into Upheaval.
[116] Lansbury, L., C. Brown and J. Nguyen-Van-Tam (2017), “Influenza in long-term care facilities”, Influenza and Other Respiratory Viruses, Vol. 11/5, pp. 356-366, https://doi.org/10.1111/irv.12464.
[165] Latinobarómetro (2015), “Degree of agreement: Women should work only if the partner does not earn enough money” [Grado de acuerdo: Mujeres deben trabajar sólo si la pareja no gana suficiente], https://www.latinobarometro.org/latOnline.jsp.
[145] Loveman, M. (2021), “The politics of a datascape transformed: ethnoracial statistics in Brazil in regional comparative perspective [A política de um cenário de dados transformado: estatísticas etnorraciais no Brasil em uma perspectiva comparativa regional]”, Sociologias, Vol. 23/56, pp. 110-153.
[131] Lustig, N. and M. Tommasi (2020), Covid-19 and social protection of poor and vulnerable groups in Latin America: a conceptual framework, UNDP, https://www.latinamerica.undp.org/content/rblac/en/home/library/crisis_prevention_and_recovery/covid-19-and-social-protection-of-poor-and-vulnerable-groups-in-.html.
[41] Michaeljon, A., E. Bell and J. Holden (2016), DFID Guidance Note: Shifting Social Norms to Tackle Violence against Women and Girls (VAWG), VAWG Helpdesk, London.
[10] Mostafa, T. (2019), ““Why don’t more girls choose to pursue a science career?””, PISA in Focus 93.
[126] NASEM (2015), Strengthening the Scientific Foundation for Policymaking to Meet the Challenges of Aging in Latin America and the Caribbean, National Academies Press, Washington, D.C., https://doi.org/10.17226/21800.
[55] OECD (2021), Man Enough? Measuring Masculine Norms to Promote Women’s Empowerment, Social Institutions and Gender Index, OECD Publishing, Paris, https://dx.doi.org/10.1787/6ffd1936-en.
[65] OECD (2021), Measuring What Matters for Child Well-being and Policies, OECD Publishing, Paris, https://dx.doi.org/10.1787/e82fded1-en.
[153] OECD (2021), “Risks that matter 2020: The long reach of COVID-19”, OECD Policy Responses to Coronavirus (COVID-19), OECD Publishing, Paris, https://dx.doi.org/10.1787/44932654-en.
[161] OECD (2020), A Territorial Approach to the Sustainable Development Goals: Synthesis report, OECD Urban Policy Reviews, OECD Publishing, Paris, https://dx.doi.org/10.1787/e86fa715-en.
[99] OECD (2020), “Combatting COVID-19’s effect on children”, OECD Policy Responses to Coronavirus (COVID-19), OECD Publishing, Paris, https://dx.doi.org/10.1787/2e1f3b2f-en.
[104] OECD (2020), “COVID-19: Protecting people and societies”, OECD Policy Responses to Coronavirus (COVID-19), OECD Publishing, Paris, https://dx.doi.org/10.1787/e5c9de1a-en.
[154] OECD (2020), OECD Employment Outlook 2020: Worker Security and the COVID-19 Crisis, OECD Publishing, Paris, https://dx.doi.org/10.1787/1686c758-en.
[135] OECD (2020), OECD Regions and Cities at a Glance 2020, OECD Publishing, Paris, https://dx.doi.org/10.1787/959d5ba0-en.
[42] OECD (2020), SIGI 2020 Regional Report for Latin America and the Caribbean, Social Institutions and Gender Index, OECD Publishing, Paris, https://dx.doi.org/10.1787/cb7d45d1-en.
[151] OECD (2020), “What is happening to middle-skill workers?”, in OECD Employment Outlook 2020: Worker Security and the COVID-19 Crisis, OECD Publishing, Paris, https://dx.doi.org/10.1787/c9d28c24-en.
[57] OECD (2020), “Women at the core of the fight against COVID-19 crisis”, OECD Policy Responses to Coronavirus (COVID-19), OECD Publishing, Paris, https://dx.doi.org/10.1787/553a8269-en.
[56] OECD (2020), Women at the Core of the Fight Against the Covid-19 Crisis, OECD.
[115] OECD (2019), Health at a Glance 2019: OECD Indicators, OECD Publishing, Paris, https://dx.doi.org/10.1787/4dd50c09-en.
[149] OECD (2019), Linking Indigenous Communities with Regional Development, OECD Rural Policy Reviews, OECD Publishing, Paris, https://dx.doi.org/10.1787/3203c082-en.
