# 3. Quality of life in Latin America

In the OECD well-being framework, Quality of Life comprises health, knowledge and skills, safety, environmental quality, civic engagement, social connections, work-life balance and subjective well-being. For each of these dimensions, this chapter provides an overview of the levels and trends across each indicator where data are available for the focal group, before discussing the potential impacts of the COVID-19 pandemic and the issues for statistical development. Overall trends in quality of life in the focal group of countries prior to the COVID-19 are encouraging and point to significant improvements in people’s well-being over the past two decades. Nonetheless, across a number of outcomes explored in this chapter, average levels for the focal group1 are being held back by certain countries, where the potential impacts of the pandemic are of particular concern.

In terms of health, indicators of health status highlight considerable progress among the focal group, but satisfaction with health care has decreased over time, and out-of-pocket expenditures remain high in four out of six countries where data are available. Despite improvements over the past two decades, indicators of knowledge and skills underscore the disparities both between and within certain countries of the focal group. This area is also highly relevant in the context of the digital transformation, as the increasing importance of digital skills means that inequalities in Internet access and ICT skills have the potential to worsen existing well-being inequalities throughout the region. Although homicides are still relatively high in certain focal group countries and have increased in others, on average fewer people reported that they had been assaulted, attacked or a victim of crime in the previous 12 months in 2018 than in 2001. However, indicators of perceived safety and of road deaths are yet to improve. Regarding environmental quality in the focal group, mean average population exposure to PM2.5 air pollution has remained reasonably stable since 2000, though in 2019, 91% of people in the countries analysed were still exposed to dangerous levels (i.e. more than 10 micrograms/m3). In some focal group countries, dissatisfaction with the public sphere has been a source of social unrest in recent years, and the indicators used to assess civic engagement in this report show a marked fall in the share of people declaring to have voiced their opinion to an official, and an increase in those who believe their country is governed by a few powerful groups for their own benefit. Between 2006-09 and 2017-19, indicators of social connections and of subjective well-being remained relatively high, close to levels recorded in the OECD.

The COVID-19 pandemic has affected people’s quality of life dramatically in the focal group of countries, as individuals have coped in every way they could with several waves of increased deaths and disease, extended lockdowns and economic hardship. Early evidence reported in this chapter suggests that in Latin America the pandemic exacerbated pre-existing deprivations in terms of access to health care, whilst increasing people’s loneliness, depressive states and substance use. School closures are likely to have affected children and adolescents unequally, as students from poorer socio-economic backgrounds risk bearing long-lasting consequences in terms of lower learning outcomes and fewer job opportunities. While extended lockdowns across most countries of Latin America and the Caribbean kept people off the streets, they had mixed consequences on crime and environmental quality. However, the social unrest and political polarisation in the lead-up to the pandemic underline the urgency for countries to create opportunities for citizens and stakeholders, and to allow them to engage in efforts to rebuild trust, improve services and enhance social cohesion.

For selected indicators where data are available from Gallup World Poll, this chapter takes a closer look at change between 2019 and 2020. On average in the focal group, the satisfaction with health care services remained relatively stable, hiding diverging trends across countries. On the other hand, satisfaction with the education system dropped across a majority of countries, amplifying disparities across the focal group in 2020. Finally, in certain countries, average levels of social network support and of life satisfaction decreased considerably between 2019 and 2020 compared to previous years, whilst higher negative affect balance reflects the burden of the crisis on people’s mental states.

Health is fundamental to people’s well-being, and it is consistently ranked as one of the most valued aspects of people’s lives.2 The ability to lead a long and healthy life not only has clear intrinsic value, but it is also instrumentally important in enhancing people’s opportunities to participate in education, the labour market and community life. Health in its broadest sense refers to “a state of complete physical, mental and social well-being and not merely the absence of disease or infirmity” (WHO, 1948[1]). While health can be understood as a multidimensional and positive concept, data limitations mean that it is most frequently measured with a focus on disease, disability and mortality, rather than the presence of more positive health states. To understand health at the broader population level, well-being frameworks often rely on indicators of longevity, years lived in good health, self-reported health, mental health symptoms and sometimes health-related behaviours.

Since 2000, life expectancy in the focal group of 11 countries has increased by 3.5 years on average, and child and maternal mortality have both decreased. However, progress across countries remains unequal, and differences in levels persist: for example, there is a 6-year gap in life expectancy at birth between the top- and bottom-performing LAC (Latin America and the Caribbean) 11 countries, while infant mortality is four times higher in the worst-performing country, relative to the top-performing country. Prior to the COVID-19 pandemic, people’s satisfaction with the availability of quality health care was already on the decline across most countries of the focal group, despite an overall improvement in health-care coverage. Around one in five people in the focal group countries report that they experience limitations in their daily activities due to poor health, which is close to OECD average levels, while the prevalence of recorded suicides remains well below the OECD average in a majority of these countries. Smoking, drinking alcohol, and especially the prevalence of overweight and obesity are critical risk factors for poor health in Latin America, but these indicators are covered in the “Human Capital” section of Chapter 4 on Resources for Future Well-being.

Latin America has been severely affected by the COVID-19 pandemic, and it has been one of the hardest-hit regions in terms of deaths worldwide (Dong, Du and Gardner, 2020[2]). Moreover, estimates suggest that 21% of the population in Latin America have at least one factor (such as obesity) that puts them at higher risk of severe COVID-19 disease (LSHTM CMMID COVID-19 working group, 2020[3]). These data are of particular concern in a context where Latin American countries face challenges for delivering accessible, affordable and safe health care due to high levels of informality and inequalities.

Life expectancy at birth is the widest-used summary measure of population health status and is often used to gauge a country’s overall health. It measures how long, on average, a new-born infant can expect to live if current death rates do not change. Life expectancy at birth has increased by 3.7 years across all countries in the focal group since 2000, from 73 years to 76.7 years on average in 2018 (Figure 3.1). Generally, this increase has been driven by the steady reduction of mortality at all ages, particularly infant and child mortality (OECD/The World Bank, 2020[4]). Convergence to the levels achieved in countries where life expectancy is highest has been relatively slow, and the gap between the focal group and the OECD has slightly widened, by 0.4 years on average since 2000. Wide differences exist among the focal group: in Costa Rica, a new-born child can expect to live over 6.2 years more than in the Dominican Republic. This is despite an improvement of 4.5 years in the Dominican Republic, which is among the countries that gained the most since 2000, along with Colombia (4.2 years), Peru (5.4 years) and Brazil (5.6 years). Over the same period, life expectancy at birth has remained relatively stable in Mexico (Figure 3.1).

In many Latin American countries, child mortality rates have historically been very high, and improvements in health outcomes for children in the first five years of life have been particularly significant in driving increased life expectancy in the region across the last two decades. Child and maternal mortality rates are especially important health indicators, since they reflect the impact of economic, social and environmental conditions on children and mothers, and they also indicate the overall effectiveness of health systems within a country.

On average in 2019, the child mortality rate (deaths per 1 000 live births) was 13.5, around half the 2000 rate (26.4) and around three times the 2019 OECD average rate (4.4). While all countries have experienced improvements, the same inter-country differences are evident among the focal group countries, with the rate in the Dominican Republic (28) being four times higher than in Chile (7). This exceeds the target set by the SDGs for 2030 (at least as low as 25 per 1 000 live births by 2030) by three points (Figure 3.2, Panel A).

Maternal mortality – the death of a woman during pregnancy or childbirth or within 42 days of the termination of pregnancy – is an important indicator of women’s health status, but also in assessing the performance of a country’s health system. This has declined from 84 deaths per 100 000 live births in 2000 to 58 in 2017, on average in the focal group of countries. Seven out of the 11 focal group countries have now achieved the SDG target of fewer than 70 maternal deaths per 100 000 live births. Nevertheless, levels in 2017 remained high when compared to those of OECD countries (on average). The largest gains have been achieved by countries that had the highest levels in 2000, and that remain above the focal group average even today: Paraguay, Peru and Ecuador. Conversely, maternal mortality increased by almost 20% in the Dominican Republic, reversing the gains achieved in the early 2000s (Figure 3.2, Panel B).

Country performance among focal group countries is very similar across the two indicators shown in Figure 3.2, with the same countries at the top and bottom of the group for both (Chile, Uruguay and Costa Rica at the top end; Dominican Republic and Paraguay at the bottom). This reflects the existence of common drivers: births unattended by health professionals, for example, are a cause of both child and maternal mortality.

Premature mortality rates offer some insights into public health and the success of government policies in tackling preventable and treatable causes of death among non-elderly populations – whether due to accidents or suicides, violence, infectious and parasitic (communicable) diseases, or non-communicable diseases (NCDs) such as cardiovascular diseases, cancers, chronic respiratory disease and diabetes. For example, effective health care systems and public policies can play an important role in mitigating some common risk factors for premature deaths due to NCDs, including tobacco use, harmful use of alcohol, unhealthy diets, physical inactivity and air pollution (Khaltaev and Axelrod, 2019[6]), while advances in medical technology and care can sometimes prevent such chronic diseases from resulting in premature death.

On average across the focal countries, adult mortality (defined as the probability of dying between ages 15 and 60, derived from life tables) was 124 per 1 000 in 2016, down from an average rate of 152 per 1 000 in 2000 (Figure 3.3). While adult mortality in the focal group was consistently below the LAC regional average between 2000 and 2016, the gap narrowed slightly towards the end of the period due to a slower rate of improvement in the focal group. Over the same period, the gap between the OECD average and the focal group average widened slightly.

Beyond levels of mortality, understanding the causes of death is key for assessing the effectiveness of a country’s health-care system, but also for identifying national priorities in terms of public health and other policy areas, such as security (OECD/The World Bank, 2020[4]). Figure 3.4 provides an overall picture of the burden of disease, injury and other risk factors for people’s health in Latin America. Non-communicable diseases (such as cardiovascular disease and cancer) are the most common cause of death globally, and the focal group is no exception, where NCDs are responsible for 79% of all deaths on average. The share is highest in Chile and Uruguay (86%) but remains below average OECD levels (89%). However, communicable diseases (CDs), such as respiratory infections, diarrhoeal diseases and tuberculosis, along with maternal and perinatal conditions, remain significant causes of death among many countries in the focal group, accounting for 11% of all deaths, on average. In Costa Rica, the share is only 6%, but it is over three times higher in Peru (20%). The remaining 10% of deaths in the focal group of countries are attributed to injuries and violence, with levels ranging from 6% in Argentina to 13% in Colombia and Ecuador.3 In the LAC region on average, the share of deaths attributed to injuries and violence (12%) is twice as high as in the OECD on average (6%).

Mental and neurological disorders (ranging from depression and anxiety to bipolar disorder) account for almost a quarter of the disease burden in Latin America and the Caribbean (WHO, 2013[8]).4 These disorders are often undertreated: in 2016, the treatment gap (i.e. the percentage of people with disorders that do not receive any treatment) for severe mental disorders in Latin America was almost 70% (Kohn et al., 2018[9]). Beyond the direct health toll, mental health can interact with and affect many other aspects of well-being, including work and job quality (e.g. through sickness absences, disengagement at work, disability and unemployment) (OECD/The World Bank, 2020[4]) as well as income, education and social connectedness. There is a two-way relationship between mental disorders and socio-economic status: mental disorders tend to lead to reduced employment and income, thereby entrenching poverty, while poverty, in turn, increases the risk of mental disorder (WHO and Calouste Gulbekian Foundation, 2014[10]).

Comparable data on the prevalence and intensity of mental health problems in the Latin American region are not available. Evidence is however available on suicides, which may be considered as the extreme manifestation of mental health problems, particularly depression. Suicides accounted for an estimated 800 000 deaths in 2018 worldwide, with 79% of them occurring in low and middle-income countries (WHO, 2019[11]). In the absence of comparable measures of mental health, suicide rates can provide some insight into levels of severe mental health problems across countries, despite issues regarding the interpretability and comparability of these data (Figure 3.5).5

Most focal group countries experience lower suicide rates than the OECD average. Unlike the OECD average, where suicides have decreased over time, focal group and regional trends since 2000 have remained relatively stable, with Brazil experiencing a notable increase (up 1.7 deaths per 100 000 to 6.5 deaths in 2016) and increases of at least 1 death per 100 000 in Mexico, Ecuador and Uruguay. However, there are marked disparities between countries in terms of level: in 2016, with fewer than five suicides per 100 000 population in Peru, but over 18 in Uruguay (Figure 3.5) – twice as high as the LAC regional average, and well above the OECD average, as a result of a steady rise in recent decades (Fachola et al., 2015[12]).

Universal health coverage (UHC) is achieved when all people, communities and social groups have access to the health services they need, when these services have a high degree of quality and when users can access these services without incurring financial hardship (OECD/WHO/World Bank Group, 2018[13]). Based on this definition, health systems in Latin American countries have significant weaknesses and are often underfinanced, segmented and fragmented, resulting in significant barriers to access (ECLAC-PAHO, 2020[14]).

Access to health care also depends on whether households can afford care services. The proportion of the population spending more than 10% of their income (or expenditure) on health care can give an idea of the financial hardship linked to direct health payments in the focal group countries (UN DESA, 2019[15]). Figure 3.6, Panel B shows that, on average among the six focal countries for which data are available, approximately 9% of households incurred out-of-pocket health-care expenditures exceeding 10% of their income over the 2010-18 period, a share that has remained broadly stable relative to the previous decade. That share has been falling in Colombia but rose by around 3 percentage points or more in Chile and Costa Rica. Just below 2% of the population has been incurring much higher out-of-pocket health-care expenditures (25% or more of their total income or expenditures on average in focal group countries), a share that has been broadly stable over time.

For certain households, the consequence of excessive out-of-pocket expenditure on health care is to be driven into poverty. In the focal group of countries where data are available, 1.7% of the population have been pushed below the “societal” poverty line by out-of-pocket health care expenditure, compared to 1.3% in OECD countries (Figure 3.7). Figure 3.6, Panel B illustrates that a relatively high proportion of the population are driven into poverty in countries where a high share of households make out-of-pocket payments exceeding 10% of their income or expenditure (for instance in Chile and Colombia). Similarly, in Mexico, where out-of-pocket health care expenditures are relatively low, less than 1% of the population have fallen below the societal poverty line as a result of them.

Worldwide, public sector organisations, departments and agencies regularly monitor users’ satisfaction with public health services to evaluate the impact of reforms and identify areas calling for further action. Data regularly collected through the Gallup World Poll allows some comparative analysis of citizens’ satisfaction with a range of public services, including health care (OECD, 2017[16]). Among the focal countries, half of the population (50%) were satisfied with the availability of quality health care in the city or area where they lived in 2017-19, which is close to the wider regional average (48%) and 20 percentage points below the OECD average of 69%. Over two out of three respondents declared to be satisfied in Uruguay (69%) and Costa Rica (64%). However, in another four countries, the majority of respondents were not satisfied (Colombia, Peru, Brazil and Chile). Across the focal group, trends are mixed: on average, satisfaction with health care fell by 5 percentage points over this period, with declines that are three to four times greater in Chile and Colombia. On the other hand, satisfaction with health care increased slightly in Paraguay (from 46% to 51%), whilst remaining relatively stable in the Dominican Republic (56%) (Figure 3.8).

Over the course of 2020, Latin America and the Caribbean was one of the regions hit hardest by the coronavirus pandemic (COVID-19), both in terms of reported cases and deaths. As of April 2021, the region accounted for 19% of confirmed cases worldwide, and for 28% of confirmed deaths, despite representing only 9% of the world population (Dong, Du and Gardner, 2020[2]; Worldometer, 2021[17]).7 In Peru, confirmed deaths per 100 000 population were higher than in any other country in the world by mid-2021 (586.41), followed by Brazil (239.15), Colombia (201.24) and Argentina (200.90). Brazil (18.1 mln), Argentina (4.3 mln) and Colombia (4 mln) were also among the top ten countries in terms of confirmed cases worldwide (Dong, Du and Gardner, 2020[2]). The pandemic has severely affected adults of all ages – including the young (PAHO, 2021[18]). However, the number of confirmed COVID-19 deaths may differ from the pandemic’s true death toll due to the way they are reported and to the way COVID-19 impacted the number of deaths occurring due to other causes (Lopez-Calva, 2020[19]).

The long-lasting consequences of the pandemic are likely to be worse for informal workers and economically vulnerable households in the region. Close to 60% of workers in the LAC are informal (OECD et al., 2020[20]). Many are self-employed in a subsistence, daily-living economy and at risk of slipping back into poverty (OECD et al., 2020[20]). Individuals lacking access to social protection must continue to work to make a living regardless of the social distancing measures put in place, limiting their capability to protect themselves and their households. As seen in Figure 3.6, Panel A, approximately a quarter of the population in Latin America as a whole did not have access to essential health-care services prior to the pandemic: these individuals will have seen their access even more restricted over the course of 2020. Moreover, in the LAC region almost 8% of people are aged 65 or more, over 80% are urban, and 21% of the urban population live in slums, informal settlements or housing where basic services are not available (OECD/The World Bank, 2020[4]). Lack of access to quality health care and information is also acute in remote rural areas, where large shares of indigenous peoples live. Another barrier affecting indigenous peoples’ access to health care is the lack of an intercultural approach encompassing native customs and languages in the management and provision of health services (UN, 2020[21]). However, as seen in Figure 3.6, Panel B, among the six focal countries for which data are available, approximately 9% of households incurred out-of-pocket health-care expenditures exceeding 10% of their income over the 2010-18 period. What is more, Figure 3.8 highlights that a majority of Latin Americans are dissatisfied with the availability of quality health care, as opposed to the OECD, where 69% of the population are satisfied on average. Combined, these factors are exacerbating the pandemic’s risks. Resolving the fragmentation, commodification and hierarchisation of health systems will hence be a crucial challenge for the region moving forward (ECLAC, 2020[22]).

Both prevention and treatment for chronic and non-communicable diseases have been heavily disrupted since the beginning of the COVID-19 pandemic, meaning that those living with them are at much higher risk of severe COVID-19-related illness and death (ECLAC-PAHO, 2020[14]; WHO, 2020[23]). Estimates suggest that 21% percent of the population in Latin America have at least one factor that puts them at higher risk of severe COVID-19 disease (LSHTM CMMID COVID-19 working group, 2020[3]).8 Obesity is one of these risk factors (Sattar, McInnes and McMurray, 2020[24]): in Latin America, 60% of the population are overweight and 25% are obese (see “Human Capital” section in the following chapter). In addition, the mortality rate for respiratory disease is far higher in Latin America than in the OECD average, particularly in focal countries such as Argentina, Brazil and Peru (WHO, 2018[25]). There has also been a reduction in access to sexual and reproductive health services during the pandemic, which are key to women’s health and may hinder country efforts in fighting maternal mortality (World Bank, 2006[26]). This could result in a lack of care for sexually transmitted infections, and a resulting increase in these infections (UNFPA, 2020[27]). Unwanted pregnancies could also become an issue of even greater importance, in a region with the second-highest adolescent pregnancy rate in the world (estimated at 66.5 births per 1 000 girls aged 15-19), after sub-Saharan Africa (PAHO/UNFPA/UNICEF, 2017[28]). Finally, the high share of older adults living with younger generations in the region (52% live with one or more of their children (UNDESA, 2017[29])) is a factor that increases the risk of infection.

While the effects of the COVID-19 pandemic on physical medical conditions have received great attention, there are also concerns about its impact on mental health, which can take the form of fear, worry or concern induced by the contagion among the population at large and specific groups. For example, in a global YouGov study, one in two Mexicans reported that the pandemic had a negative impact on their mental health (51%), and almost one in four reported suffering from at least one mental health condition in 2020 (22%) (YouGov, 2020[30]). More widely, 27% of young Latin Americans (aged 13-29) reported feeling anxiety and 15% depression in the previous seven days, during the first months of the pandemic (UNICEF, 2020[31]). Lockdown measures are likely to have increased people’s loneliness, substance abuse and self-harm (WHO, 2020[32]). It is therefore vital to include mental health and psychosocial support in national response plans to the pandemic. In a survey carried out across 29 countries of the Americas (27 of which belong to Latin America and the Caribbean) with designated respondents, 93% of countries reported that such support systems were indeed included in their response plans, yet only 7% (2 countries) ensured full funding for them in their government budget, while another 31% (9 countries) reported having no funding for mental health activities (PAHO, 2020[33]). One country that has taken action in the region is Chile: in 2018, it allocated the lowest share of health spending to mental health amongst all OECD countries, at 2.1% of government health spending, and it announced in February 2021 that the budget for mental health would increase by 310% compared to the previous budget (Ministerio de Salud, 2021[34]; OECD, 2021[35]). Moving forward, it will be crucial to understand how well countries are delivering the services and policies that matter for achieving good mental health outcomes, as measured in the OECD Mental Health System Performance Benchmark, for instance (OECD, 2021[35]).

Finally, Gallup World Poll data from 2020 show that during the first year of the COVID-19 pandemic, satisfaction with health care was impacted in different ways across the focal group, compared to 2019. On average, the level of satisfaction remained relatively stable at 48% (Figure 3.9). Nonetheless, a handful of countries recorded clear decreases in the share of people satisfied with the availability of quality health care in the city or area where they live: some of the largest declines were recorded in Brazil (-6 percentage points), the Dominican Republic (-8), Paraguay (-10) and Peru (-14). Conversely, the share increased in Argentina (by 4 percentage points), Colombia (+5 points), Costa Rica (+8 points) and Chile (+12 points). Overall, these trends have widened the disparities among focal group countries, as levels of satisfaction with health care are now almost three times higher in Costa Rica (71%) than in Peru (25%). Further analysis of satisfaction with health care in the region following the pandemic will be provided in (OECD, forthcoming[36]).

The frequent and timely publication of data on life expectancy, mortality and co-morbidity is key to gain insight on health trends in a country, but practices vary across Latin America. Both life expectancy and mortality data rely on vital registration systems that are incomplete in many developing countries, with about one-third of countries in Latin America not having recent data. Unregistered deaths are common in Peru and are also high in Colombia and Ecuador (OECD/The World Bank, 2020[4]). Furthermore, although administrative data on specific conditions such as cancer and diabetes are available, they do not address co-morbidity (different conditions affecting the same individual). This, however, is vital for understanding the prevailing incidence of different diseases across the population and to provide insight on people’s health-related quality of life (OECD, 2020[39]).

The measure of life expectancy used in this chapter refers to length of life, regardless of health conditions during those years. Measures of “healthy” life expectancy (also referred to as “disability-free life expectancy” exist but are not yet internationally comparable (except for Europe). Furthermore, although measures of people’s functioning (i.e. their capacity to perform daily activities) have been recommended by the Washington Group on Disability Statistics, and international guidance exists, harmonised measures are not available for the region (United Nations, 2005[40]; Washington Group on Disability Statistics, 2016[41]). This area of statistical development is highly relevant for Latin American countries, since previous estimates suggested that approximately 66 million people in the region live with at least one disability (ECLAC, 2013[42]).

Comparable measures of mental health outcomes are globally scarce, including in Latin America. Identifying comparable measures at the population level (as opposed to people diagnosed or treated by medical professionals) remains a challenge. Moreover, the stigma associated with mental health may lead to further difficulties such as under-reporting, which could potentially impact cross-country comparability or the interpretation of changes in prevalence rates over time (OECD, 2020[39]). Data on suicides, such as those reported in Figure 3.5, under-represent the scale of the phenomenon, whilst also failing to account for suicide attempts – which are often much higher. Estimates suggest that for each adult who died by committing suicide globally, there may have been over 20 other attempts (WHO, 2021[43]). Furthermore, self-reports of suicide attempts could also be exposed to considerable under-reporting and comparability issues, arguably even more so than other symptoms of psychological distress.

Finally, in the context of the COVID-19 pandemic, international statistics have been developed at an unprecedented speed. Nonetheless, divergences in reporting mortality statistics (mentioned above) are particularly problematic for assessing the pandemic’s health impacts in Latin America. While most countries have published mortality statistics related to COVID-19, death certificates are filled out differently from one country to another, and testing practices for the virus also vary. As a result, certain fatalities may be categorised as being related to the pandemic in some jurisdictions and not in others. What is more, certain patients may have died from the disruption that the pandemic caused to health-care systems, rather than from the virus itself. The international comparability of mortality statistics related to COVID-19 is hindered by the differences in coding and reporting practices, which underscores the importance of other measures, such as high-frequency data on the number of deaths from all causes – from which excess mortality statistics may be derived (Morgan et al., 2020[44]). By comparing overall numbers with the level of expected deaths in a given country based on the same period in previous years, excess mortality statistics can provide an indication of the overall impact of COVID-19. This can be achieved by accounting not only for deaths directly attributed to COVID-19, but also those that may be uncounted or indirectly linked, such as deaths caused by delayed or foregone treatment due to an overloaded health system (Morgan et al., 2020[44]).

Moving forward, leveraging digital solutions and data to better detect, prevent, respond to and recover from the sanitary and economic crises will be a major challenge for the region. It will also be vital to adequately manage the risks of diverting resources to ineffective digital tools, the exacerbation of inequalities and the violation of privacy, both during and after the outbreak (OECD/The World Bank, 2020[4]).

Education and skills bring a wide range of benefits to society, including higher economic growth, stronger social cohesion and less crime (OECD, 2011[45]). At an individual level, receiving a good education is of intrinsic value and responds to the basic need to learn and to adapt to a changing environment. Knowledge and skills have a positive impact on material living conditions, since higher levels of education lead to higher earnings and greater employability, better health status as well as an increased chance of working in an environment with fewer health hazards. People with a higher level of education are also more likely to report higher levels of support from friends and relatives and are more satisfied with their lives overall (OECD, 2017[46]). Finally, education provides individuals both with the knowledge to enjoy some leisure activities such as reading and participating in cultural events, and more importantly with the skills to integrate fully into society, by fostering civic awareness and political participation (OECD, 2011[45]; OECD, 2016[47]).

In Latin America, educational attainment has improved over the past two decades, but several indicators show that the region is lagging in other areas, and that disparities persist both within and between countries. Results from the OECD Programme for International Student Assessment (PISA) are a sign of the progress achieved across all participating focal countries, yet on average, student competencies in the region remain well below those attained in OECD countries. This is also highlighted by a greater share of low achievers in the region (below Level 2), particularly among socio-economically disadvantaged students. Evidence also shows that adults’ skills have also improved, as witnessed by the current literacy rate of 95%. Finally, satisfaction with education varies greatly across the Latin American countries included in the focal group, showing improvements in some cases but deteriorations in others.

COVID-19 has disrupted the learning cycle for approximately 154 million students in the region, as most schools remained closed in an effort to contain the pandemic (Basto-Aguirre, Cerutti and Nieto-Parra, 2020[48]; OECD, forthcoming[36]). This risks interrupting the progress in students’ skills made in the focal group, whilst widening disparities between countries and exacerbating inequalities within them.

Increasing educational attainment has been an important goal in OECD and focal countries alike. Upper secondary education is considered today as the minimum qualification level for successful integration in society and labour markets (OECD, 2017[49]). On average, the share of the population aged 25 and above having completed at least upper secondary education is 26 percentage points lower in the focal group of countries (46%) than in the OECD (72%). Across these countries, disparities are wide: in Uruguay, only 30% of the population has attained an upper secondary education, which is roughly half the share attained in Chile (59%) (Figure 3.10, Panel A).

Trends in upper secondary educational attainment have been positive. As a result of strong gains in six countries (where educational attainment since 2000 has increased by 15 percentage points or more), the focal group average has increased by 13 percentage points. Over this period, attainment rates have improved in all focal countries, although some countries are lagging. For instance, both Argentina and Uruguay had approximately the same share of the population aged 25 or above having completed at least an upper secondary education in the first years of the century. However, since then, upper secondary attainment has increased by 23 percentage points in Argentina and fallen by four in Uruguay (Figure 3.10, Panel A).

Tertiary education opens up further opportunities to people. For example, in OECD countries, adults with a tertiary degree are 10 percentage points more likely to be employed, and their life expectancy is longer than that of people with a low level of education (8 years longer for men, and 5 years longer for women (Murtin et al., 2017[50]).9 People with a tertiary degree are also less likely to suffer from depression than their less-educated peers (OECD, 2019[51]). Students who complete university also earn higher salaries later on in life: in Brazil, Chile, Colombia and Costa Rica, relative earnings for full-time, full-year 25-64 year-old workers with a tertiary education are over twice as high as for those with upper secondary education (against 54% on average in OECD and partner countries) (OECD, 2020[52]).10

In six focal group countries, approximately 20% of adults aged 25 or above have attained tertiary education, compared to 30% in OECD countries where data are available (Figure 3.10, Panel B). Cross-country disparities are somewhat lower than for upper secondary education, although overall levels are much lower: 9 percentage points separate Uruguay (13%) and Costa Rica (22%). As with trends in upper secondary educational attainment, both the focal group average and the Latin America regional average have experienced large gains in the share of the population having completed tertiary education since 2000, reaching 19% on average (with gains of 7 percentage points for both). Considerable gains were achieved in Paraguay, where the share has risen by 11 percentage points between 2005 and 2018, but the level remains at just 15%.

While educational attainment is a measure of the quantity of education received, the quality of the skills acquired during schooling years also has a major impact on people's life chances. The OECD Programme for International Student Assessment (PISA) assesses what students know and can do in reading, mathematics and science towards the end of their compulsory schooling (at age 15). Results have been used to assess the quality of learning outcomes attained by students around the world, as well as how these learning outcomes differ across students with different characteristics. As such, they allow educators as well as policy makers to learn from the policies and practices applied in other countries (OECD, 2019[55]).

Among the eight focal group countries that participated in PISA 2018, 15-year-old students in Chile, Uruguay, Mexico and Costa Rica tended to have the highest cognitive skills scores across the three subjects, while the Dominican Republic, Peru, Argentina, Brazil and Colombia fell below the focal group average (Figure 3.11). Despite overall improvement, 15-year-old students in Latin America are yet to achieve the cognitive skills of OECD countries. Performance gaps are wide, with the Dominican Republic lagging other focal group countries by a significant margin (e.g. PISA scores in Science are almost 25% lower than in Chile, the highest-performing country).

Trends in PISA scores are generally positive across all three subjects among the focal countries (Figure 3.11). On average, the improvement in scores was highest in reading (a 16-point gain) and lowest in mathematics (+8 points). Progress since around 2006 has been greatest in Peru where, on average, 15-year-old students have improved their grades by 31 points in reading, and 35 points in mathematics and in science – but where, nevertheless, grades remain below the LAC 8 average. Students in Brazil and Colombia improved considerably across all three subjects. In Costa Rica, learning outcomes declined over the past two decades, particularly in reading and science – a similar trend was recorded in the Dominican Republic for reading and to a lesser extent in Uruguay for mathematics.11

Figure 3.12 shows the share of top-performing and low-achieving 15-year-old students in the LAC region.12 One striking finding is that, in 7 out of the 10 Latin American countries participating in PISA, fewer than 1% of students perform at the highest levels of proficiency (Level 5 or above) in mathematics, reading and science.13 In Chile, where the share of students attaining Level 5 or above in all three subjects is the highest, this figure was only 3% in reading, 1% in mathematics and 1% in science (compared to 9%, 11% and 7% in the OECD average, respectively) (Figure 3.12, Panel B). In PISA, Level 2 is considered the baseline level of skills required for productive participation in society.14 Yet within the focal group of countries, on average, 50% of students failed to reach Level 2 in reading, 64% in mathematics and 53% in science (Figure 3.12, Panel A). In the Dominican Republic, at least 8 out of 10 students achieved results below Level 2 across all three subjects. This presents a major challenge for Latin American countries that are transitioning into knowledge-based economies, where people need to innovate, adapt and leverage advanced human capital (OECD/CAF/UN ECLAC, 2016[56]).

Deep socio-economic inequalities stand out when comparing students’ proficiency by socio-economic status. PISA data can be disaggregated based on the Index of Economic, Social and Cultural Status (ESCS) – with the top quarter of ESCS scores representing the most advantaged students, and the bottom quarter the least advantaged. This index is a composite score built by the indicators parental education, highest parental occupation, and home possessions including books in the home (OECD, 2017[57]). Differences in students’ achievement are particularly pronounced when considering the highest levels of proficiency (Figure 3.13). In 8 out of 9 countries of the focal group with available data, fewer than 0.5% of disadvantaged students were top performers in reading, with the exception of only Chile, which nevertheless remained six times lower than the OECD average (Figure 3.13, Panel B). On average for the focal group of countries, the share of students reaching Level 5 in reading was over 30 times higher for the most advantaged students compared to the least affluent ones, whereas in the OECD the share was six times higher. Similarly, the share of low achievers amongst disadvantaged students was more than twice as high as among advantaged students, on average in focal countries (68% against 28% respectively), whereas in the OECD it was more than three times higher (36% against 11%) (Figure 3.13, Panel A).

Schooling is just one element in individuals’ development of cognitive skills (Hanushek, 2015[59]). Across and within countries, individuals having attained similar levels of education have different levels of skills once they reach adult age. What’s more, acquiring skills does not depend only on having obtained a certain certificate or diploma but also on other factors, such as the quality of educational systems, socio-economic contexts, networks, families and various life experiences (OECD/CAF/UN ECLAC, 2016[56]). The availability of cognitive skill measures allows to draw a clearer picture of what adults have learned to do throughout their schooling years in Latin America.

There are two different ways of assessing cognitive skills among adults. The first is through the literacy rate, defined as the percentage of people aged 15 or above who can both read and write a short simple statement about their everyday life.15 It is measured by national census and household surveys and is generally considered as an outcome indicator to evaluate educational attainment. It is also used as a proxy to evaluate the effectiveness of education systems: a high literacy rate suggests that the education system has provided a large share of the population with basic literacy skills (World Bank, 2020[60]).

Based on this measure, close to 95% of the adult population in focus group countries was literate, a slightly higher share than in the Latin American region as a whole, on average (Figure 3.14, Panel A). The literacy rate reached 99% in Argentina in 2018, and 93% in Ecuador (in 2017). Across all countries, the trend since 2000 has been mostly positive. In Brazil, Costa Rica, the Dominican Republic, Mexico and Peru, the literacy rate increased by 3 percentage points or more, enabling them to catch up with other focal group countries.

The Survey of Adult Skills defines numeracy as the ability to access, use, interpret and communicate mathematical information and ideas in order to engage in and manage the mathematical demands of a range of situations in adult life (OECD, 2019[62]).16 Figure 3.15, Panel A shows the percentage of adults who scored at each of the six proficiency levels on the numeracy scale in the four focal group countries with available data. In Mexico, the share of adults scoring below Level 1 (23%) is over 3 times higher than in the OECD average (7%), while in Ecuador and Peru this share is at least 6 times higher (at 42% and 46%, respectively). In Chile, although the share of the population scoring below Level 1 is also high (31%), the share of adults reaching levels 3 (10%) and 4/5 (2%) is above that of the three other Latin American countries participating in PIAAC. Nonetheless, when compared to the OECD average (31% for Level 3, and 11% for Level 4/5) these levels remain relatively low.

Today, the capacity to solve problems in technology-rich environments – i.e. to access, evaluate, analyse and communicate information – is crucial. Information and communication technology (ICT) applications have become a common feature in most workplaces, but also in education and everyday life (OECD, 2013[63]). In the Survey of Adult Skills, the scale of problem-solving in technology-rich environments is divided into four levels of proficiency (Levels 1 to 3, as well as below Level 1). Across participating OECD countries, roughly one-third of adults (30%) are proficient at the two highest levels (Level 2 or Level 3), demonstrating the capacity to use both generic and more specific technology applications. However, only one in ten adults or less managed to achieve these levels in Ecuador (5%), Peru (7%) and Mexico (10%), compared to 15% in Chile (Figure 3.15, Panel B).

Although relatively few adults in the focal countries perform at Level 1 or below for problem-solving in technology-rich environments, many are unable to display any proficiency at all. In all the countries participating in the PIAAC assessment, a considerable proportion of adults were unable to display their abilities in problem-solving in technology-rich environments, since they took the assessment in the paper-based format (OECD, 2016[47]). Among the countries of the focal group with available data, particularly large shares of adults opted out of the computer-based assessment in Ecuador and Mexico (approximately 18%), compared to Peru (11%). Furthermore, Ecuador (33%), Mexico (39%) and Peru (44%) stand out as countries where a very large proportion of the adult population have no prior computer experience or very poor ICT skills, particularly compared to the OECD average (16%) (Figure 3.15, Panel B). This means that they failed the “ICT core” test and thus did not have the basic computer skills needed for the computer-based assessment. As a result, smaller shares of adults may be scoring at Level 1 and below in countries such as Peru and Mexico, because these countries registered larger proportions of adults who were unable to display sufficient proficiency in problem-solving to have scored at even the lowest levels (OECD, 2019[62]; OECD, forthcoming[36]).

While people learn in a variety of settings, the educational system is the main vehicle through which communities attend to the learning needs of their students. Both the public and private sectors have invested significant amounts of resources in the educational system, with various features of this system (ranging from costs to location, accessibility and quality of teaching) shaping people’s satisfaction with the services delivered.

The Gallup World Poll collects data on the share of people who are satisfied with the education system in the city or area where they live. This measure remained relatively stable between 2006-09 and 2017-19 in the focal group countries, at around 63% on average (Figure 3.16). However, this average hides significant cross-country differences, as well as diverging trends. For example, in 2017-19, satisfaction with educational services had increased by 5 percentage points or more in Ecuador, the Dominican Republic (by 5 percentage points), Argentina and Peru (7 points) since 2006-09, while dropping by more than 10 points in Uruguay (-11 points), Colombia and Chile (-13 points). The decrease in Chile meant that less than half of the population were satisfied with the educational system in 2017-19, widening the gap with countries such as Costa Rica, where approximately eight out of 10 people were satisfied in 2006-09 and 2017-2019 alike.