[93] OECD (2019), OECD Economic Surveys: Argentina 2019, OECD Publishing, Paris, https://dx.doi.org/10.1787/0c7f002c-en.
[92] OECD (2019), OECD Economic Surveys: Colombia 2019, OECD Publishing, Paris, https://dx.doi.org/10.1787/e4c64889-en.
[46] OECD (2019), SIGI 2019 Global Report: Transforming Challenges into Opportunities, Social Institutions and Gender Index, OECD Publishing, Paris, https://dx.doi.org/10.1787/bc56d212-en.
[2] OECD (2019), Under Pressure: The Squeezed Middle Class, OECD Publishing, Paris, https://dx.doi.org/10.1787/689afed1-en.
[94] OECD (2018), OECD Economic Surveys: Brazil 2018, OECD Publishing, Paris, https://dx.doi.org/10.1787/eco_surveys-bra-2018-en.
[6] OECD (2017), How’s Life? 2017: Measuring Well-being, OECD Publishing, Paris, https://dx.doi.org/10.1787/how_life-2017-en.
[152] OECD (2017), OECD Employment Outlook 2017, OECD Publishing, Paris, https://dx.doi.org/10.1787/empl_outlook-2017-en.
[66] OECD (2017), Preventing Ageing Unequally, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264279087-en.
[69] OECD (2015), How’s Life? 2015: Measuring Well-being, OECD Publishing, Paris, https://dx.doi.org/10.1787/how_life-2015-en.
[150] OECD (2011), How’s Life?: Measuring Well-being, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264121164-en.
[59] OECD (forthcoming), COVID-19 and Well-Being Evidence Scan.
[80] OECD/CAF/ECLAC (2016), Latin American Economic Outlook 2017: Youth, Skills and Entrepreneurship, OECD Publishing, Paris, https://dx.doi.org/10.1787/leo-2017-en.
[128] OECD/European Commission (2020), Cities in the World: A New Perspective on Urbanisation, OECD Urban Studies, OECD Publishing, Paris, https://dx.doi.org/10.1787/d0efcbda-en.
[96] OECD/IDB/The World Bank (2014), Pensions at a Glance: Latin America and the Caribbean, OECD Publishing, Paris, https://dx.doi.org/10.1787/pension_glance-2014-en.
[21] OECD/ILO (2019), Tackling Vulnerability in the Informal Economy, Development Centre Studies, OECD Publishing, Paris, https://dx.doi.org/10.1787/939b7bcd-en.
[79] OECD/The World Bank (2020), Health at a Glance: Latin America and the Caribbean 2020, OECD Publishing, Paris, https://dx.doi.org/10.1787/6089164f-en.
[155] Paakkari, L. and O. Okan (2020), COVID-19: health literacy is an underestimated problem, Elsevier Ltd, https://doi.org/10.1016/S2468-2667(20)30086-4.
[101] PAHO (2020), “Weekly Press Briefing on the COVID-19 Situation in the Americas”, https://www.paho.org/en/media/weekly-press-briefing-covid-19-situation-americas.
[78] PAHO (2017), Health in the Americas+, 2017 Edition. Summary: Regional Outlook and Country Profiles, https://iris.paho.org/bitstream/handle/10665.2/34321/9789275119662_eng.pdf?sequence=6&isAllowed=y.
[36] Russell, D. and N. Van de Ven (eds.) (1976), Crimes Against Women: The Proceedings of the International Tribunal. East Palo Alto, CA: Frog in the Well; 1976., Frog in the Well, East Palo Alto, CA.
[1] Sánchez-Ancochea, D. (2021), The Costs of Inequality in Latin America: Lessons and Warnings for the Rest of the World, I. B. Tauris, London.
[73] Santana, V., L. Kiss and A. Andermann (2019), “The scientific knowledge on child labor in Latin America”, Cadernos de Saúde Pública, Vol. 35/7, https://doi.org/10.1590/0102-311x00105119.
[134] Santos, M. (2019), “Non-monetary indicators to monitor SDG targets 1.2 and 1.4: Standards, availability, comparability and quality”, Statistics series, No. No. 99 (LC/TS.2019/4), ECLAC, Santiago.
[119] SENADIS (2015), Segundo Estudio Nacional de la Discapacidad [Second National Study on Disability], https://www.senadis.gob.cl/pag/355/1197/ii_estudio_nacional_de_discapacidad.
[84] Soto, H. and D. Trucco (eds.) (2015), Youth: Realities and Challenges for Achieving Development with Equality, https://repositorio.cepal.org/bitstream/handle/11362/40015/1/S1501235_en.pdf.
[7] Stiglitz, J., J. Fitoussi and M. Durand (eds.) (2018), For Good Measure: Advancing Research on Well-being Metrics Beyond GDP, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264307278-en.
[166] Sudre, C., B. Murray and T. Varsavsky (2020), “Attributes and predictors of Long COVID: analysis of COVID cases and their symptoms collected by the Covid Symptoms Study App”, https://www.medrxiv.org/content/10.1101/2020.10.19.20214494v1.full.pdf.
[141] Telles, E. (2014), Pigmentocracies: Ethnicity, Race, and Color in Latin America, University of North Carolina Press.
[144] Telles, E. and T. Paschel (2014), “Who is Black, White or Mixed Race? How skin color, status and nation shape racial classification in Latin America”, American Journal of Sociology, pp. 864-907.
[70] Thévenon, O. et al. (2018), “Child poverty in the OECD: Trends, determinants and policies to tackle it”, OECD Social, Employment and Migration Working Papers, No. 218, OECD Publishing, Paris, https://dx.doi.org/10.1787/c69de229-en.
[157] Tsirigotis, K., W. Gruszczynski and M. Tsirigotis (2011), “Gender differentiation in methods of suicide attempts”, Medical Science Monitor, Vol. 17/8, pp. 65-70, https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3539603/.
[43] Ullman, H. (2018), Main challenges faced by young people in Latin America and the Caribbean, https://www.ohchr.org/Documents/Issues/Youth/ECLAC.pdf.
[47] UN Women (2021), Towards parity and inclusive participation in Latin America and the Caribbean: Regional overview and contributions to CSW65, UN Women, New York, https://www.cepal.org/sites/default/files/document/files/lac_consultation_csw65.pdf.
[58] UN Women (2020), “The pandemic’s impact due to COVID-19 on violence against women”, https://lac.unwomen.org/en/noticias-y-eventos/articulos/2020/11/impacto-de-la-pandemia-covid-en-violencia-contra-las-mujeres.
[48] UN Women (2020), From Insights to Action: Gender Equality in the wake of Covid-19, https://www.unwomen.org/-/media/headquarters/attachments/sections/library/publications/2020/gender-equality-in-the-wake-of-covid-19-en.pdf?la=en&vs=5142.
[25] UN Women (2020), “Sexual Harassment in the Informal Economy: Farmworkers and domestic workers”, https://www.unwomen.org/-/media/headquarters/attachments/sections/library/publications/2020/discussion-paper-sexual-harassment-in-the-informal-economy-en.pdf?la=en&vs=4145.
[27] UN Women (2019), Progress of the world’s women 2019–2020: Families in a changing world, https://www.unwomen.org/-/media/headquarters/attachments/sections/library/publications/2019/progress-of-the-worlds-women-2019-2020-en.pdf?la=en&vs=3512.
[74] UNDESA (2020), Sustainable Development, https://sdgs.un.org/.
[127] UNDP (2020), UNDP in Latin America and the Caribbean, https://www.latinamerica.undp.org/content/rblac/en/home.html.
[159] UNESCO (2021), UNESCO Science Report: The race against time for smarter development, UNESCO, Paris, https://www.unesco.org/reports/science/2021/en/race4smarter-development.
[85] UNESCO (2017), School violence and bullying: global status report, https://unesdoc.unesco.org/ark:/48223/pf0000246970?posInSet=1&queryId=e2a1a7e5-847e-4351-8eed-92dfc642211c.
[98] UNICEF (2021), UNICEF data: Covid 19 and children, https://data.unicef.org/covid-19-and-children/.
[121] UNICEF (2021), UNICEF MICS, http://mics.unicef.org/.
[103] UNICEF (2020), COVID-19: Más del 95 por ciento de niños y niñas está fuera de las escuelas de América Latina y el Caribe [More than 95 per cent of children are out of school in Latin America and the Caribbean], https://www.unicef.org/mexico/comunicados-prensa/covid-19-m%C3%A1s-del-95-por-ciento-de-ni%C3%B1os-y-ni%C3%B1as-est%C3%A1-fuera-de-las-escuelas-de.
[102] UNICEF (2020), Impact of COVID-19 on Children and Families in Latin America and the Caribbean, https://www.unicef.org/lac/media/14381/file/UNICEF_LACRO_COVID19_impact.pdf.