The COVID crisis will have a profoundly negative impact on education. According to data from the United Nations Educational, Scientific and Cultural Organization (UNESCO), by mid-May 2020, more than 160 million students at all levels of education had stopped having face-to-face classes in Latin America and the Caribbean. The total duration of school closures in the focal group of countries was generally over 41 weeks, with the exception of Uruguay where it was 31-40 weeks (UNESCO, 2021[64]). Early estimates suggest that, worldwide, COVID-19 could result in a loss of 0.6 years of schooling, adjusted for quality, which would bring down the effective years of basic schooling achieved by students from 7.9 to 7.3 years. For today’s cohort in primary and secondary education, this could also mean facing a reduction in yearly earnings of $872 at present value (World Bank, 2020[65]). Latin American universities face a challenging environment as well, with 84% of them expecting reduced enrolment, of which half expect declines by 10%-25% (Hershberg, Flinn-Palcic and Kambhu, 2020[66]). Being out of school and losing family livelihoods due to the pandemic may leave girls particularly vulnerable (due to an increased burden of care work and/or increased likelihood of teenage pregnancies linked to abuse), whilst exacerbating exclusion and deprivation, especially for persons with disabilities or members of other marginalised groups (World Bank, 2020[65]). Most countries in the focal group moved to use education technology in order to deliver remote-learning solutions, but many students and schools were not sufficiently prepared for the transition, thereby amplifying socio-economic gaps in education (Gropello, 2020[67]; OECD et al., 2020[20]). While online education can help alleviate the immediate impacts of school closures, only 34% of students in primary, 41% in secondary and 68% in tertiary education in Latin America have access to an Internet-connected computer at home. The transition to online study has excluded many students from poorer households: fewer than 14% of poor students (those living with less than USD 5.5 per day, PPP 2011) in primary education have a computer connected to the Internet at home, in contrast to over 80% of affluent students (i.e. those living with more than USD 70 per day) (Basto-Aguirre, Cerutti and Nieto-Parra, 2020[48]). Furthermore, technological tools are only as effective as their use: on average, 58% of 15-year-olds in the region attended schools whose principals considered that teachers had the necessary technical and pedagogical skills to integrate digital devices into the curricula (OECD et al., 2020[20]). Students from poorer socio-economic backgrounds therefore risk bearing particularly long-lasting consequences in terms of learning outcomes and job opportunities, because they lack the resources and support to transition to distance learning (both in school and at home) (Basto-Aguirre, Cerutti and Nieto-Parra, 2020[48]; OECD et al., 2020[20]). The experience of Chile further underlines how face-to-face learning is difficult to replace, despite efforts to facilitate distance learning during the pandemic: when considering effectiveness and coverage indicators, distance education in the country offset only between 30% and 12% of learning losses linked to school closings, and effectiveness decreased to 6% in public schools, affecting mostly disadvantaged students (Ministerio de Educación, Centro de Estudios, 2020[68]). Furthermore, beyond the impact on learning outcomes, students’ social relationships can be harmed due to isolation (Loades et al., 2020[69]), and many may also miss out on school meals, which in some cases are a lifeline (WFP, 2020[70]). Data from the Gallup World Poll show a clear drop in the share of people satisfied with the educational system in 2020, compared to 2019. The year-on-year drop of 11 percentage points left the average level among countries in the focal group at 52% in 2020 (Figure 3.17). Falls were limited in Uruguay, Costa Rica and Argentina (-6 percentage points.), while exceeding 10 points in six other countries: Brazil (-14 percentage points), the Dominican Republic (-15), Ecuador (-19), Mexico (-19), Paraguay (-22) and Peru (-31). As a result, in Ecuador, Paraguay, Mexico and Brazil, barely one in two people declare themselves satisfied with the educational system or the schools in the city or area where they live, and only one in four people in Peru (26%). On the other hand, this share increased by 6 percentage points in Uruguay, reaching 70% in 2020. Overall, in Latin America and the Caribbean, there is a lack of country-level comparable data on individual skills, with relatively little comparative evidence on literacy, numeracy, problem-solving and technical skills. Information is also lacking on what types of higher-level technical and professional skills businesses in the region require both now and in future (OECD/CAF/UN ECLAC, 2016[56]). Much like the rest of the world, in the context of the digital transformation most Latin Americans will need to be equipped with access to the Internet and ICT problem-solving skills – on top of solid reading, numeracy and general problem-solving skills – in order to be able to benefit from digital technologies in their daily life and in the workplace. Moreover, the increasing importance of digital skills means that inequalities in Internet access and ICT skills have the potential to worsen existing inequalities in well-being (OECD, 2019[71]). While access to the Internet is addressed in the previous chapter, metrics on ICT skills (drawn from international studies such as PIAAC) are currently available only for a small subset of countries in the region. An important priority for future statistical work is therefore to assess additional aspects of people’s knowledge and skills, once the measurement of the core “building blocks” (reading, mathematics, sciences and digital skills) has been consolidated. For instance, non-cognitive abilities, such as social and emotional skills – including resourcefulness, perseverance, adaptability and team-working – can also be considered as essential competencies. The OECD Study on Social and Emotional Skills (SSES), which aims to capture non-cognitive abilities in childhood and adolescence, shows that valid, reliable, comparable information on social and emotional skills can be produced across diverse populations and settings. Bogota and Manizales (Colombia) are among the ten cities for which data should soon be available (2023) (OECD, 2020[72]). A new module on socio-emotional skills has also been included in the latest round of PIAAC (2018), aiming to provide insights on individual attributes, behaviours and beliefs such as conscientiousness, open-mindedness and relationships with others (OECD, 2021[73]). An ideal set of indicators for knowledge and skills would also address the challenge that drop-out rates represent in terms of school performance, both at primary and secondary levels, in Latin America. Generally, the “road to disengagement” from school begins during childhood, either at home or at school (Lessard et al., 2008[74]). For those who complete their primary education, students often have the illusion of faring relatively well. However, once they enter lower secondary school, they may experience learning difficulties, and the rigour expected of them can lead to disengagement, impeding the learning process (Bautier, 2003[75]; Blaya, Catherine; Hayden, Carol, 2003[76]). Young adults who have left secondary school without attaining a formal qualification are at high risk of poor employment, suffer worse health conditions and are over-represented among those committing crimes (Belfield and Levin, 2007[77]; Lochner, 2011[78]; Machin, Marie and Vujić, 2011[79]). The cumulative drop-out rates to the last grade of primary and to the last grade of lower secondary were available on the UNESCO Institute for Statistics (UIS) database for a majority of focal group countries up until September 2020, when the indicators were discontinued, in order to produce a smaller number of core indicators based on the SDGs.17 Other indicators produced by the UIS may help in capturing certain elements such as the completion rates (primary education, lower secondary education, upper secondary education) or the survival rate to the last grade of primary education, from which the cumulative drop-out rate to last grade of primary education can be derived.18 Personal security or freedom from harm is a key component of people’s well-being. The range of threats to people’s lives is vast, from political and ethnic conflicts to environmental hazards, industrial accidents and terrorism. However, one of the more common threats to personal security in emerging and developed countries alike is crime. This includes a large number of criminal offences, such as crimes against property (e.g. car theft, burglary in one’s own home), contact crimes (e.g. assault, mugging), non-conventional crimes (e.g. consumer fraud, corruption) and murders. However, according to the International Classification of Crime for Statistical Purposes (ICCS): “the vast disparity in approaches and sources used in the establishment of criminal laws by different countries makes it impossible to create a consistent and comprehensive definition of crime” (UNODC, 2016[80]). Therefore, the concept must be delimited for the sake of cross-country comparison and analysis. Addressing high levels of criminal violence is a top priority for many countries in Latin America and the Caribbean. According to the most recent Latinobarometro survey (2018), crime and public security were the top concern for 21% of citizens in the focal group of countries – more so than unemployment, the economy or corruption (Latinobarometro, 2020[81]). The homicide rate for the focal countries (13 per 100 000 population) is six times higher than the OECD average (3 per 100 000), and the share of people who feel safe when walking alone at night (44%) is very low compared with the OECD average (72%). Among the countries with the highest homicides rates worldwide, 17 of the top 20, and 40 of the top 50, are in Latin America (Muggah, 2018[82]). Although the data show some progress over the past decade, much remains to be done to meet people’s expectations and international commitments. Overall, both objective and subjective measures of safety in this report point to very high levels of insecurity that have not always improved for all countries. High urbanisation in the region also contributes to some of these trends, as crime rates tend to be higher in urban and peri-urban areas (Muggah and Szabó, 2016[83]). Cartel and gang-related violence is a significant contributor to high violence rates in Latin America, although the manifestations and drivers of gang activity differ from country to country, and comparable evidence on the subject is scarce (Dammert, 2017[84]). One clear, yet indirect, measure of the extent of gang violence is the greater incidence of homicides among young men, as the vast majority of participants in and victims of gang violence tend to be adolescent and young adult males (Chioda, 2017[85]). The impact of this violence is felt across society, however, not only through the loss of life experienced by affected families and communities, but also through a heightened awareness and fear of violent crime. For example, one in three people in Mexico and Argentina, and one in ten in Chile, report being frequently aware of shootings in their area of residence (UNODC, 2020[86]). Threats to safety can also come from within the home or family, particularly for women and children. In almost every Latin American and Caribbean country with nationally representative survey data, more than 40% of children experienced violence in the past month, and this is usually higher for boys (Lenzer, 2017[87]). In contrast, women and girls are much more likely to experience physical, sexual or psychological abuse (OECD, forthcoming[36]). The World Health Organization (WHO) estimates that 30% of women in the Americas have experienced physical and/or sexual violence by a partner, while 11% have experienced sexual violence by a non-partner (WHO, 2013[88]). Further data on this issue are presented in Chapter 5 of this report. The COVID-19 pandemic changed the nature of crime risks that people face on a daily basis, whilst further increasing the economic hardship that contributes to high crime rates in the region (Crisis Group Latin America, 2020[89]; UN, 2020[21]; UNODC, 2020[90]). Although the extended lockdowns made some types of crime less likely (e.g. property crime), early evidence from focal group countries shows that in certain regions violence continued as usual or even increased. Restrictive measures taken by governments to contain the coronavirus also provided criminal organisations with a window of opportunity to solidify their power, competing with governments to gain the support of local populations by providing essential services to hard-to-reach groups (Asmann, 2020[91]; Felbab-Brown, 2020[92]; Rivard Piché, 2020[93]). In the context of lockdowns due to the pandemic, criminality online has soared (Austin, 2020[94]), along with the risk of domestic violence and abuse (Statista, 2020[95]). The homicide rate in the focal group of countries (14 per 100 000 population) is almost five times higher than in the OECD on average (3 per 100 000 population), yet lower than the wider regional average (22 per 100 000 population) (Figure 3.18, Panel A). In a majority of the focal group, the homicide rate is 10 per 100 000 population or below, yet it is more than twice as high in Mexico (29), Brazil (27) and Colombia (25), while it is much lower, and not far from the OECD average, in Chile (4 per 100 000 population). Although the homicide rate has fallen by four points on average in the focal group of countries since 2000, trends widely differ across countries, with a drastic decrease in Colombia (-42 points), considerable falls in Paraguay (-12 points) and Ecuador (-9 points), and substantial rises in Mexico (+18 points), Peru (+8 points) and Uruguay (+6 points). Homicides represent only a fraction of the security risks faced by people, and self-reported victimisation rates indicate the prevalence of other criminal threats to safety. One measure of victimisation, drawn from the Latinobarometro survey, considers the share of individuals who answered “yes” to the question: “Have you (or a member of your family) been assaulted, attacked or a victim of crime in the previous 12 months?”. Based on this measure (Figure 3.18, Panel B), Mexico, Chile and Colombia are among the countries with higher victimisation rates in the region, ranging from 33% in Mexico to 19% in Paraguay. The victimisation rate in the focal countries has fallen from 43% in 2001 to 25% in 2018 on average, with high volatility in most countries. Self-reported victimisation was already low in the Dominican Republic in 2004, and since then it has decreased by 3 percentage points. Conversely, it is highest in Mexico (closely followed by Chile), where it has however more than halved in the space of seventeen years. Detailed and comparable data on specific types of crime are not systematically available in the region, although some national victimisation surveys provide insightful information. The most common form of violence in focal group countries where data are available is robbery, affecting almost one in ten people in Peru (9.4%) and Mexico (8.4%). Physical violence linked to injuries is most common in Argentina (2.3%), which also features the highest levels of psychological (4%) and sexual violence (1.7%) (Figure 3.19). In addition to the risk of crime and violence, people’s perceptions about their own safety may have large impacts on people’s well-being through increased concern and anxiety (OECD, 2015[96]). In 2017-19, the share of people who said they felt safe when walking alone at night among the focal group of countries (44% on average) was relatively low when compared to the OECD (72%) (Figure 3.20). Prior to the COVID-19 pandemic, the latest data available showed that Ecuador and Paraguay were the only countries of the focal group where half of the population declared to feel safe walking alone at night (50%), 15 percentage points higher than in the lowest-performing country, Brazil (35%). The share of people feeling safe when walking alone at night remained roughly stable between 2006-09 and 2017-19, but declined in focal group countries where the situation was already ominous. Trends over time vary across countries: considerable declines were registered in Mexico (-11 percentage points), the Dominican Republic (-6 points), Brazil and Colombia (-5 points), while a clear improvement was recorded in Ecuador (+9 points), Paraguay (+6 points) and Chile (+5 points). Levels remained relatively stable in Argentina, Uruguay, Peru and Costa Rica. Fear of crime can negatively impact people’s well-being by affecting their behaviour and their perceived freedom to do the things that they value doing. At 55%, the share of people in the focal group who reported that crime was the greatest threat to their personal safety was twice that of the OECD average (22%) in 2019 (Gallup World Poll, 2021[97]). Rates range from 39% of respondents in Chile to 68% in Brazil.19 Figure 3.21 shows some of the behaviours that have been constrained by fear of crime in Argentina, Mexico and Peru (the only three focal countries for which data are available). In Argentina and Mexico, a majority of people have stopped allowing their children to go out alone and stopped carrying cash. In all three countries, a large share of people have stopped going out at night completely, with this share ranging from 39% in Peru to 53% in Mexico. Every year, roughly 1.35 million people die from road traffic crashes around the world, with over half of them affecting pedestrians, cyclists and motorcyclists, according to the World Health Organization. Mortality rates tend to be higher in low and middle-income countries than in higher-income countries (WHO, 2020[98]). In 2019, the focal group average (18 deaths per 100 000 population) was twice as high as the OECD average (9 deaths) (Figure 3.22). Ecuador was the worst-performing country, with 27 deaths per 100 000 population due to road traffic injuries. The Dominican Republic, Paraguay and Brazil also registered over 20 deaths per 100 000 population. At the other end of the spectrum, Argentina (14), Peru (14) and Chile (13) had the lowest road death rates among the focal group, approximately half the rate recorded in Ecuador. Trends between 2000 and 2019 have diverged across countries, with most best-performing countries (e.g. Chile and Peru) suffering fewer road deaths than two decades ago, whereas a number of low-performing countries (e.g. Paraguay) suffered more. In Brazil, despite decreasing more than anywhere else in the focal group, deaths from road traffic crashes (at 21 per 100 000 population) remain above average. Extended lockdowns in Latin America and the Caribbean kept people off the streets, with mixed consequences on crime. On the one hand, they made some types of crime less likely (e.g. property crime). Mandatory confinement and strict social control across the region led to fewer opportunities for petty crimes such as mugging, whilst many criminals were dissuaded by the risks of infection (Semple and Ahmed, 2020[99]). In the first semester of 2020, 22% of households in Mexico fell victim to robbery, burglary or theft, compared to 35% a year earlier (2019) (INEGI, 2020[100]), with crimes committed outside of private dwellings falling from 17% to 9%. Furthermore, in January 2021, the adult population expressed a higher level of satisfaction with security than a year earlier, despite the level remaining relatively low (5.5 out of 10 in 2021, against 5.2 in 2020) (INEGI, 2021[101]). In Central America, the homicide rate per 100 000 population decreased by almost a third on average, from 31 to 21 – representing 2 607 fewer homicides (Infosegura, 2021[102]).20 However, early evidence shows that for other countries in the focal group violence continued as usual. For instance, rural communities in Colombia fell victim to armed conflict even during national lockdowns (El Espectador, 2020[103]), while the number of homicides remained stable in Mexico following the introduction of lockdown measures, with similar levels during the first semesters of 2019 and 2020 ( (Gobierno de Mexico, 2020[104]; UNODC, 2020[90]). COVID-19 also opened a window of opportunity for groups that partake in organised crime to solidify their power. In Brazil, Mexico and Colombia, cartels and armed groups engaged in charitable activities (e.g. by handing out basic food packages (Felbab-Brown, 2020[92])) during lockdowns in an attempt to expand their social base, and imposed their own restrictions on communities – separate from those instituted by national governments (Asmann, 2020[91]). By capitalising on their ability to enforce key measures at a local level, these groups can entrench themselves more deeply within communities, making it harder for governments to regain authority. Further consequences of the pandemic such as rising levels of poverty and of unemployment among youth may also provide these groups with an environment in which they can thrive, raising the appeal of illegal activities to vulnerable groups (Nugent, 2020[105]). Data from the Gallup World Poll on the share of people declaring that they felt safe when walking alone at night show little year-on-year change in 2020 compared to 2019 on average in the focal group (from 45% in 2019 to 46% in 2020) (Gallup World Poll, 2021[97]). Nonetheless, this hides diverging trends across countries, for instance a 7 percentage point drop in Chile (from 48% to 41%) and a 6 percentage point increase in the Dominican Republic (from 39% to 45%) and Uruguay (from 46% to 52%). Finally, with many services, shops and offices shut, as well as a significant share of the population in self-isolation, a higher number of people have relied on purchasing goods and services online. As a result, criminal organisations have turned to ransomware attacks, online scams and phishing e-mail schemes, which proliferated throughout Latin-American countries during the pandemic, posing dangers to people, but also to banks and governments (Austin, 2020[94]). Reports of domestic violence during the first weeks of quarantine showed an increase in four focal group countries (Argentina, Chile, Colombia and Mexico) (Statista, 2020[95]). Due to the isolation measures and income shortages entailed by the sanitary and economic crises, the situation is likely to heighten the risk of violence and abuse in Latin American homes. Early evidence suggests that call volumes to helplines in the region increased post-quarantine (López-Calva, 2020[106]): examples include an increase of 32% in helpline calls in Buenos Aires (Perez-Vincent et al., 2020[107]), following the introduction of restrictions on mobility, and an increase of 48% between April and July 2020 to helpline calls in Peru (Agüero, 2020[108]). Data from Línea Mujeres in Mexico City suggest there was little effect of the lockdown on calls regarding interpersonal violence, but an increase in calls for psychological services along with a fall in calls for legal services (Silverio-Murillo and Balmori de la Miyar, 2020[109]). The homicide rate is a key indicator of violent crime, but it represents only the “tip of the iceberg”. In this report, this has been complemented by the self-reported victimisation rate to provide a wider view of how crime affects individuals. More data are needed from both police registers and crime victimisation surveys to cover a wider range of experiences, as cross-country comparability of existing data is limited and no central repository of international data currently exists. Furthermore, crime victimisation surveys from the region show that few people report crimes to the competent authorities and that, when they do, most of them report a negative experience. In Peru, for example, the share of crimes reported to the police makes up only 13% of all crimes experienced, and dissatisfaction when reporting crime reaches 83% – the main reason being a lack of action by the authorities (UNODC, 2020[86]). Feelings of safety affect people’s well-being and behaviours. Nevertheless, available indicators sourced from the Gallup World Poll have a narrow scope (feelings of safety “when walking alone at night”). Moreover, there is no indication of the types of threats that people may fear, nor on the contextual predictors (such as social cohesion, incivilities or neighbourhood disruption, for instance), which limits the identification of potential policy levers. Given the extent to which the region is marked by violence, the generation of comparable insecurity statistics that include people’s perceptions is a priority for Latin America’s statistical agenda. The indicator used in this report is hence only a placeholder until better quality data become available. The scope of the road safety indicator used in this report could be improved by extending it to (non-fatal) road injuries. In some Latin American countries, however, the institutional capacity to monitor road injuries and crash data is still limited. Deaths from conflict are also missing from the data set used in this report. The evidence on the impacts of the COVID-19 pandemic highlights certain key areas where measures of people’s safety may be improved. Domestic violence is an important aspect of safety highlighted in the Sustainable Development Goals (Target 5.2.1 refers to women and girls subject to intimate partner violence). However, existing data often come from specialised surveys that are conducted infrequently and focus mainly on women (rather than on the entire population). These specialised surveys must also follow the required safety and ethical measures for this type of research: trained interviewers must collect data in a private space in a non-judgmental way, interviewing one person per household in the absence of their partner (WHO, 2013[88]). In Latin America, only five countries (which are also in the focal group for this report) have implemented surveys that come close to this standard: Chile, Costa Rica, Ecuador, Mexico and Uruguay (UNDP, 2017[110]). One alternative countries have used to collect information from representative population samples is via the inclusion of a module on domestic violence in an existing survey. A total of 12 countries in Latin America have relied on this approach – and Ecuador is the only one to also have a specialised survey, the “Encuesta Nacional Sobre Relaciones Familiares y Violencia de Género contra la Mujeres - ENVIGMU(UNDP, 2017[110]; INEC, 2019[111]). Finally, the ongoing digital transformation also implies new risks to people’s safety. As mentioned above, in the absence of effective regulatory, legal and ethical frameworks, both Internet users and organisations can be exposed to substantial economic, social, emotional and even physical risks. Measuring cybersecurity risks is challenging, however, as online criminal activity may go unnoticed by Internet users, and no centralised reporting mechanism for small-scale online security incidents currently exists in the region. While self-reports of cybercrimes are common, they have methodological limits (OECD, 2019[71]), implying that greater efforts are needed to develop a more general and objective measure of cybersecurity risks. The quality of the environment affects human health directly through the quality of air, water and soil, and through the presence, density and toxicity of hazardous substances. Environmental quality also matters intrinsically to people who attach importance to its beauty and value amenities that affect their life choices (e.g. a place to live) (Balestra and Dottori, 2011[112]). Furthermore, people benefit from environmental services and assets. In particular, access to green spaces has been associated with numerous benefits, including psychological relaxation, stress reduction, enhanced physical activity, mitigation of exposure to air pollution, excessive heat and noise, improved social capital and pro-environmental behaviours (WHO Regional Office for Europe, 2016[113]; Engemann et al., 2019[114]). Environmental quality depends on how natural resources and land are used, as human activities have the potential to pollute through by-products that end up on land or in rivers or lakes, the ocean and the atmosphere (ECLAC, 2010[115]). The countries of Latin America and the Caribbean are endowed with a rich base of natural resources (see Chapter 4), particularly minerals, oil deposits, forest area and arable land (Solbrig, 1998[116]). In addition, the Pacific and South Atlantic coasts are rich in seafood. However, the region also faces some of the most threatening environmental problems. Most cities confront huge air quality challenges as a result of urban growth, transport emissions and energy consumption. These factors, in addition to relatively inefficient vehicles, weak fuel standards and biomass burning for heating and cooking, further contribute to alarming PM2.5 levels (NRDC, 2014[117]; CAF, 2015[118]; IQ Air, 2019[119]). Other challenges include contaminated water due to industrial waste and soil erosion (ECLAC, 2010[115]; UNEP, 2018[120]), as well as deforestation (discussed in Chapter 4). Finally, although Latin America bears relatively little historical responsibility for greenhouse gas emissions, it is heavily exposed to some of the consequences, such as extreme weather events. Rising sea levels, for example, could have dramatic impacts on Caribbean islands over the next century, while high-intensity tropical cyclones are of great concern to Central American countries, and increasing temperatures are expected to exacerbate droughts in areas such as the northeast of Brazil (FIDA, 2020[121]). This section covers the key environmental aspects that impact people’s well-being based on available data for the region. Wider threats to the environment linked to natural capital, such as endangered species, water stress and greenhouse gas emissions, are covered in Chapter 4 on resources for future well-being. Early evidence suggests that the COVID-19 pandemic has improved outdoor air quality in many respects, which could prove to be beneficial for many individuals considered among the most vulnerable. However, these benefits are likely to be short-lived. As countries recover from the pandemic, resumptions in air travel, people’s movement within and between cities, and production levels in factories will see an increase in outdoor air pollution. Furthermore, inefficient waste treatment systems in the region are a cause for concern due to the additional hazardous waste generated during the outbreak. Air pollution is one of the main immediate environmental risks for people’s health in the Americas (WHO, 2016[122]). In Latin America and the Caribbean in particular, phenomena such as wildfires, the widespread use of wood for heating and/or cooking and the increasing number of vehicles (CAF, 2019[123]) are leading to people’s high exposure to both indoor and outdoor air pollution. The sources of air pollutants vary both within and across countries, as does the severity of people’s exposure.21 The potential costs include shorter life expectancy, increased health-care costs and reduced labour productivity. Further consequences include reduced agricultural output and damage to ecosystems (OECD, 2017[124]). Fine particulate matter (PM2.5) is a common air pollutant that is inhaled and may cause serious health disorders, including both cardiovascular and respiratory diseases (OECD, 2020[39]). In the focal group of countries, just over 90% of the population are exposed to levels of PM2.5 above the first WHO threshold level of risk to human health (10 micrograms per cubic metre) (WHO, 2006[125]) (Figure 3.23, Panel A). However, different thresholds of exposure can be used in order to assess air pollution at different levels of severity, revealing a nuanced picture of the situation in each country. For instance, in Ecuador, Mexico and Costa Rica, the average share of the population exposed to levels of fine particulate matter air pollution above 15 micrograms/m3 exceeds 97%, yet less than 9% are exposed to levels above 25 micrograms/m3 (i.e. less than the averages in the focal group and OECD of 14% and 11%, respectively). By contrast, in Peru, Chile and Colombia, over 40% of the population are exposed to the highest threshold level – which affects less than 1% of the population in five of the other seven focal group countries. Although the average mean exposure to PM2.5 is less straightforward to interpret, it is a useful measure for assessing changes in air pollution over time as opposed to the share of the population exposed to certain thresholds, since the share of the population moving from one side of a threshold to another may distort the trend in overall exposure. Between 2000 and 2019, the average mean exposure to PM2.5 fell by 9% on average in the focal group of countries. The largest improvements occurred in Brazil, Paraguay, Mexico and Colombia, where levels fell by 20% or more. Conversely, the average mean exposure to PM2.5 increased slightly in Peru (11%) and Ecuador (6%) (Figure 3.23, Panel B). Air pollution is generally associated with urbanisation, industry and transport. However, the contribution of biomass burning from household cooking and/or agriculture to local air pollution is considerable (Brezzi and Sanchez-Serra, 2014[128]). Thus, exposure to air pollution, and its causes, vary greatly according to whether people live in cities or in rural areas, or in developed or developing countries. The following set of estimates are based on political and administrative boundaries established by territorial grid units and Global Administrative Unit Layers, respectively developed by the OECD and the FAO (OECD, 2020[129]; FAO, 2021[130]). According to 2019 estimates, in 90% of the regions in the selected countries, average annual exposure to air pollution was higher than the World Health Organization’s recommended maximum of 10 μg/m3 (Figure 3.24). Of the remaining 10% of regions, over half were in Uruguay – the only country of the focal group where total exposure is below the WHO threshold.22 Very high values of exposure to fine particulate matter are found in some regions of Peru, where 20 regions have an annual average exposure over 25 µg/m3, but also in Colombia (8) and Chile (6). In Figure 3.24, relatively large regional disparities can be observed in Chile, Peru, Colombia and Argentina (above 15 µg/m3), as opposed to countries such as Uruguay, Paraguay and Costa Rica (below 5 µg/m3). Aisén del General Carlos Ibañez del Campo, Chile’s most polluted region, is also the most polluted region in the focal group countries and the OECD, according to this measure. Chile’s southernmost region of Magallanes is the focal group’s least polluted region, with a mean population exposure to PM2.5 of 6 µg/m3, i.e. one-third of the focal group average (18 µg/m3).23 Latin America and the Caribbean is a region that is prone to natural hazards. Between 2000 and 2019, it was the second most impacted region in the world, with a total of 152 million people affected by 1 205 natural disasters, with floods being the most common (OCHA, 2019[131]). The region is exposed to a large variety of disasters: between 1990 and 2020, 1 412 disasters with natural hazards as a source were registered, 87% of which were climate-related (i.e. wet mass movements, storms, floods, fires and extreme temperature events) and 13% geophysical (dry mass movements, volcanic eruptions and earthquakes).24 Floods were recorded most frequently, affecting roughly 49 million people. Despite being less frequent, droughts affected approximately 70 million people (ECLAC, 2021[132]).25 Climate change has been shown to be worsening a number of climate-related disasters in the region (OECD, 2019[133]). As seen in the Housing section of the previous Chapter, Latin America is one of the most urbanised regions on the planet, and its metropolitan areas are expected to face a higher level of risk in the years to come (Fisher and Gamper, 2017[134]). The region’s cities are also among the most unequal worldwide, home to an increased concentration of poor and hence vulnerable people, who are potentially exposed to natural hazards (Hardoy and Pandiella, 2009[135]; Fisher and Gamper, 2017[134]). A large share of the urban expansion that has taken place in Latin America in recent decades has been up mountain slopes, flood-plains and other areas prone to sea surges or seasonal storms (Warn and Adamo, 2014[136]). Examples includes cities such as Quito, in Ecuador (built on steep slopes at the foot of the Pichincha volcano) and Santa Fe, Argentina (expansion onto the Río Salado floodplain) (Hardoy and Pandiella, 2009[135]). Within agglomerations, many of the most affected neighbourhoods are inhabited by low-income groups in informal settlements that lack access to services and infrastructure (OECD, 2019[133]). In 2017-18, the number of deaths, missing persons and directly affected persons attributed to disasters was slightly lower than in 2005-06, though there was a sharp peak during 2009-10 (Figure 3.25, Panel A). The peak may be explained by repeated “El Nino-Southern Oscillation” (ENSO) phenomena in 2006-07 and 2009-2010 (Cai et al., 2020[137]), coupled with strong earthquakes in Haiti and Chile in 2010. In a context where the data are so volatile and have frequent spikes, looking at point changes for individual years may be misleading, which is why data have been pooled across years in Figure 3.25, Panel B. Over the 2012-2018 period, the count of people who died, went missing or who were directly affected by disasters stood at 534 persons per 100 000 population in focal countries, close to the regional average (524) (Figure 3.25, Panel B). However, the count is over 10 times higher in Paraguay (1222) than Chile (112). That being said, Chile is among the three countries of the focal group to have registered the largest fall (along with Colombia and Mexico) when comparing 2012-2018 data to 2005-11 data. This may be attributed to spikes due to certain events over the period, such as the 2010 earthquake in Chile, or floods in Colombia and Mexico (CERF, 2007[138]; IFRC, 2010[139]; IFRC, 2011[140]). Argentina is subject to intense thunderstorms, which bring severe weather including damaging winds and hail, torrential rains and lightning that can start wildfires. It is one of the countries in the focal group where the number of people who died, went missing or were directly affected by disasters increased between 2005-11 and 2012-2018, with particularly deadly storms registered in 2013 and 2015 (IFRC, 2013[141]; IFRC, 2013[141]; Penn State, 2020[142]). Nonetheless, evidence from this indicator must be interpreted with caution due to methodological differences among reporting systems: countries where change is more visible over time may simply report the data more accurately, rather than being more or less prepared for disasters of natural origin. Nearly 9 out of 10 Latin Americans in the focal group countries are exposed to a level of small particulate matter air pollution that puts their health at risk (Figure 3.24), and high levels of air pollution can be a risk factor for worse outcomes if infected with COVID-19 (Pozzer et al., 2020[145]; Wu et al., 2020[146]). Worldwide, pre-COVID outdoor air pollution caused more than 3 million premature deaths in 2010, with elderly people and children the most affected. OECD projections imply a doubling, or even tripling, of premature deaths from dirty air by 2060 (OECD, 2016[147]). Data show that ambient nitrogen dioxide and sulfur dioxide concentrations in Latin American cities decreased during the quarantines — mainly at the beginning — while PM2.5 levels show no clear overall trend before and during the period of restrictions (ECLAC, 2020[148]). Reductions in air pollution will provide temporary respite to people with respiratory problems or asthma, who are considered more susceptible to COVID-19, as well as reducing negative side-effects of pollution such as increased inflammation and lowered immunity (Glencross et al., 2020[149]). However, as countries begin to recover from the pandemic, resumptions in air travel, movements of people within and between cities, and production levels in factories will likely see an increase in outdoor air pollution (OECD, 2020[150]). Although ambient air pollution during COVID-19 lockdowns were fairly well documented (Amoatey et al., 2020[151]), studies on indoor air pollution were relatively scarce – particularly in Latin America and the Caribbean. However, this is a major issue in low and middle-income countries, and if people spend more time inside their homes the role of indoor air pollution takes on new significance as people suffer from a higher risk of exposure (Du and Wang, 2020[152]). According to evidence drawn from international research, important factors that can impact indoor air pollution include heating and/or cooking fuel and household fuel consumption (Shen et al., 2017[153]; Du et al., 2018[154]), cooking with oil (Zhao et al., 2019[155]), smoking (Kanchongkittiphon et al., 2015[156]), and the use of home ventilation or air conditioning (Zhang et al., 2011[157]; Liu et al., 2018[158]). Effective and environmentally sound waste management, an essential service, is particularly important in response to emergencies such as the COVID-19 pandemic. During the outbreak, various types of additional hazardous waste (such as medical waste) were generated, including gloves, masks and protective equipment. There are significant weaknesses in waste treatment facilities in the region, and unsound management of this type of waste could potentially lead to unforeseen “knock-on” effects on the environment, as well as on human health. Several measures have been identified as regional priorities for environmental policy during the recovery phase post-COVID-19, including the progressive closure of dumpsites, increasing the capacity of health-care waste treatment, strengthening the resilience of the waste sector, prioritising circular approaches and promoting institutional frameworks for sustainable waste management (UNEP, 2020[159]). The protection of the region’s biodiversity will be vital in the recovery process from COVID-19. As discussed in detail in Chapter 4 on resources for future well-being, Latin America is one of the most important regions of the world in