[107] UNICEF (2020), UNICEF data: Covid 19 and children, https://data.unicef.org/covid-19-and-children/.
[108] UNICEF (2020), Youth speak up about violence during COVID-19, https://www.unicef.org/lac/en/youth-speak-about-violence-during-covid-19.
[72] UNICEF/CEPAL (2019), Las mediciones multidimensionales de pobreza infantil en América Latina y el Caribe y a nivel internacional [Multidimensional measures of child poverty in Latin America and the Caribbean and at the international level], https://www.unicef.org/lac/sites/unicef.org.lac/files/2019-10/PDF%20Las%20mediciones%20multidimensionales%20de%20pobreza%20infantil%20en%20Am%C3%A9rica%20Latina%20y%20el%20Caribe%20y%20a%20nivel%20internacional.pdf.
[162] United Nations (1948), Universal Declaration of Human Rights, https://www.un.org/en/about-us/universal-declaration-of-human-rights.
[82] UNODC (2019), Global Study on Homicide, https://www.unodc.org/documents/data-and-analysis/gsh/Booklet1.pdf.
[64] Villatoro, P. (2017), Indicadores no monetarios de pobreza: avances y desafíos para su medición: Memoria del seminario regional realizado en Santiago, los días 15 y 16 de mayo de 2017 [Non-monetary indicators of poverty: measurement achievements and challenges, Summary of a regional seminar in Santiago, 15-16 May 2017], ECLAC, Santiago.
[105] WFP (2020), El impacto de COVID-19 en programas de comidas escolares en América Latina y el Caribe [The impact of COVID-19 on school meal programmes in Latin America and the Caribbean], https://historias.wfp.org/26-de-33-paises-han-suspendido-sus-programas-de-comidas-escolares-en-america-latina-y-el-caribe-5687c79e75a3.
[32] WHO (2013), Global and regional estimates of violence against women: prevalence and health effects of intimate partner violence and non-partner sexual violence, http://www.who.int.
[51] World Bank (2020), Gender Dimensions of the COVID 19 Pandemic.
[50] World Bank (2020), Policy Note: Gender dimensions of the Covid-19 pandemic, World Bank Group.
[68] World Bank (2020), World Development Indicators, https://data.worldbank.org/indicator/SP.POP.65UP.TO.ZS?locations=ZJ.
[143] World Bank (2015), Indigenous Latin America in the Twenty-First Century: the first decade, World Bank Group, Washington, DC, https://documents1.worldbank.org/curated/en/145891467991974540/pdf/Indigenous-Latin-America-in-the-twenty-first-century-the-first-decade.pdf.
[137] World Bank Group (2018), Afro-descendants in Latin America: Towards a Framework of Inclusion, World Bank Group, Washington, DC.
Notes
← 1. As noted by Deere, Kanbur and Stewart (2018[7]), “any significant horizontal inequality is unjust since there is no reason why people should receive unequal rewards or have unequal political power merely because they are black rather than white, women rather than men, or of one ethnicity rather than another”; at the same time, “horizontal inequalities have been shown to raise the risk of violent conflict significantly”, (as they) “provide powerful grievances which leaders can use to mobilise political protest, by calling on cultural markers (e.g. a common history or language or religion) and pointing to group exploitation” (p. 87).
← 2. These issues are not completely absent from data collections. On the contrary, migration has long been included as a background variable in censuses, administrative records and some household surveys, although the under-measurement of the migration population remains a challenge. Disability has also been considered in many recurrent measurement instruments in the region, but there is a lack of standardisation and therefore of comparable measures on this issue. And as for sexual orientation and gender identity, there is an almost total lack of data.
← 3. Throughout this report, the eleven focal countries refer to Argentina, Brazil, Chile, Colombia, Costa Rica, Dominican Republic, Ecuador, Mexico, Paraguay, Peru and Uruguay.
← 4. One explanation for men’s increased suicide rates may be the pressure of restrictive norms of masculinity. When men fail to comply with the masculine norms dictated by society, this may induce important psycho-social consequences (OECD, 2021[55]). However, it should also be noted that while worldwide men are two to three times more likely to commit suicide, women are more likely to suffer episodes of major depression, and when counting both successful and unsuccessful suicide attempts, women are more likely than men to attempt suicide (Tsirigotis, Gruszczynski and Tsirigotis, 2011[157]).
← 5. It should also be emphasised that these differences are likely not due to natural differences in ability by gender, but are rather the consequence of social conditioning through discriminatory norms and the educational environment, and that this has the effect of encouraging boys’ performance and discouraging girls’ performance in these areas (UNESCO, 2021[159]).