terms of biodiversity and ecosystems. Biodiversity underpins current and future well-being as well as economic prosperity, and it is essential that it be a key part of the regional COVID-19 response and recovery plans (OECD, 2018[160]). Its protection is also vital in order to avoid the next pandemic: close to three-quarters of emerging infectious diseases in humans come from other animals. Wildlife exploitation and land-use change increase the risk of infectious disease, by bringing domestic animals and people into close proximity to wildlife that carry pathogens and by disrupting ecological processes that help keep diseases in check (OECD, 2020[161]). An ideal set of indicators of Environmental Quality in Latin America and the Caribbean would inform on people’s access to environmental services and amenities (OECD, 2020[39]), particularly with regards to water quality and recreational green space. The latter is even more relevant in the context of COVID-19: under confinement conditions, movement is restricted, and public spaces and parks may be closed. As mentioned in the section on Housing in Chapter 2 of this report, Latin America is one of the most urbanised regions in the word, and its cities are often afflicted by social and spatial segregation (Loret de Mola et al., 2017[162]). In the context of a pandemic, many Latin American urban families are confined to small, often inadequately built apartments. Access to basic services in these conditions is clearly a primary concern, but so is access to green space, as it provides numerous health and well-being benefits, including psychological relaxation; stress reduction; enhanced physical activity; mitigation of exposure to air pollution, excessive heat and noise; improved social capital; and pro-environmental behaviours (WHO Regional Office for Europe, 2016[113]; Engemann et al., 2019[114]). Although there is currently no universally accepted definition of green space,26 recent studies have helped to assess access to green areas in European cities using satellite data (Poelman, 2018[163]). The underlying method determines an area of easy walking distance – approximately 10 minutes’ walking time (at an average speed of 5 km per hour) – near an inhabited Urban Atlas polygon (Copernicus Land Monitoring Service, 2021[164]). Urban areas are defined as cities with an urban centre of at least 50 000 inhabitants (Dijkstra and Poelman, 2012[165]). The accuracy of estimates of air pollution exposure shown in this chapter varies considerably by location. Worldwide, accuracy is particularly poor in areas with few monitoring stations, and generally good in regions with dense networks of monitoring stations (such as most advanced economies) (Shaddick et al., 2018[166]). In addition, for some regions, particularly snow-covered areas, small islands and coastal areas, there are no PM2.5 concentration estimates for part of the region, because satellite-based measurements of aerosol optical depth are not reliable in areas where the dominant land cover is very reflective (Mackie, Haščič and Cárdenas Rodríguez, 2016[167]). Inequalities in exposure to air pollution, particularly by gender, age and education, are challenging to produce due to the nature of the data – which are collected at increasingly fine spatial levels, but not attributable to specific households or individuals (and therefore cannot be disaggregated by household and individual characteristics). In 2018, the OECD launched “The Geography of Well-Being”, a project aimed at building a comprehensive database of exposure to environmental risks disaggregated by socio-economic status, using metrics that are harmonised across countries and which can be considered a first step in this direction (OECD, 2020[39]). Civic engagement allows people to express their voice and to contribute to the political life of their society. Political voice is one of the basic freedoms and rights that people have reason to value (Sen, 1999[168]). People who have the opportunity to participate in a decision are more likely to endorse the decision and to consider it fair (Stutzer and Frey, 2006[169]). Civic engagement may also increase people’s sense of personal efficacy and control over their lives, (Barber, 1984[170]) and allows individuals to develop a sense of belonging to their community, trust in others and a feeling of social inclusion. With some exceptions, Latin America has made considerable progress in providing citizens with political voice, moving away from military dictatorships, human rights violations and internal conflicts over the past two decades. In fact, most Latin American people live in democracies today, and according to the recent assessment of the Economist Intelligence Unit, the democracies of Costa Rica and Uruguay are among the most robust in the world (EIU, 2020[171])). Nonetheless, dissatisfaction with the public sphere has been a source of social unrest in recent years, often linked to the State’s limited capacity to ensure its monopoly of violence and to run its institutions in accordance with the rule of law (ECLAC, 2021[172]) This dissatisfaction has the potential to hinder governance and the way democracies work: for instance, the share of the population having voiced their opinion to a public official has dropped from approximately one in five to one in six among countries in the focal group over the past decade. As described in the section on Income and Consumption in Chapter 2, limited progress in reducing inequality over the past decade has affected the way people perceive fairness in their societies as well as their trust in public institutions (ECLAC, 2013[42]; Busso and Messina, 2020[173]). This perceived lack of fairness and legitimacy in Latin American democracies contributes to the belief that economic and political elites enjoy privileges denied to most citizens, and that government institutions are the preserve of a few powerful groups that use them for their own benefit. Since 2019, several countries in the region experienced a wave of citizen protests and mobilisations, often led by youth demanding change, particularly with regards to long-standing structural inequalities and perceptions that governments are not responsive to the needs of all citizens. Meeting citizen expectations is as vital as ever, since the most vulnerable sections of society have been hit hardest by the crisis. Governments must prioritise effective, inclusive and non-discriminatory public participation in decision-making in order to guarantee institutional legitimacy, political voice and long-term stability. The most fundamental form of democratic engagement is participation in national elections. Voter turnout differs widely across focal group countries, partially reflecting differences in electoral systems, including the existence of compulsory voting.27 In recent years, voter turnout ranged from 47% in Chile (where voting is no longer compulsory since 2012) to 90% in Uruguay (where it is compulsory and enforced with sanctions) (Figure 3.26, Panel. A). On average, 7 out of 10 people who were registered to vote in the focal group of countries cast a ballot in the last election (70%), a share that has remained relatively stable over the past two decades. This stability masks gains of 6 to 7 percentage points in Argentina, Mexico and Colombia, and a 17-point increase in Ecuador. Although voter turnout in Ecuador was considerably higher at the height of the COVID-19 pandemic than it was 20 years ago, it was much lower in Peru (-8 percentage points) and the Dominican Republic (-21) (IDEA, 2021[174]).28 Voting is, however, only one aspect of political voice, and contacting public officials is also an important form of civic engagement (OECD, 2020[39]). In focal group countries, the share of people declaring to have voiced their opinion to a public official was three percentage points below that of OECD countries in 2017-19, at 16% on average (Figure 3.26, Panel. B). Over this period, shares ranged from 10% in Argentina to 22% in Colombia. Since 2006-09, the only country that recorded an increase was Paraguay (by 5 percentage points), lifting its share just above the focal group average. The share of people declaring to have voiced their opinion to a public official remained relatively stable in Ecuador and Uruguay, as well as in Argentina, where it was lowest among the focal group countries. Elsewhere in the focal group, the share of people who voiced their opinion to an official dropped considerably between 2006-09 and 2017-19, particularly in Colombia and Costa Rica, where it fell by 10 percentage points or more, yet remained relatively high. While all the 11 countries of the focal group are electoral democracies, their democratic experience is often relatively recent, and the political process is still perceived as being the preserve of powerful groups, with limited accountability for their decisions. When asked the question, “In general terms, would you say that your country is governed by a few powerful groups for their own benefit, or that it is governed for the good of all the people?”, four in five people (81%) in focal group countries answer the former, on average, with this share being close to 90% in Brazil. Chile, Costa Rica and Uruguay are the only countries where this proportion is 75% or lower (Figure 3.27). The share of the population who believe that their country is governed by a few powerful groups for their benefit increased in 8 focal group countries between 2004 and 2018, remaining relatively stable (at high levels) in the Dominican Republic and in Peru, while falling in Uruguay from 78% to 64%. Notable increases between these two years include Argentina (11 percentage points), Colombia (21 points) and Brazil (25 points) (Figure 3.27). The COVID-19 pandemic has disturbed electoral processes in a number of Latin American countries, with elections postponed in Chile, the Dominican Republic, Paraguay and Uruguay. Even when elections were maintained, there were considerable disruptions, ranging from changes in voter turnout to campaigning difficulties for candidates (Querido, 2020[175]). Moreover, each country adopted its own approach to the introduction of precautionary safety measures, which generally included social distancing, mask wearing, sanitising, temperature checks, and the single use of voting pencils (IDEA, 2020[176]). Certain countries also extended voting hours, increased the number of polling stations, offered mobile polling stations, or even made accommodations for advance voting, particularly for certain groups at risk (Asplund et al., 2021[177]; López-Calva, 2021[178]). While these special voting arrangements proved to be useful to mitigate the effects of the sanitary crisis on electoral calendars, they were not implemented systematically across Latin American countries. Instances in which this was particularly problematic include mandatory quarantine periods for voters who had recently returned from abroad or tested positive, and who were consequently disenfranchised from their voting right (Asplund et al., 2020[179]). Early evidence across 14 parliamentary and presidential elections suggests that the pandemic may have affected voting behaviour in the region (López-Calva, 2021[178]). When comparing the elections that took place during the pandemic to historical averages, voter turnout slightly increased in half of the countries, and decreased in the other half. However, when compared with previous elections, a majority of countries (11) registered a decrease in voter turnout and, whether compared to historical averages or to previous elections, these decreases were larger than the increases (López-Calva, 2021[178]). In due course, it will be important to take a closer a look at disaggregated data as well, to assess changes in voting behaviour across different groups of the population. Although there may be several, cross-cutting drivers behind these findings, a starting point for reflection is that trust in elections was already fragile prior to the pandemic (LAPOP, 2021[180]), in a context of increased social unrest. Although the 2009-13 period showed signs of greater optimism and confidence, there has been growing disenchantment and political polarisation in more recent years (ECLAC, 2021[172]). Countries that have reduced opportunities for public participation in decision-making should reverse this trend, noting the benefits of more inclusive governance, civic empowerment, and greater government legitimacy as a result. Civic space is considered a core component of any democratic, open society, and is guaranteed by the fundamental freedoms of association, assembly and expression (OECD, 2020[181]; CIVICUS, 2021[182]). Across the region, certain emergency responses to contain the pandemic sometimes led to restricted freedoms and liberties (OECD, 2020[181]; ICNL, 2021[183]). It is key for these measures to have sunset clauses (i.e. they must be time-bound) and to be strictly proportionate, in order to protect civic space and allow public engagement to resume in due course. Evidence suggests that there is a positive correlation between the protection of civic space and a country’s levels of economic and human development (BTEAM, 2021[184]). Examples of potential threats to civic space during the pandemic in Latin America include citizen’s reduced capacity to collectively voice their opinion on government responses - highlighted by reports of the overuse of force - as well as restrictive COVID-19-related disinformation laws on the freedom of expression (OECD, 2020[181]; CIVICUS, 2021[185]; ICNL, 2021[183]). Finally, 2020 data from the Gallup World Poll on the share of people having voiced their opinion to an official show relatively little year-on-year change compared to 2019 in the focal group, on average (17% in both years). It increased by 5 percentage points to reach 21% in Brazil, but declined by the same amount in Colombia and Costa Rica to 19% and 16%, respectively (Gallup World Poll, 2021[97]). An ideal set of civic engagement indicators would measure whether citizens are involved in a range of important civic and political activities that enable them to shape the society where they live. In well-functioning democracies, civic engagement shapes the institutions that govern people’s lives. The quality of these institutions per se is considered in the section on Social Capital of Chapter 4. Voting is the most traditional form of political voice. Much like voicing one’s opinion to a public official, further methods of civic expression are important, such as signing a petition, attending a political meeting or a demonstration, and participating in campaigns and protest via social media (Boarini and Díaz, 2015[186]). Guidance to statistical offices on how to measure political participation, as well as other aspects of governance, is provided by the 2020 Praia Group Handbook on Governance Statistics (UN, 2020[187]), but comparable official data in this field are still in their infancy. Comparable measures of these forms of participation are available for European countries (via the European Quality of Life Survey), and similar measures for Latin America and the Caribbean moving forward would be highly relevant – particularly in light of the social unrest in 2019. Analysis based on 30 European countries shows that people’s attitudes towards their ability to influence and engage in political life – or their “political efficacy” – affect their political behaviour, including different forms of participation (Prats and Meunier, 2021[188]). Additional data used in OECD countries for this area of study include survey data on “having a say in what the government does”. The indicator used in the flagship publication How’s Life? (OECD, 2020[39]) is sourced from PIAAC, which is run only every 10 years and whose main waves were last conducted by the OECD in 2012. The European Social Survey (ESS), conducted every two years, includes a similar question (“How much would you say the political system in [country] allows people like you to have a say in what the government does?”), but covers only European countries. In future rounds, PIAAC will also use a similar question wording to increase comparability. As of now, the measure of having a say in government included in How’s Life? refers only to a belief in the (external) responsiveness of public institutions and government officials to citizens’ demands, while excluding (internal) feelings of having the personal competence to participate in politics (Hoskins, Janmaat and Melis, 2017[189]), while the OECD Government at a Glance report includes also a measure of internal political efficacy for European countries (OECD, 2019[190]). In the 2019 revision of the Inter-Agency and Expert Group list of Sustainable Development indicators, both internal and external aspects were added under Goal 16 (Peace, Justice and Strong Institutions) (OECD, 2020[39]; UN, 2020[191]), and analysis of the accuracy and validity of the available measures of political efficacy showed that these indicators could be expanded to other regions outside Europe (González, 2020[192]). Social connections are essential for people’s well-being. Beyond the intrinsic pleasure that people derive from spending time with others, those with extensive and supportive networks have better health, tend to live longer and are more likely to be employed. At the same time, the lack of social connections worsens individuals’ mental and physical health (Cacioppo, Hawkley and Thisted, 2010[193]). Research in the field of social connections in Latin America stresses the relevance of friendship for people in their efforts to overcome poverty (Garcia et al., 2016[194]). More specifically, social connections play an important part in the survival strategies of vulnerable households for poverty alleviation. The sense of community and “togetherness” in Latin American societies is illustrated by the high value given to family and close friends, and their influence on individuals’ decisions in life (Husted, 2002[195]). Likewise, informal networks are often a vector for the transfer of resources amongst friends and family members (Uthoff and Beccaria, 2007[196]). In terms of economic behaviour, authors also mention the preference of Latin Americans for establishing friendship before engaging in business transactions (Ogliastri, 1997[197]). In other areas of life such as health, social support from friends has a positive influence on the experience of caring for chronic illness, by way of informational, material, emotional and affective support (Vega Angarita and González Escobar, 2009[198]). Findings in this section suggest that social network support in the focal group of countries is relatively high and close to the OECD on average, with little change over time. During the COVID-19 pandemic, many Latin Americans endured extended lockdowns and confinement restrictions, impacting their ability to maintain social relationships beyond immediate household members. Although online technologies can be harnessed to provide social support and a sense of belonging in the context of a pandemic, disparities in access to or literacy in digital resources remain a major concern in the region. The share of people reporting that they have relatives or friends whom they can count on to help them in times of need saw little change in the focal group of countries between 2006-09 and 2017-19, much like for the OECD average. At 87% in 2017-19, this remained below the OECD average by 3 percentage points and was similar to the regional average of 85% (Figure 3.28). Social network support was highest in Uruguay (at 91%) but significantly lower in Peru, Ecuador and Mexico (at 83%). Broad stability in this measure of social support across countries in the focal group over this period hides diverging patterns at a national level. In Chile, social network support increased by 6 percentage points, more than in any other country of the focal group. On the other hand, in Mexico it declined by 4 percentage points (Figure 3.28). Despite these findings, other evidence suggests that, beyond support networks, other aspects of social connections are important for people’s well-being and are particularly strong in certain Latin American countries. According to Rojas (2019[199]), people find a sense of identity and purpose through “close and warm person-based interpersonal relationships” (quality), and they report positive emotions to others thanks to the number and frequency of their relations (quantity). Representative surveys fielded in 2018 in Colombia, Costa Rica, Mexico and to the white/Caucasian population of the United States also suggest that the quality of interpersonal relations is higher in Latin American countries than in the United States (Rojas, 2019[199]). 