← 6. Possible explanations could include the fact that men tend to participate more in the public space and therefore experience some forms of discrimination that women do not. Another possible explanation is that social norms normalise some types of discrimination, making women less likely to be aware of the discriminatory aspects of their situations. However, more research is needed to fully understand this result and to confirm its validity.
← 7. Once again, the disproportionate burden of unpaid care and domestic work taken on by women plays a role. In the LAC region, 57.8% of women aged 15-29 who are not in employment, education or training (categorised as NEET) are engaged in unpaid care and domestic work (as are 66.1% of women aged 25-29), compared with only 7%of men aged 15-29 in the NEET category. (ECLAC, 2020[160]).
← 8. Throughout this chapter, as for the rest of this report, ‘poverty’ refers to the absolute poverty rate as calculated by ECLAC, and ‘extreme poverty’ refers to the extreme poverty rate as calculated by ECLAC, unless otherwise stated (see Chapter 2, Box 2.1).
← 9. As explained in Box 2.1 (Chapter 2), the extreme poverty threshold is calculated as the value necessary to purchase the basic food basket without additional goods and services, while absolute poverty adds to the costs of the food basket those of essential non-food components.
← 10. It should be noted, however, that women with zero income need not be poor (and indeed, due to the dominance of traditional family structures many of the region’s most affluent families are likely to be headed by a sole male earner with the wife earning no income). This indicator therefore says as much about women’s agency and overall economic autonomy as it does about their poverty outcomes.
← 11. Results across the focal countries ranged from 17.5% of respondents in Brazil to 51.3% in Mexico (Latinobarómetro, 2015[165]).
← 12. The gap is even larger when considering non-agricultural employment: on average, across the focal countries, 50% of women’s non-agricultural employment was informal in 2019, compared with 46% of men’s (ILO, 2021[164]).
← 13. However, it should also be noted that the methodology used in LAC countries to record time use is different from that used in most OECD countries, hence the two values are not fully comparable.
← 14. Argentina, Brazil, Chile, Colombia, Ecuador, El Salvador, Guatemala, Honduras, Mexico, Peru and Uruguay.
← 15. According to survey data from 11 Latin American countries in the mid to late 2000s (ranging from 2004 to 2009), in most countries the reported prevalence of partner violence was two to three times greater among women whose first live birth occurred before age 17 (or age 15) compared with those whose first birth occurred after age 24 (Bott et al., 2012[33]).
← 16. From (Bott et al., 2012[33]): “[There are] a number of findings that suggest exposure to violence in childhood may have long-term and intergenerational effects. For example, after controlling for other factors, the most consistent risk factor for experiencing physical or sexual intimate partner violence against women across all countries was a history of ‘father beat mother’. Similarly, the prevalence of intimate partner violence was significantly higher (usually around twice as high) among women who reported having experienced physical abuse in childhood compared with those who did not. Partner violence was also significantly higher (usually more than twice as high) among women who reported experiencing sexual abuse in childhood compared with those who did not. In addition, children living in households where women had experienced intimate partner violence were significantly more likely than other children to be punished with hitting, beating, spanking, or slapping (note that surveys did not always identify who punished the children).”
← 17. Gender imbalances in the health impact of Covid also need to be considered beyond overall infection and death rates. A preliminary study in the United Kingdom, based on data collected via a symptom tracker app, indicated that women under 60 were much more likely to suffer from “long Covid” symptoms (lasting longer than a month, and with the potential of leading to long-term illness), with women in the 40-50 age group twice as likely as similarly-aged men to be affected (Sudre, Murray and Varsavsky, 2020[166]).
← 18. Latin American governments state the importance of gender-specific information systems in the Montevideo Strategy (ECLAC, 2017[9]), a joint declaration of the priorities for the implementation of the Regional Gender Agenda in the context of the UN 2030 Agenda. Statistical offices in the region also emphasised the importance of the gender perspective when formulating the prioritised set of SDG indicators for Latin America and the Caribbean. As a result, the regional framework agreed by the Conference of Statisticians of the Americas (CEA) underlines the importance of monitoring the structural challenges faced in the pursuit of gender equality, particularly with respect to time use and women’s physical and economic autonomy (ECLAC, 2019[156]).
← 19. Chapter 3 on Quality of Life includes the mortality rate of children aged under five, as it is an important indicator of overall health status and systems. Chapter 4 on the Resources for Future Well-being includes the NEET rate (Youth not in education, employment or training), youth informal employment, youth educational attainment and child malnutrition, given the importance of these indicators not only at an individual level, but also as a reflection of the stock of human capital within societies.