65% of respondents in the Latin American participating countries agreed with the statement, “In this society interpersonal relations are warm and close”, against only 38% among white/Caucasians in the United States. When focusing on specific types of personal relations such as extended family, 62% of Latin Americans report visiting their grandparents frequently or very frequently during their childhood, compared to only 42% among white/Caucasians in the United States (Rojas, 2019[199]). Moreover, the quality of people’s social relations is linked to their perceptions of loneliness, a pattern that holds regardless of people’s age (OECD, 2019[200]). Loneliness and isolation are related to a number of factors, including lower levels of daily activity and mobility, higher depression and risk of death (OECD, 2019[200]). Although comparable official data on these issues are lacking in Latin America, ad-hoc studies have allowed to assess perceived loneliness among certain age groups in the region. The Global School-based Student Health Survey (GSHS), for instance, found that approximately one in six students in Latin America and the Caribbean reported being lonely most or all of the time and/or having no close friends (Sauter, Kim and Jacobsen, 2019[201]) – despite the fact that a relatively large share of Latin Americans tend to live with their parents compared to Western European and Anglo-Saxon countries (Helliwell, Layard and & Sachs, 2018[202]). The prevalence of loneliness among older adults (aged 65 or above) varies between 25% and 32% in Latin America, and is significantly higher for women, widows, less educated people and those with fewer household assets (Gao et al., 2020[203]). According to (Gerst-Emerson and Jayawardhana, 2015[204]), social isolation among older adults is a “serious public health concern” due to their heightened risk of cardiovascular, autoimmune, neurocognitive and mental health problems. Digitalisation is already having an impact on the way in which people interact with one another. The frequency of interactions via social media has risen and is likely to continue to do so as access to social interaction technologies increases. These technologies foster a wider network with weaker ties, rather than smaller networks with stronger ties (OECD, 2019[71]). Only few time-use surveys ask respondents to report on the use of information technology (OECD, 2020[39]). However, there is evidence that social media usage in Latin America is higher than in any other world region (in Q2 of 2019, 100% of people aged 16-64 had used or visited a social media network in the past month (Global Web Index, 2019[205]), with 54% reporting “staying in touch with what friends are doing” as the main reason for using social media, and 66% declared they follow people they know in real life – more than they do brands (51%), singers, musicians and bands (49%) (Global Web Index, 2019[205]). All told, these findings imply that much greater efforts are needed to develop high-quality, nationally representative and comparable data on social connections and the various facets of social support available to people today. Official measures on these issues are lacking not just in the Latin America and Caribbean region, but also across OECD countries (see discussion below). During the first wave of the COVID-19 pandemic, Latin Americans endured some of the longest lockdowns in the world (Parkin, Phillips and Agren, 2020[206]). They were also subject to some of the strictest mobility and contact restrictions in the spring of 2020, when approximately 85% of individuals in the region were distancing themselves from friends and relatives (Hale et al., 2021[207]; Alicea-Planas, Trudeau and Vásquez Mazariegos, 2021[208]). Both voluntary social distancing and mandatory lockdown policies have implications for people’s ability to maintain social relationships beyond immediate household members – whether for instrumental or emotional support, or simply for companionship (OECD, 2020[209]). Figure 3.29 shows that a majority of people in 2020 felt they have people to count on in time of need, with the focal group average standing at 83%, ranging from 74% in Peru to 92% in Uruguay. Nevertheless, this share decreases considerably (by 4 percentage points) relative to 2019, with strong declines in Mexico and Costa Rica (-8 percentage points), as well as in Brazil, Colombia, the Dominican Republic and Peru (-7 points), (Figure 3.29).29 Overall, fewer individuals were at risk of being confined alone in Latin America than in Europe or North America (Esteve et al., 2020[210]). In Colombia, data on people’s perceptions and expectations throughout the pandemic collected through the “Social Pulse Survey” conducted by the National Statistics Office (NSO) shows that over two-thirds (68%) of people in the country’s 23 main cities had spoken to family or friends to feel better over the previous 7 days during September 2020 and February 2021, ranging from 38% in Cúcuta to 97% in Quibdó (DANE, 2021[211]). Findings from the latest wave of this survey in February 2021 also show that feelings of loneliness were higher among women (12%) than men (9%). As noted above, while online technologies could be harnessed in order to provide social support and a sense of belonging (Newman and Zainal, 2020[212]), disparities in access to or literacy in digital resources remain a major concern in Latin America. The Internet usage gap between the richest and poorest across Latin America is almost 40 percentage points, and that between urban and rural households is above 25 percentage points (OECD et al., 2020[20]). Overcoming such digital divides will be critical to reduce the isolation and loneliness that many people in vulnerable groups experience. Social isolation and loneliness imply high risks for both physical and mental health and need to be addressed through interventions rooted in communities, civil society, social services and volunteering (House, Landis and Umberson, 1988[213]; Holt-Lunstad, Smith and Layton, 2010[214]; Pantell et al., 2013[215]; Klinenberg, 2016[216]; Sauter, Kim and Jacobsen, 2019[201]). Anecdotal reporting suggests that the pandemic may have prompted more solidarity worldwide (World Economic Forum, 2020[217]) and that confinements in Latin America generated a high social mobilisation in the digital space (Duque Franco et al., 2020[218]).30 A large psychological literature has documented the important direct and buffering roles that social support may play during times of stress (Cohen and Wills, 1985[219]; Cohen, 2005[220]; Cohen et al., 2014[221]; Bowen et al., 2014[222]). In the face of extended social distancing measures, it is key to sustain social connectedness and solidarity, particularly whilst enduring prolonged lockdown measures as well (OECD, 2020[209]). The measure of social support included in this report is limited: as a simple “yes/no” question, it provides no information about the frequency, intensity, quality or type (e.g. financial or emotional) of the support received. Moreover, it is not possible to assess gaps in support between the top and the bottom of the distribution from a simple “yes/no” question. Finally, the small sample sizes of the Gallup World Poll raise issues regarding measurement errors, especially when exploring change over time. An extensive psychological literature dating back several decades exists on social support measurement, and National Statistical Offices are taking increasing interest in such measures. However, beyond Europe, there is currently little consistency across NSO practices in collecting these measures (Fleischer, Smith and Viac, 2016[223]). As a dimension of the well-being framework used in this report, social support is currently undermeasured, and as a result it is rarely present in policy discussions, meaning that further research is needed. Advances in Latin America on this front have been made by the Colombian NSO, which developed a Social Capital module as part of its Political Culture Survey (Encuesta de cultura política, ECP). This module allows evaluating various areas of social capital, such as the importance of family ties or being able to count on a close network of social support (DANE, 2020[224]). An ideal indicator set for social connections would also provide information about the quantity of social interactions, both face to face (e.g. frequency and amount of time individuals spend with household members, their family, friends, colleagues and other acquaintances) and via social networks.31 Time-use surveys are fairly widely employed in the LAC region, with 19 countries having implemented some form of time-use survey by 2019 (ECLAC, 2019[225]). However, despite the existence of a harmonised Classification of Time-Use Activities for Latin America and the Caribbean (CAUTAL) (ECLAC/ INEGI/I NMUJERES/ UN-Women, 2016[226]), this system is not yet universally applied across countries. More regular and harmonised data collection on time use would increase the potential for better statistics on social activities. The quality of social connections (e.g. satisfaction with social interactions, perceived loneliness) is also relevant, as discussed above. However, survey questions on satisfaction with personal relationships are rare and infrequent. An example of the type of indicator that could be developed is the “Satisfaction with personal relationships” included in the OECD publication How’s Life? 2020, which shows mean values on an 11-point scale, with responses ranging from 0 (not at all satisfied) to 10 (fully satisfied). Data are sourced from the EU-SILC ad hoc modules (well-being) from 2013 and 2018, as well as from the Canadian General Social Survey and the Well-being survey for Mexico (OECD, 2020[39]). Information on whether social interactions take place face-to-face or via social networks is also sparse. As mentioned above, the frequency of the latter has risen and is likely to continue to do so with digitalisation. The way in which people spend the daily time available to them is a key determinant of their well-being. In the OECD framework for measuring well-being, the Work-life balance dimension refers to a “satisfactory state of equilibrium between an individual’s work and private life”, and is therefore about assessing people’s capacity to combine family commitments, leisure and work – including both paid and unpaid work (OECD, 2011[45]; OECD, 2020[39]). On the one hand, not working enough can potentially prevent individuals from earning sufficient income or developing as a professional and may even reduce their sense of purpose in life. On the other, working too much reduces the time individuals can devote to themselves, their family and their friends, and contributes to worsening their health, particularly when combined with inadequate working conditions (Wong, Chan and Ngan, 2019[227]). Establishing what counts as “too little” or “too much” is a key to assessing work-life balance, and this may depend on individual characteristics such as age, income, job quality, family size and personal preferences. To a certain extent, the section on Work and Job Quality in Chapter 2 informs on these issues in Latin America, as it covers unemployment and people working very long hours. However, long working hours matter for well-being in terms of both paid work (e.g. in salaried employment, as covered in Chapter 2) and unpaid work (e.g. caring responsibilities, cooking, and cleaning in the home). Figure 3.30, Panel A shows that the average weekly hours of unpaid work of the total population in the focal group of countries stands at 27 hours, well above the OECD average of 23 hours. Unpaid work is 37 hours per week in Argentina, over twice as high as in Brazil (18 hours). As a result, the employed population in these two countries face very different working days each week: for Brazilians with a paid job, weekly hours of unpaid work (15 hours) represent just over a third of weekly hours of paid work (40 hours), whereas those in Argentina face almost a “double day” burden of both paid work (39 hours) and unpaid work (33 hours) (Figure 3.30, Panel B). The issue of unpaid work is particularly important from a gender perspective, as women and girls tend to face a disproportionate burden. This is explored further in Chapter 5. The OECD’s Guidelines on Measuring Subjective Well-Being define the concept as “good mental states, including all of the various evaluations, positive and negative, that people make of their lives and the affective reactions of people to their experiences” (OECD, 2013[228]). This definition encompasses three key elements: life evaluation (a reflective assessment on a person’s life or some specific aspect of it); affect (a person’s feelings, emotions and states, typically measured with reference to a particular point in time); and eudaimonia (a sense of meaning and purpose in life, or good psychological functioning). Generally speaking, both the affect scores and the evaluations of life reported in Latin America tend to be relatively high – particularly considering not only what average income levels would predict (Rojas, 2018[229]), but also what might be expected based on objective measures of health or political voice. In this respect, research has drawn attention to the existence of a “Latin American paradox” (Box 3.1). To a certain extent, these favourable results encapsulate the inadequacy of traditional welfare measures for assessing progress, as well as the need to bring into greater focus measures that capture the quality of people’s lives. By taking people’s values into account and by recognising human universality in the experience of well-being, measures of subjective well-being are of upmost relevance in a range of policy debates and strategies to achieve sustainable development. Life satisfaction reflects the way people evaluate their lives as a whole and is measured through survey questions. In OECD countries, information about current levels of life satisfaction can be derived from estimates provided by National Statistics Offices, based on national surveys that rely on broadly comparable questions (OECD, 2017[46]). However, in order to assess changes over time in the focal group of countries and in Latin America overall, the Gallup World Poll is a better source of information, as it provides longer time series and enables the assessment of most countries on a comparable basis.32 The average life satisfaction score across the focal group for the 2017-19 period was slightly above 6, as compared to values close to 7 across OECD countries. Average scores ranged from below 5.7 in the Dominican Republic to 7.1 in Costa Rica. Average satisfaction among the focal group of countries in 2017-19 was very similar to that recorded in 2006-09. This has also been the case for the OECD average, although several OECD member countries experienced marked falls in life satisfaction during the 2008 global financial crisis (OECD, 2013[230]; OECD, 2017[46]). Five countries (the Dominican Republic, Ecuador, Paraguay, Peru and Uruguay) experienced life satisfaction gains of 8% or more between 2006-09 and 2017-19, while respondents in Argentina and Mexico (-4% each) and Brazil (-5%) reported slightly lower life satisfaction in 2017-19 (Figure 3.31, Panel A). At the low end of the scale, 19% of respondents in the focal countries reported life satisfaction of 4 or lower in 2017-19, as compared to 11% in the OECD average (Figure 3.31, Panel B). The share was, however, almost four times higher in the Dominican Republic (33%) than in Costa Rica (9%). It has nonetheless decreased in most countries, namely in Ecuador (-10 percentage points), Peru (-9 points), the Dominican Republic (-6 points.) and Chile (-5 points), whilst increasing in Argentina (5 points). Affect is a term often used in psychology to describe a person’s feelings. Therefore, the different measures of affect reflect particular emotional states, typically referring to a specific point in time (OECD, 2013[228]). The negative affect balance measure shown below is a summary calculated from a battery of items, to which respondents indicate “yes” or “no” to having felt a lot of each emotion or state on the previous day. The negative items considered here relate to anger, sadness and worry, and the positive items to enjoyment, feeling well-rested and laughing or smiling. A negative affect balance refers to respondents who report more negative than positive feelings or states on the previous day (OECD, 2020[39]). The balance of emotional states in the focal group of countries was, on average, slightly more positive than among OECD countries in 2017-19: only 13% of people in the focal group report a negative affect balance, a rate similar to that recorded in OECD countries, on average. Across the focal group, the rate ranges from 17% in Brazil and Peru to 8% or less in Mexico and Paraguay. Between 2006-09 and 2017-19, negative affect balance remained relatively stable both in the focal group and in the OECD average. Negative affect balance increased (implying a deterioration of the situation) by three percentage points or more in Costa Rica and Brazil, and it decreased in Uruguay by just over three percentage points (Figure 3.31, Panel C). While early evidence from the region suggests that the pandemic had certain effects on people’s anxiety and stress, trends in life satisfaction in Latin America are currently less clear-cut. In Colombia, for instance, official data suggest that between September 2020 and February 2021 just under half of the population (43%) felt worried or nervous in the 23 main cities, and that between December 2020 and February 2021, approximately 16% felt sad. However, this data come from a survey launched by the National Statistics Office in 2020 to monitor people’s perceptions and expectations during the crisis, and no reference point is available from previous years (DANE, 2021[211]). In Uruguay, on the other hand, a web survey suggested that 32% of the population felt sad and 67% felt nervous at the end of March 2020 (i.e. at the very beginning of the pandemic once the first restrictions were in place) – figures that were respectively 20 and 37 percentage points higher than the previous year (Bericat and Acosta, 2020[244]). In Argentina, a telephone survey carried out during lockdown in Buenos Aires in May 2020 found that one in five people (21%) reported symptoms of anxiety or depression, expressed as “psychological discomfort” (Rodríguez Espínola, Filgueira and Paternó Manavella, 2020[245]). In contrast, in Mexico, where the National Statistics Office has been measuring life satisfaction in a comparable way over time, life satisfaction for the urban population in January 2021 was similar to that in January 2015 and January 2018 (8.2 on the Cantril Ladder). This represents a very slight fall, however, compared to January 2019 (8.4) and January 2020 (8.3) (INEGI, 2021[101]). In this regard, it is important to note that averages in life satisfaction may mask disparities within the national population. In Mexico for example, women reported slightly lower life satisfaction (8) than men (8.3). Data from the Gallup World Poll show clear impacts of the COVID-19 pandemic across indicators of subjective well-being. In the wake of the pandemic, life satisfaction declined by 0.4 points in the focal group of countries, on average, representing a 7% decrease (Figure 3.32, Panel A). This drop, the largest recorded since 2015, has brought average life satisfaction below the levels recorded in 2006-08. The fall affected all countries in the focal group, with the exceptions of Argentina, Chile and Paraguay, where life satisfaction remained relatively stable. In Mexico, Ecuador, Colombia, Costa Rica and the Dominican Republic, life satisfaction has fallen by 0.5 to 0.8 points, representing changes from -7% to -14% (Figure 3.32, Panel B). The largest drop was recorded in Peru, where life satisfaction fell from 6 to 5 (-17%), leaving it lower than anywhere else in the focal group.33 Between 2019 and 2020, the share of the population reporting a low life satisfaction score increased on average in the focal group, echoing the detrimental effects of the pandemic mentioned above. In 2020, one in four individuals reported a score of 4 or below on a scale of 0-10, compared to approximately one in five just a year earlier. In 9 out of the 11 focal group countries the share increased by 3 percentage points or more, notably in Mexico and Ecuador (by 9 percentage points) and Peru (13 points). On the other hand, this share slightly fell in Paraguay, whilst remaining stable in Chile (Figure 3.32, Panel C). The pandemic has also increased the share of the population experiencing a negative affect balance. On average, 17% of respondents in focal group countries experienced more negative than positive feelings in a typical day in 2020, roughly 4 percentage points more than a year earlier. In 6 out of the 11 focal group countries, the share increased by 3 percentage points or more, particularly in Costa Rica and Mexico (6 points) and Peru (11 points) (Figure 3.32, Panel D). In the remaining countries of the focal group, levels remained broadly stable over the past two years. An ideal set of subjective well-being indicators would encompass different measures of life evaluations, affect, and eudaimonia.34 For example, the OECD Guidelines proposed a core module of five questions, considered to be the minimum necessary to capture these three elements (OECD, 2013[228]). Within that core module, the question on life evaluation (in this case, a question about life satisfaction, rated on a 0 to 10 scale) was selected as the primary measure – i.e. in a scenario where only one question could be included in a survey, it would be the single recommended question. This is largely due to the fact that it is the question for which there is the greatest degree of international consensus on both its construction and use, as well as the strongest evidence base regarding the validity, relevance and reliability of the measure. A majority of OECD national statistical offices are now collecting measures of life satisfaction in an internationally harmonised way, although some methodological variations persist (OECD, 2020[39]). In Chile, for example, life satisfaction data have been collected by the Instituto Nacional de Estadísticas (INE), using a response-scale format that is not comparable with that used in other OECD countries. Mexico also reports life satisfaction for the population on a biannual basis, as part of a module called “BIARE” (“Bienestar Autoreportado”, or self-reported well-being) included in the National Survey of Consumer Confidence (ENCO) by the INEGI; while in Colombia, the statistical office has made steps in this direction in recent years as well. However, despite progress towards harmonisation, life satisfaction data collections in official statistics are still scarce in Latin America, and where they do exist long time series are still lacking. Both the life satisfaction and the negative affect balance data reported in this section are sourced from the Gallup World Poll, due to the lack of harmonised data across statistical offices in the region. The World Poll offers a standardised measurement approach that covers all the focal group countries and provides a consistent time series, collected annually in most cases since 2005/6. As mentioned previously, the Gallup World Poll measure for negative affect is based on people’s feelings and affective states “yesterday”, rather than over a longer time period, to reduce the risk of retrospective recall bias. This is also the case in the BIARE module used by the INEGI (INEGI, 2021[246]). When adopted in conjunction with large sample sizes, this question framing should suffice to extract information on a typical day’s experiences, but estimates may be more volatile over disaggregations across population groups or smaller samples more generally. An alternative framing for survey questions is to ask respondents about states and feelings over a period of several weeks, hence reducing the impact of uncommon events, yet increasing the role of dispositional tendencies influencing the data and the risk of retrospective recall bias. Data on affective experiences collected through time-use surveys are likely to yield the most accurate and useful results (OECD, 2013[228]), but are yet to be included in those undertaken in the focal group of countries, such as Chile or Costa Rica (INE, 2015[247]; INEC, 2017[248]). Eudaimonia measures are absent from this section, due to a lack of internationally harmonised data collected at regular time intervals. Nonetheless, the BIARE module developed by the INEGI is an example of how this measure could be included in national surveys throughout the region moving forward. It includes several positive statements and a negative statement, to which respondents are asked to rate the level of agreement on a scale of 0 to 10 (INEGI, 2021[101]). Recent statistical developments in the field of subjective well-being in Latin America are contributing to advancing this agenda in the region. The Colombian NSO has made considerable progress on developing new tools to measure people’s perceptions, particularly on subjective well-being, in the Social Capital module of its Political Culture Survey (Encuesta de cultura política, ECP). During the COVID-19 pandemic, it has also carried out its “Social Pulse” survey on a more regular basis, in which it has included measures of affect (DANE, 2021[211]). Furthermore, the Mexican statistical office is working towards a new national survey on income and well-being, in close co-ordination with external experts. The survey will include three major dimensions of well-being, including subjective well-being. Finally, Chile’s Ministry of Social Development has been seeking to measure well-being through the Complementary Social Welfare Survey, to complement the existent national socio-economic survey (Encuesta de Caracterización Económica Nacional, CASEN), which is based on the OECD framework. ## References [108] Agüero, J. (2020), COVID-19 and The Rise of Intimate Partner Violence, University of Connecticut, Department of Economics, https://ideas.repec.org/p/uct/uconnp/2020-05.html. [208] Alicea-Planas, J., J. Trudeau and W. Vásquez Mazariegos (2021), “COVID-19 risk perceptions and social distancing practice in Latin America”, Hispanic Health Care International, p. 154041532098514, http://dx.doi.org/10.1177/1540415320985141. [151] Amoatey, P. et al. (2020), “Impact of building ventilation systems and habitual indoor incense burning on SARS-CoV-2 virus transmissions in Middle Eastern countries”, Science of The Total Environment, Vol. 733, p. 139356, http://dx.doi.org/10.1016/j.scitotenv.2020.139356. [91] Asmann, P. (2020), What Does Coronavirus Mean for Criminal Governance in Latin America?, https://www.insightcrime.org/news/analysis/criminal-governance-latin-america-coronavirus/. [177] Asplund, E. et al. (2021), Elections and Covid-19: How special voting arrangements were expanded in 2020, IDEA, https://www.idea.int/news-media/news/elections-and-covid-19-how-special-voting-arrangements-were-expanded-2020. [179] Asplund, E. et al. (2020), People with COVID-19 and those self-isolating must not be denied the vote, London School of Economics, https://blogs.lse.ac.uk/covid19/2020/10/23/people-with-covid-19-and-those-self-isolating-must-not-be-denied-the-vote/. [94] Austin, A. (2020), Latin America Under Threat of Cybercrime Amid Coronavirus, https://www.insightcrime.org/news/analysis/threat-cyber-crime-coronavirus/. [261] Balestra, C., R. Boarini and E. Tosetto (2018), “What matters the most to people? : Evidence from the OECD Better Life Index users’ responses”, OECD Statistics Working Papers, No. 2018/3, OECD Publishing, Paris, https://dx.doi.org/10.1787/edf9a89a-en. [112] Balestra, C. and D. Dottori (2011), “Aging society, health and the environment”, Journal of Population Economics, Vol. 25/3, pp. 1045-1076, http://dx.doi.org/10.1007/s00148-011-0380-x. [170] Barber, B. (1984), Strong Democracy: Participatory Politics for a New Age, University of California Press. [48] Basto-Aguirre, N., P. Cerutti and S. Nieto-Parra (2020), Is COVID-19 widening educational gaps in Latin America? Three lessons for urgent policy action, OECD Development Centre, https://oecd-development-matters.org/2020/06/04/is-covid-19-widening-educational-gaps-in-latin-america-three-lessons-for-urgent-policy-action/. [75] Bautier, E. (2003), “Décrochage scolaire : Genèse et logique des parcours”, Ville-école-intégration Enjeux, Vol. 132, https://halshs.archives-ouvertes.fr/halshs-00808806/document. [77] Belfield, C. and H. Levin (2007), “The Price We Pay: Economic and Social Consequences of Inadequate Education”, p. 273, https://www.jstor.org/stable/10.7864/j.ctt126269. [244] Bericat, E. and M. Acosta (2020), El impacto del COVID-19 en el bienestar emocional de los trabajadores en Uruguay, https://fes-sociologia.com/uploads/public/Bericat%20&%20Acosta%20(2020)%20Impacto%20del%20COVID-19%20en%20el%20bienestar%20emocional%20de%20los%20trabajadores%20en%20Uruguay.pdf. [76] Blaya, Catherine; Hayden, Carol (2003), Constructions Sociales des Absenteismes et des Decrochages Scolaires en France et en Angleterre, https://pdfs.semanticscholar.org/73e9/dc77b35362b53a372057a7971f61cce249f8.pdf. [186] Boarini, R. and M. Díaz (2015), “Cast a ballot or protest in the street - did our grandparents do more of both?: An age-period-cohort analysis in political participation”, OECD Statistics Working Papers, No. 2015/2, OECD Publishing, Paris, https://dx.doi.org/10.1787/5js636gn50jb-en. [260] Bond, T. and K. Lang (2019), “The sad truth about happiness scales”, Journal of Political Economy, Vol. 127/4, pp. 1629-1640, http://dx.doi.org/10.1086/701679. [222] Bowen, K. et al. (2014), “The stress-buffering effects of functional social support on ambulatory blood pressure.”, Health Psychology, Vol. 33/11, pp. 1440-1443, http://dx.doi.org/10.1037/hea0000005. [128] Brezzi, M. and D. Sanchez-Serra (2014), “Breathing the same air? Measuring air pollution in cities and regions”, OECD Regional Development Working Papers, No. 2014/11, OECD Publishing, Paris, https://dx.doi.org/10.1787/5jxrb7rkxf21-en. [184] BTEAM (2021), The Business case for protecting civic rights, https://bteam.org/assets/reports/The-Business-Case-for-Protecting-Civic-Rights.pdf. [173] Busso, M. and J. Messina (eds.) (2020), The Inequality Crisis: Latin America and the Caribbean at the Crossroads, Inter-American Development Bank, http://dx.doi.org/10.18235/0002629. [193] Cacioppo, J., L. Hawkley and R. Thisted (2010), “Perceived social isolation makes me sad: 5-year cross-lagged analyses of loneliness and depressive symptomatology in the Chicago Health, Aging, and Social Relations Study”, Psychol Aging, Vol. 25/2, pp. 453-63, https://doi.org/10.1037/a0017216. [123] CAF (2019), Infraestructura en el desarrollo de América Latina, https://scioteca.caf.com/bitstream/handle/123456789/1465/IDEAL_2017-2018_Infraestructura_en_el_Desarrollo_de_Am%c3%a9rica_Latina_%28documento_principal%29.pdf?sequence=4&isAllowed=y. [118] CAF (2015), Transporte en América Latina, vital para frenar el calentamiento global, https://www.caf.com/es/actualidad/noticias/2015/11/transporte-en-america-latina-vital-para-frenar-el-calentamiento-global/. [137] Cai, W. et al. (2020), “Climate impacts of the El Niño–Southern Oscillation on South America”, Nature Reviews Earth & Environment, Vol. 1/4, pp. 215-231, http://dx.doi.org/10.1038/s43017-020-0040-3. [138] CERF (2007), Annual Report of the Humanitarian/Resident Coordinator on the use of CERF grants, Mexico, 2007, https://cerf.un.org/sites/default/files/resources/2007%20Mexico%20final%20slr.pdf. [259] Chen, L. et al. (2019), Have Econometric Analyses of Happiness Data Been Futile? A Simple Truth About Happiness Scales, https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3390139. [85] Chioda, L. (2017), Stop the Violence in Latin America: A look at prevention from cradle to adulthood, World Bank Group. [182] CIVICUS (2021), Civic Space in Numbers, https://monitor.civicus.org/quickfacts/. [185] CIVICUS (2021), Freedom of expression and the COVID-19 pandemic: a snapshot of restrictions and attacks, https://monitor.civicus.org/COVID19May2021/. [220] Cohen, S. (2005), “Keynote presentation at the eight international congress of behavioral medicine Mainz, Germany August 25–28, 2004”, International Journal of Behavioral Medicine, Vol. 12/3, pp. 123-131, http://dx.doi.org/10.1207/s15327558ijbm1203_1. [221] Cohen, S. et al. (2014), “Does Hugging Provide Stress-Buffering Social Support? A Study of Susceptibility to Upper Respiratory Infection and Illness”, Psychological Science, Vol. 26/2, pp. 135-147, http://dx.doi.org/10.1177/0956797614559284. [219] Cohen, S. and T. Wills (1985), “Stress, social support, and the buffering hypothesis”, Psychological Bulletin, Vol. 98/2, pp. 310-357, http://dx.doi.org/10.1037/0033-2909.98.2.310. [164] Copernicus Land Monitoring Service (2021), Urban Atlas, https://land.copernicus.eu/local/urban-atlas. [89] Crisis Group Latin America (2020), Virus-proof Violence: Crime and COVID-19 in Mexico and the Northern Triangle, https://d2071andvip0wj.cloudfront.net/083-virus-proof-violence.pdf. [84] Dammert, L. (2017), Gang Violence in Latin America, Wiley - Blackwell, https://onlinelibrary.wiley.com/doi/abs/10.1002/9781119057574.whbva102. [211] DANE (2021), Encuesta Pulso Social: Históricos [Social Pulse Survey: Historical], https://www.dane.gov.co/index.php/estadisticas-por-tema/encuesta-pulso-social/encuesta-pulso-social-historicos. [224] DANE (2020), Encuesta de cultura política [Political Culture Survey], https://www.dane.gov.co/index.php/estadisticas-por-tema/cultura/cultura-politica-encuesta#modulo-capital-social. [238] Deaton, A. and A. Stone (2013), “Two happiness puzzles”, American Economic Review, Vol. 103/3, pp. 591-597, http://dx.doi.org/10.1257/aer.103.3.591. [239] Diener, E. and R. Biswas-Diener (2018), Social Well-being: Research and Policy Recommendations, Sustainable Development Solutions Network, https://s3.amazonaws.com/ghc-2018/UAE/GHPR_Ch6.pdf. [165] Dijkstra, L. and H. Poelman (2012), Cities in Europe, the new OECD-EC definition, https://ec.europa.eu/regional_policy/sources/docgener/focus/2012_01_city.pdf. [2] Dong, E., H. Du and L. Gardner (2020), “An interactive web-based dashboard to track COVID-19 in real time”, The Lancet Infectious Diseases, Vol. 20/5, pp. 533-534, http://dx.doi.org/10.1016/s1473-3099(20)30120-1. [218] Duque Franco, I. et al. (2020), “Mapping repertoires of collective action facing the COVID-19 pandemic in informal settlements in Latin American cities”, Environment and Urbanization, Vol. 32/2, pp. 523-546, http://dx.doi.org/10.1177/0956247820944823. [152] Du, W. and G. Wang (2020), “Indoor air pollution was nonnegligible during COVID-19 lockdown”, Aerosol and Air Quality Research, Vol. 20/9, pp. 1851-1855, http://dx.doi.org/10.4209/aaqr.2020.06.0281. [154] Du, W. et al. (2018), “Field-based emission measurements of biomass burning in typical Chinese built-in-place stoves”, Environmental Pollution, Vol. 242, pp. 1587-1597, http://dx.doi.org/10.1016/j.envpol.2018.07.121. [132] ECLAC (2021), Building forward better: action to strengthen the 2030 Agenda for Sustainable Development t (LC/FDS.4/3/Rev.1)., https://www.cepal.org/sites/default/files/publication/files/46696/S2100124_en.pdf. [266] ECLAC (2021), Environmental Statistics and Indicators: Natural extreme events and disasters, https://cepalstat-prod.cepal.org/cepalstat/tabulador/SisGen_MuestraFicha_puntual.asp?id_aplicacion=1&id_estudio=1&indicador=1837&idioma=i. [172] ECLAC (2021), Social Panorama of Latin America, 2020, https://www.cepal.org/sites/default/files/publication/files/46688/S2100149_en.pdf. [148] ECLAC (2020), Effects of the quarantines and activity restrictions related to the coronavirus disease (COVID-19) on air quality in Latin America’s cities, https://repositorio.cepal.org/bitstream/handle/11362/45885/4/S2000475_en.pdf. [22] ECLAC (2020), The Social Challenge in Times of COVID-19, https://www.cepal.org/sites/default/files/publication/files/45544/S2000324_en.pdf. [225] 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. [42] ECLAC (2013), Social Panorama of Latin America 2012, ECLAC, https://www.cepal.org/cgi-bin/getProd.asp?xml=/publicaciones/xml/4/48454/P48454.xml&xsl=/publicaciones/ficha.xsl&base=/publicaciones/top_publicaciones.xsl#. [115] ECLAC (2010), Sustainable Development in Latin America and the Caribbean: Trends, progress and challenges in sustainable consumption and production, mining, transport, chemicals and waste management, United Nations, https://sustainabledevelopment.un.org/content/documents/LAC_background_eng.pdf. [226] ECLAC/ INEGI/I NMUJERES/ UN-Women (2016), Classification of Time-Use Activities for Latin America and the Caribbean (CAUTAL), https://www.cepal.org/en/publications/40170-classification-time-use-activities-latin-america-and-caribbean-cautal. [14] ECLAC-PAHO (2020), Health and the economy: A convergence needed to address COVID-19 and retake the path of sustainable development in Latin America and the Carbbean, https://repositorio.cepal.org/bitstream/handle/11362/45841/4/S2000461_en.pdf. [171] EIU (2020), Democracy Index 2019:, http://www.eiu.com/Handlers/WhitepaperHandler.ashx?fi=Democracy-Index-2019.pdf&mode=wp&campaignid=democracyindex2019. [103] El Espectador (2020), La guerra en Colombia no se detiene, a pesar del coronavirus [The war in Colombia does not stop, despite the coronavirus], https://www.elespectador.com/colombia2020/territorio/la-guerra-en-colombia-no-se-detiene-pesar-del-coronavirus-articulo-911860/. [114] Engemann, K. et al. (2019), “Residential green space in childhood is associated with lower risk of psychiatric disorders from adolescence into adulthood”, Proceedings of the National Academy of Sciences, Vol. 116/11, pp. 5188-5193, http://dx.doi.org/10.1073/PNAS.1807504116. [210] Esteve, A. et al. (2020), “Living alone over the life course: Cross‐national variations on an emerging issue”, Population and Development Review, Vol. 46/1, pp. 169-189, http://dx.doi.org/10.1111/padr.12311. [242] Etheridge, B. and L. Spantig (2020), “The gender gap in mental well-being during the Covid-19 outbreak: Evidence from the UK”, ISER Working Paper Series, https://www.iser.essex.ac.uk/research/publications/working-papers/iser/2020-08.pdf. [127] European Commission, Joint Research Centre (JRC (2015), Dataset: Global Human Settlement Layer (GHSL) project, archived, https://data.jrc.ec.europa.eu/dataset/jrc-ghsl-ghs_pop_gpw4_globe_r2015a. [265] Exton, C. and L. Fleischer (forthcoming), The Future of the OECD Well-being Dashboard, OECD Publishing, Paris. [231] Exton, C., C. Smith and D. Vandendriessche (2015), “Comparing happiness across the world: Does culture matter?”, OECD Statistics Working Papers, No. 2015/4, OECD Publishing, Paris, https://dx.doi.org/10.1787/5jrqppzd9bs2-en. [12] Fachola, M. et al. (2015), “Tentativa e ideación de suicidio en adultos mayores en Uruguay”, Ciência & Saúde Coletiva, Vol. 20/6, pp. 1693-1702, http://dx.doi.org/10.1590/1413-81232015206.02252015. [130] FAO (2021), The Global Administrative Unit Layers (GAUL) 2014 dataset, http://www.fao.org/geonetwork/srv/en/main.home. [257] Feigelman, W., B. Gorman and J. Jordan (2009), “Stigmatization and Suicide Bereavement”, Death Studies, Vol. 33/7, pp. 591-608, http://dx.doi.org/10.1080/07481180902979973. [92] Felbab-Brown, V. (2020), “Mexican cartels and the COVID-19 pandemic.”, in Mexican cartels are providing COVID-19 assistance. Why that’s not surprising., https://www.brookings.edu/blog/order-from-chaos/2020/04/27/mexican-cartels-are-providing-covid-19-assistance-why-thats-not-surprising/. [256] Ferré-Grau, C. et al. (2011), “El estigma del suicidio vivencias de pacientes y familiares con intentos de autolisis”, Index de Enfermería, Vol. 20/3, pp. 155-159, http://dx.doi.org/10.4321/s1132-12962011000200004. [255] Ferrer‐i‐Carbonell, A. and P. Frijters (2004), “How important is methodology for the estimates of the determinants of happiness?”, The Economic Journal, Vol. 114/497, pp. 641-659, http://dx.doi.org/10.1111/j.1468-0297.2004.00235.x. [121] FIDA (2020), “Spotlight on climate and environment: Latin America and the Caribbean”, https://www.ifad.org/fr/web/latest/photo/asset/39019826. [134] Fisher, M. and C. Gamper (2017), Policy Evaluation Framework on The Governance of Critical Infrastructure Resilience in Latin America, https://publications.iadb.org/publications/english/document/Policy-Evaluation-Framework-on-the-Governance-of-Critical-Infrastructure-Resilience-in-Latin-America.pdf. [223] Fleischer, L., C. Smith and C. Viac (2016), “A review of general social surveys”, OECD Statistics Working Papers, No. 2016/9, OECD Publishing, Paris, https://dx.doi.org/10.1787/bb54d16f-en. [254] Frey, B. and A. Stutzer (2000), “Happiness, economy and institutions”, The Economic Journal, Vol. 110/466, pp. 918-938, http://dx.doi.org/10.1111/1468-0297.00570. [97] Gallup World Poll (2021), Gallup World Poll, https://www.gallup.com/analytics/232838/world-poll.aspx. [203] Gao, Q. et al. (2020), Loneliness among older adults in Latin America, China, and India: prevalence, correlates and association with mortality, Research Square, http://dx.doi.org/10.21203/rs.3.rs-19700/v1. [194] Garcia, A. et al. (2016), “Friendship in Latin American social comparative studies”, Interpersona: An International Journal on Personal Relationships, Vol. 10/1, pp. 1-12, http://dx.doi.org/10.5964/ijpr.v10i1.227. [204] Gerst-Emerson, K. and J. Jayawardhana (2015), “Loneliness as a public health issue: The impact of loneliness on health care utilization among older adults”, American Journal of Public Health, Vol. 105/5, pp. 1013-1019, http://dx.doi.org/10.2105/ajph.2014.302427. [149] Glencross, D. et al. (2020), “Air pollution and its effects on the immune system”, Free Radical Biology and Medicine, Vol. 151, pp. 56-68, http://dx.doi.org/10.1016/j.freeradbiomed.2020.01.179. [205] Global Web Index (2019), Global Web Index: Global Social Media Landscape, https://www.globalwebindex.com/hubfs/Downloads/The_Global_Social_Media_Landscape.pdf?utm_campaign=Global%20Social%20Media%20Landscape%202019&utm_medium=email&_hsmi=79595728&_hsenc=p2ANqtz-9el-TBd7FK6mv-u3j664Ei7Ic716cRTPHRhldj_g8T9_DpScyc-j85pSyxQ5bpecVTi. [104] Gobierno de Mexico (2020), Informe Anual de Seguridad 2020, https://www.gob.mx/cms/uploads/attachment/file/603367/CPM_Informe_Anual_de_Seguridad_2020__31dic20.pdf. [192] González, S. (2020), “Testing the evidence, how good are public sector responsiveness measures and how to improve them?”, OECD Working Papers on Public Governance, No. 38, OECD Publishing, Paris, https://dx.doi.org/10.1787/c1b10334-en. [236] Graham, C. and E. Lora (2009), Paradox and Perception: Measuring Quality of Life in Latin America, Brookings Institution Press, http://www.jstor.org/stable/10.7864/j.ctt12805r. [67] Gropello, D. (2020), The costs of COVID-19 in education in Latin America. Acting now to protect the future of our children, https://blogs.worldbank.org/education/costs-covid-19-education-latin-america-acting-now-protect-future-our-children. [207] Hale, T. et al. (2021), “A global panel database of pandemic