← 20. Although gender differences in child labour may also be due to under-reporting for girls, who are more likely to be involved in less visible forms of labour such as domestic work in households (ILO, 2017[158]). Girls are also more likely to be involved in unpaid work: global estimates show that 55% of children performing household chores are female (Thévenon et al., 2018[70]).
← 21. In the context of Indigenous peoples, it is important to distinguish exploitative forms of child labour from domestic and productive activities that take place in childhood as part of family support and knowledge transferral strategies based on the formative processes of their own culture. It is a fundamental element in the processes of upbringing and the transmission of ancestral knowledge and traditions and constitutes a way of progressively developing skills and abilities for adult life. Therefore, it is part of their “right freely to participate in the cultural life of the community”, as stated in Article 27 of the Universal Declaration of Human Rights (United Nations, 1948[162]). The available data do not allow to differentiate between Indigenous people living in traditional communities and those who do not.
← 22. For example in the Children’s World survey, which uses visual and story-telling techniques to elicit meaningful responses (OECD, 2021[65]).
← 23. No information is, however, available on urban slums and informal settlements, which are excluded from the data on overcrowding in urban settlements.
← 24. Beyond digital infrastructure, the share of jobs that are amenable to remote work (which is linked to the skills profile of predominant occupations) is also an important determinant of exposure to the virus.. Recent OECD work has shown that capital regions have the highest potential for remote working, with rates that are 8 percentage points higher than the respective country average (OECD, 2020[135]).
← 25. Regions with sub-national government are also responsible, including through public spending, for many public policies that matter for well-being and SDGs – particularly for Federal countries such as Mexico, Brazil and Argentina. A 2016 review of OECD countries found that OECD sub-national governments were responsible for around 40% of total public expenditure and 60% of total public investment. Of these public resources, at least 70% were invested in core areas of the SDGs, such as education, public services, economic affairs and environmental protection (OECD, 2020[161]).
← 26. Figure 5.24 gives only a general indication of differences rather than a precise assessment of the current situation, as data availability and timeliness vary widely depending on the measure (see the Statlink and the Note to Figure 5.24 for more detail). These issues reflect the overall shortcomings of available data and the need for more timely and comprehensive data on well-being by ethnicity and race in the region (see the later section on “Issues for Statistical Development”). Nonetheless, the over-arching message that Indigenous and Afro-descendant people experience worse outcomes than their comparison group in most well-being measures is valid and clear.
← 27. Afro-descendants were 3% more likely to work in the informal sector in Brazil (2015) and Uruguay (2005) and 1.3% more likely in Colombia (2015). However in Ecuador, Afro-descendants were 3.5% less likely to work in the informal sector (World Bank Group, 2018[137]).
← 28. However, it should be noted that interpreting the overcrowding indicator is not straightforward for Indigenous communities, as living in close proximity may be associated with residential and kinship patterns specific to each culture and – in that sense – would denote cultural robustness.
← 29. Defined as follows: “Indigenous peoples in voluntary isolation are Indigenous peoples or segments of Indigenous peoples who do not maintain sustained contacts with the majority non-Indigenous population, and who generally reject any type of contact with persons not part of their own people. They may also be peoples or segments of peoples previously contacted and who, after intermittent contact with the non-Indigenous societies, have returned to a situation of isolation and break the relations of contact that they may have had with those societies….. Indigenous peoples in initial contact are Indigenous peoples or segments of Indigenous peoples who maintain intermittent or sporadic contact with the majority non-Indigenous population, generally used in reference to peoples or segments of peoples who have initiated a process of contact recently. However, ‘initial’ should not necessarily be understood as a temporal term, but as a reference to the scant extent of contact and interaction with the majority non-Indigenous society.” (Inter-American Commission on Human Rights /Rapporteurship on the Rights of Indigenous Peoples, 2013[163]).
← 30. Afro-descendant female workers were more likely to be domestic workers than non-Afro-descendant female workers across all five focal countries with available data from the latest census (Brazil, 2010; Costa Rica, 2011; Ecuador, 2010; Mexico, 2015; Peru, 2018) (ECLAC, 2020[139]).
← 31. For example, in Chile, the share of the Indigenous population in the abbreviated 2017 Census was the basis for determining the number of reserved seats for Indigenous representatives in the Constitutional Convention process to reform the Chilean constitution.
← 32. Also relevant for Afro-descendant peoples.