policies (Oxford COVID-19 Government Response Tracker)”, Nature Human Behaviour, http://dx.doi.org/10.1038/s41562-021-01079-8. [59] Hanushek, E. (2015), “Why standard measures of human capital are misleading”, KDI Journal of Economic Policy, Vol. vol. 37(2), pp. 22-37, https://ideas.repec.org/a/zbw/kdijep/v37y2015i2p22-39.html. [135] Hardoy, J. and G. Pandiella (2009), “Urban poverty and vulnerability to climate change in Latin America”, Environment and Urbanization, Vol. 21/1, pp. 203-224, http://dx.doi.org/10.1177/0956247809103019. [38] Helliwell, J. et al. (2021), World Happiness, Trust and Deaths under COVID-19, Sustainable Development Solutions Network, https://worldhappiness.report/ed/2021/. [232] Helliwell, J. et al. (2018), International Migration and World Happiness, Sustainable Development Solutions Network, https://s3.amazonaws.com/happiness-report/2018/CH2-WHR-lr.pdf. [202] Helliwell, J., R. Layard and J. & Sachs (2018), Latin American Happiness has Social Foundations, Sustainable Development Solutions Network., https://worldhappiness.report/ed/2018/. [229] Helliwell, J., R. Layard and J. Sachs (eds.) (2018), Happiness in Latin America Has Social Foundations, Sustainable Development Solutions Network., https://worldhappiness.report/ed/2018/. [233] Helliwell, J. et al. (2021), World Happiness Report 2021, https://worldhappiness.report/ed/2021/. [237] Helliwell, J. and R. Putnam (2004), “The social context of well-being”, Philosophical Transactions of The Royal Society B Biological Sciences, Vol. 359/1449, https://royalsocietypublishing.org/doi/10.1098/rstb.2004.1522. [66] Hershberg, Flinn-Palcic and Kambhu (2020), The COVID-19 Pandemic and Latin American Universities, https://www.american.edu/centers/latin-american-latino-studies/upload/la-higher-ed-covid-final.pdf. [214] Holt-Lunstad, J., T. Smith and J. Layton (2010), “Social relationships and mortality risk: A meta-analytic review”, PLoS Medicine, Vol. 7/7, p. e1000316, http://dx.doi.org/10.1371/journal.pmed.1000316. [189] Hoskins, B., J. Janmaat and G. Melis (2017), “Tackling inequalities in political socialisation: A systematic analysis of access to and mitigation effects of learning citizenship at school”, Social Science Research, Vol. 68, pp. 88-101, http://dx.doi.org/10.1016/j.ssresearch.2017.09.001. [213] House, J., K. Landis and D. Umberson (1988), “Social relationships and health”, Science, Vol. 241/4865, pp. 540-545, http://dx.doi.org/10.1126/science.3399889. [195] Husted, B. (2002), “Culture and international anti-corruption agreements in Latin America”, Journal of Business Ethics, Vol. 37/413–422, https://doi.org/10.1023/A:1015248921716. [183] ICNL (2021), COVID-19 Civic Freedom Tracker, https://www.icnl.org/covid19tracker/. [174] IDEA (2021), Compulsory voting, https://www.idea.int/data-tools/data/voter-turnout/compulsory-voting. [176] IDEA (2020), Elecciones y COVID-19: lecciones de América Latina, https://www.idea.int/es/news-media/news/es/elecciones-y-covid-19-lecciones-de-am%C3%A9rica-latina. [141] IFRC (2013), Argentina: Floods, https://reliefweb.int/sites/reliefweb.int/files/resources/Argentina%20Floods%20DREF%20operation%20n%20MDRAR007.pdf. [140] IFRC (2011), Colombia: Floods and Landslides, https://reliefweb.int/disaster/fl-2011-000042-col. [139] IFRC (2010), Factsheet Chile earthquake, https://reliefweb.int/report/chile/factsheet-chile-earthquake-08032010. [247] INE (2015), Síntesis de Resultados. La dimensión personal del tiempo, https://www.ine.cl/docs/default-source/uso-del-tiempo-tiempo-libre/publicaciones-y-anuarios/publicaciones/sintesis-resultados-actividades-personales-enut.pdf?sfvrsn=fd9a7cea_6. [111] INEC (2019), Encuesta Nacional Sobre Relaciones Familiares y Violencia de Género contra la Mujeres - Boletin, https://www.ecuadorencifras.gob.ec/documentos/web-inec/Estadisticas_Sociales/Violencia_de_genero_2019/Boletin_Tecnico_ENVIGMU.pdf. [248] INEC (2017), Encuesta Nacional de Uso del Tiempo (ENUT 2017). Resultados Generales, https://www.inec.cr/multimedia/enut-2017-encuesta-nacional-de-uso-del-tiempo. [246] INEGI (2021), Bienestar subjetivo - BIARE Ampliado [Subjective well-being - extended BIARE], https://www.inegi.org.mx/investigacion/bienestar/ampliado/. [101] INEGI (2021), Indicadores de Bienestar Autorreportado de la Poblacion Urbana. Cifras de Enero 2021, https://www.inegi.org.mx/contenidos/saladeprensa/boletines/2021/biare/biare2021_02.pdf. [100] INEGI (2020), Encuesta Nacional de Seguridad Publica Urbana (Septiembre 2020), https://www.inegi.org.mx/contenidos/saladeprensa/boletines/2020/ensu/ensu2020_10.docx. [102] Infosegura (2021), Homicidios en el año del COVID-19: Centroamérica y República Dominicana. [119] IQ Air (2019), 2019 Air Quality Report, https://www.iqair.com/. [253] Jaen-Varas, D. et al. (2014), “Mental health and psychiatric care in Bolivia: what do we know?”, International Journal of Mental Health Systems, Vol. 8/1, http://dx.doi.org/10.1186/1752-4458-8-18. [267] JCE (2020), Junta central electoral, https://jce.gob.do/. [156] Kanchongkittiphon, W. et al. (2015), “Indoor environmental exposures and exacerbation of asthma: An update to the 2000 review by the Institute of Medicine”, Environmental Health Perspectives, Vol. 123/1, pp. 6-20, http://dx.doi.org/10.1289/ehp.1307922. [258] Kenneth C. Land, M. (ed.) (2012), National Accounts of Well-being, https://ink.library.smu.edu.sg/cgi/viewcontent.cgi?article=2309&context=soss_research. [6] Khaltaev, N. and S. Axelrod (2019), “Chronic respiratory diseases global mortality trends, treatment guidelines, life style modifications, and air pollution: preliminary analysis”, Journal of Thoracic Disease, Vol. 11/6, pp. 2643-2655, http://dx.doi.org/10.21037/jtd.2019.06.08. [216] Klinenberg, E. (2016), “Social isolation, loneliness, and living alone: Identifying the risks for public health”, American Journal of Public Health, Vol. 106/5, pp. 786-787, http://dx.doi.org/10.2105/ajph.2016.303166. [9] Kohn, R. et al. (2018), “Mental health in the Americas: An overview of the treatment gap”, Revista Panamericana de Salud Pública, Vol. 42, http://dx.doi.org/10.26633/rpsp.2018.165. [180] LAPOP (2021), Latin American Public Opinion Project: Data, https://www.vanderbilt.edu/lapop/interactive-data.php. [81] Latinobarometro (2020), Latinobarómetro: Data, http://www.latinobarometro.org/latOnline.jsp. [252] Lau, J. et al. (2006), “Positive mental health-related impacts of the SARS epidemic on the general public in Hong Kong and their associations with other negative impacts”, Journal of Infection, Vol. 53/2, pp. 114-124, http://dx.doi.org/10.1016/j.jinf.2005.10.019. [87] Lenzer, G. (2017), Violence against children: Making human rights real, Taylor and Francis, http://dx.doi.org/10.4324/9781351248433. [74] Lessard, A. et al. (2008), “Cheminement de décrocheurs et de décrocheuses”, Dossier thématique, Revue des sciences de l’éducation, Vol. 33/3, pp. 647-662, http://dx.doi.org/10.7202/018962ar. [158] Liu, J. et al. (2018), “Indoor air quality and occupants’ ventilation habits in China: Seasonal measurement and long-term monitoring”, Building and Environment, Vol. 142, pp. 119-129, http://dx.doi.org/10.1016/j.buildenv.2018.06.002. [69] Loades, M. et al. (2020), “Rapid systematic review: The impact of social isolation and loneliness on the mental health of children and adolescents in the context of COVID-19”, Journal of the American Academy of Child & Adolescent Psychiatry, Vol. 59/11, pp. 1218-1239.e3, http://dx.doi.org/10.1016/j.jaac.2020.05.009. [78] Lochner, L. (2011), Non-Production Benefits of Education: Crime, Health, and Good Citizenship, National Bureau of Economic Research, Cambridge, MA, http://dx.doi.org/10.3386/w16722. [19] Lopez-Calva (2020), A greater tragedy than we know: Excess mortality rates suggest that COVID-19 death toll is vastly underestimated in LAC, https://www.latinamerica.undp.org/content/rblac/en/home/presscenter/director-s-graph-for-thought/a-greater-tragedy-than-we-know--excess-mortality-rates-suggest-t.html. [178] López-Calva, L. (2021), The Virus and the Votes: How is COVID-19 changing voter turnout in LAC?, UNDP, https://www.latinamerica.undp.org/content/rblac/en/home/presscenter/director-s-graph-for-thought/the-virus-and-the-votes--how-is-covid-19-changing-voter-turnout-.html. [106] López-Calva, L. (2020), No safer place than home?: The increase in domestic and gender-based violenceduring COVID-19 lockdowns in LAC, UNDP, https://www.latinamerica.undp.org/content/rblac/en/home/presscenter/director-s-graph-for-thought/no-safer-place-than-home---the-increase-in-domestic-and-gender-b.html. [162] Loret de Mola, U. et al. (2017), “On the use of hedonic price indices to understand ecosystem service provision from urban green space in five Latin American megacities”, Forests, Vol. 8/12, p. 478, http://dx.doi.org/10.3390/f8120478. [3] LSHTM CMMID COVID-19 working group (2020), How many are at increased risk of severe COVID-19 disease? Rapid global, regional and national estimates for 2020, Cold Spring Harbor Laboratory, http://dx.doi.org/10.1101/2020.04.18.20064774. [79] Machin, S., O. Marie and S. Vujić (2011), “The Crime Reducing Effect of Education”, The Economic Journal, Vol. 121/552, pp. 463-484, http://dx.doi.org/10.1111/j.1468-0297.2011.02430.x. [167] Mackie, A., I. Haščič and M. Cárdenas Rodríguez (2016), “Population exposure to fine particles: Methodology and results for OECD and G20 countries”, OECD Green Growth Papers, No. 2016/2, OECD Publishing, Paris, https://dx.doi.org/10.1787/5jlsqs8g1t9r-en. [251] Mascayano, F. et al. (2015), “Suicide in Latin America: A growing public health issue”, Revista de la Facultad de Ciencias Médicas, https://www.researchgate.net/publication/298213467_SUICIDE_IN_LATIN_AMERICA_A_GROWING_PUBLIC_HEALTH_ISSUE_SUICIDIO_EN_LATINO_AMERICA_UN_CRECIENTE_PROBLEMA_DE_SALUD_PUBLICA (accessed on 30 April 2021). [68] Ministerio de Educación, Centro de Estudios (2020), Impacto del COVID-19 en los resultados de aprendizaje y escolaridad en Chile, http://documents1.worldbank.org/curated/en/112721598898527225/pdf/Impacto-del-COVID-19-en-los-Resultados-de-Aprendizaje-y-Escolaridad-en-Chile-Analisis-con-Base-en-Herramienta-de-Simulacion-Proporcionada-por-el-Banco-Mundial.pdf. [34] Ministerio de Salud (2021), Subsecretaria de Salud Pública detalla aumento de 310% del presupuesto para Salud Mental, https://www.minsal.cl/subsecretaria-de-salud-publica-detalla-aumento-de-310-del-presupuesto-para-salud-mental/. [44] Morgan, D. et al. (2020), “Excess mortality: Measuring the direct and indirect impact of COVID-19”, OECD Health Working Papers, No. 122, OECD Publishing, Paris, https://dx.doi.org/10.1787/c5dc0c50-en. [82] Muggah, R. (2018), Citizen Security in Latin America: Facts and Figures, Igarapé Institute. [83] Muggah and Szabó (2016), Latin America’s cities: Unequal, dangerous and fragile. But that can change, https://www.weforum.org/agenda/2016/06/latin-america-s-cities-unequal-dangerous-and-fragile-but-that-can-change/. [50] Murtin, F. et al. (2017), “Inequalities in longevity by education in OECD countries: Insights from new OECD estimates”, OECD Statistics Working Papers, No. 2017/2, OECD Publishing, Paris, https://dx.doi.org/10.1787/6b64d9cf-en. [212] Newman, M. and N. Zainal (2020), “The value of maintaining social connections for mental health in older people”, The Lancet Public Health, Vol. 5/1, pp. e12-e13, http://dx.doi.org/10.1016/s2468-2667(19)30253-1. [117] NRDC (2014), Limpiando el aire de América Latina: la reducción de las emisiones de carbono negro puede beneficiar rápidamente al clima y la salud pública, https://www.nrdc.org/sites/default/files/latin-america-diesel-pollution-IB-sp.pdf. [105] Nugent, C. (2020), Why Armed Groups in Latin America Are Enforcing COVID-19 Lockdowns, https://time.com/5870054/coronavirus-latin-america-armed-groups/. [131] OCHA (2019), Natural disasters in Latin America and the Caribbean, https://www.humanitarianresponse.info/sites/www.humanitarianresponse.info/files/documents/files/20191202-ocha-desastres_naturales2.pdf. [35] OECD (2021), A New Benchmark for Mental Health Systems: Tackling the Social and Economic Costs of Mental Ill-Health, OECD Health Policy Studies, OECD Publishing, Paris, https://dx.doi.org/10.1787/4ed890f6-en. [73] OECD (2021), OECD Skills Surveys: PIAAC, https://www.oecd.org/fr/sites/evaluationdescompetencesdesadultespiaac/programme/. [161] OECD (2020), “Biodiversity and the economic response to COVID-19: Ensuring a green and resilient recovery”, OECD Policy Responses to Coronavirus (COVID-19), OECD Publishing, Paris, https://doi.org/10.1787/d98b5a09-en (accessed on 12 October 2020). [181] OECD (2020), “COVID-19 in Latin America and the Caribbean: An overview of government responses to the crisis”, OECD Policy Responses to Coronavirus (COVID-19), OECD Publishing, Paris, https://doi.org/10.1787/0a2dee41-en. [209] OECD (2020), “COVID-19: Protecting people and societies”, OECD Publishing, Paris, https://doi.org/10.1787/e5c9de1a-en. [52] OECD (2020), Education at a Glance 2020: OECD Indicators, OECD Publishing, Paris, https://dx.doi.org/10.1787/69096873-en. [150] OECD (2020), “Environmental health and strengthening resilience to pandemics”, OECD Policy Responses to Coronavirus (COVID-19), OECD Publishing, Paris, https://doi.org/10.1787/73784e04-en. [39] OECD (2020), How’s Life? 2020: Measuring Well-being, OECD Publishing, Paris, https://dx.doi.org/10.1787/9870c393-en. [129] OECD (2020), OECD Territorial grids, OECD, Paris, http://www.oecd.org/regional/regional-statistics/territorial-grid.pdf (accessed on 9 March 2021). [54] OECD (2020), “Sources, methods and technical notes”, in Education at a Glance 2020: OECD Indicators, OECD Publishing, Paris, https://dx.doi.org/10.1787/442eac2f-en. [72] OECD (2020), “The OECD Survey on Social and Emotional Skills”, OECD, Paris, https://www.oecd.org/education/ceri/social-emotional-skills-study/. [51] OECD (2019), Education at a Glance 2019: OECD Indicators, OECD Publishing, Paris, https://dx.doi.org/10.1787/f8d7880d-en. [190] OECD (2019), Government at a Glance 2019, OECD Publishing, Paris, https://dx.doi.org/10.1787/8ccf5c38-en. [71] OECD (2019), How’s Life in the Digital Age?: Opportunities and Risks of the Digital Transformation for People’s Well-being, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264311800-en. [55] OECD (2019), PISA 2018 Results (Volume I): What Students Know and Can Do, PISA, OECD Publishing, Paris, https://dx.doi.org/10.1787/5f07c754-en. [263] OECD (2019), PISA 2018 Results (Volume II): Where All Students Can Succeed, PISA, OECD Publishing, Paris, https://dx.doi.org/10.1787/b5fd1b8f-en. [133] OECD (2019), Risk Governance Scan of Colombia, OECD Reviews of Risk Management Policies, OECD Publishing, Paris, https://dx.doi.org/10.1787/eeb81954-en. [62] OECD (2019), Skills Matter: Additional Results from the Survey of Adult Skills, OECD Skills Studies, OECD Publishing, Paris, https://dx.doi.org/10.1787/1f029d8f-en. [200] OECD (2019), Trends Shaping Education 2019, OECD Publishing, Paris, https://dx.doi.org/10.1787/trends_edu-2019-en. [160] OECD (2018), Biodiversity Conservation and Sustainable Use in Latin America: Evidence from Environmental Performance Reviews, OECD Environmental Performance Reviews, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264309630-en. [262] OECD (2018), “PISA for Development Science Framework”, in PISA for Development Assessment and Analytical Framework: Reading, Mathematics and Science, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264305274-6-en. [49] OECD (2017), Education in Costa Rica, Reviews of National Policies for Education, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264277335-en. [16] OECD (2017), Government at a Glance 2017, OECD Publishing, Paris, https://dx.doi.org/10.1787/gov_glance-2017-en. [124] OECD (2017), Green Growth Indicators 2017, OECD Green Growth Studies, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264268586-en. [46] OECD (2017), How’s Life? 2017: Measuring Well-being, OECD Publishing, Paris, https://dx.doi.org/10.1787/how_life-2017-en. [57] OECD (2017), “PISA 2015 Technical Background”, in PISA 2015 Results (Volume III): Students’ Well-Being, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264273856-19-en. [58] OECD (2016), PISA 2015 Results (Volume I): Excellence and Equity in Education, PISA, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264266490-en. [47] OECD (2016), Skills Matter: Further Results from the Survey of Adult Skills, OECD Skills Studies, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264258051-en. [147] OECD (2016), The Economic Consequences of Outdoor Air Pollution, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264257474-en. [61] OECD (2016), “What does low proficiency in literacy really mean?”, Adult Skills in Focus, No. 2, OECD Publishing, Paris, https://dx.doi.org/10.1787/5jm0v427jl9p-en. [96] OECD (2015), How’s Life? 2015: Measuring Well-being, OECD Publishing, Paris, https://dx.doi.org/10.1787/how_life-2015-en. [230] OECD (2013), How’s Life? 2013: Measuring Well-being, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264201392-en. [228] OECD (2013), OECD Guidelines on Measuring Subjective Well-being, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264191655-en. [63] OECD (2013), OECD Skills Outlook 2013: First Results from the Survey of Adult Skills, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264204256-en. [45] OECD (2011), How’s Life?: Measuring Well-being, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264121164-en. [36] OECD (forthcoming), Latin American Economic Outlook (LEO) 2021. [241] OECD (forthcoming), Life in the first year of the pandemic: Evidence on COVID-19 and Well-Being, OECD Publishing. [56] OECD/CAF/UN ECLAC (2016), Latin American Economic Outlook 2017: Youth, Skills and Entrepreneurship, OECD Publishing, Paris, https://dx.doi.org/10.1787/leo-2017-en. [20] OECD et al. (2020), Latin American Economic Outlook 2020: Digital Transformation for Building Back Better, OECD Publishing, Paris, https://dx.doi.org/10.1787/e6e864fb-en. [4] 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. [13] OECD/WHO/World Bank Group (2018), Delivering Quality Health Services: A Global Imperative, World Health Organization, Geneva 27, https://dx.doi.org/10.1787/9789264300309-en. [197] Ogliastri, E. (1997), Una introducción a la negociación internacional: La cultura latinoamericana frente a la angloamericana, japonesa, francesa y del Medio Oriente, http://conocimiento.incae.edu/~ogliaste/mono-49%20intro%20n%20int%20%5B1%5D.pdf. [18] PAHO (2021), Hospitalizations and deaths of younger people soar due to COVID-19, PAHO Director reports, https://www.paho.org/en/news/5-5-2021-hospitalizations-and-deaths-younger-people-soar-due-covid-19-paho-director-reports. [33] PAHO (2020), The Impact of COVID-19 on Mental Neurological and Substance Use Services in the Americas: Results of a Rapid Assessment, https://iris.paho.org/bitstream/handle/10665.2/52999/PAHONMHMHCovid-19200044_eng.pdf?sequence=1&isAllowed=y. [28] PAHO/UNFPA/UNICEF (2017), Accelerating progress toward the reduction of adolescent pregnancy in Latin America and the Caribbean, https://iris.paho.org/bitstream/handle/10665.2/34493/9789275119761-eng.pdf?sequence=1&isAllowed=y&ua=1. [215] Pantell, M. et al. (2013), “Social isolation: A predictor of mortality comparable to traditional clinical risk factors”, American Journal of Public Health, Vol. 103/11, pp. 2056-2062, http://dx.doi.org/10.2105/ajph.2013.301261. [206] Parkin, J., D. Phillips and D. Agren (2020), Covid warnings ring out as Latin America bids to return to normality, The Guardian, https://www.theguardian.com/world/2020/sep/19/latin-america-covid-coronavirus-warnings. [142] Penn State (2020), “’Gargantuan’ hail in Argentina may have smashed world record”, https://www.sciencedaily.com/releases/2020/04/200429191903.htm. [107] Perez-Vincent, S. et al. (2020), COVID-19 lockdowns and domestic violence: Evidence from two studies in Argentina, IADB, p. 48, https://publications.iadb.org/publications/english/document/COVID-19-Lockdowns-and-Domestic-Violence-Evidence-from-Two-Studies-in-Argentina.pdf. [163] Poelman, H. (2018), A walk to the park? Assessing access to green areas in Europe’s cities, update using completed Copernicus urban atlas data“, https://ec.europa.eu/regional_policy/sources/docgener/work/2018_01_green_urban_area.pdf. [145] Pozzer, A. et al. (2020), “Regional and global contributions of air pollution to risk of death from COVID-19”, Cardiovascular Research, Vol. 116/14, pp. 2247-2253, http://dx.doi.org/10.1093/cvr/cvaa288. [188] Prats, M. and A. Meunier (2021), “Political efficacy and participation: An empirical analysis in European countries”, OECD Working Papers on Public Governance, No. 46, OECD Publishing, Paris, https://dx.doi.org/10.1787/4548cad8-en. [175] Querido, L. (2020), “Elections under the stress of the pandemic”, Transparencia Electoral, https://www.transparenciaelectoral.org/elections-under-the-stress-of-the-pandemic/. [93] Rivard Piché, G. (2020), In El Salvador, criminal gangs are enforcing virus-related restrictions. Here’s why., https://www.washingtonpost.com/politics/2020/06/01/el-salvador-criminal-gangs-are-enforcing-coronavirus-curfews-heres-why/. [245] Rodríguez Espínola, S., P. Filgueira and M. Paternó Manavella (2020), Acceso a Servicios de Salud y Efectos en el Bienestar Subjetivo en Tiempos de Aislamiento, https://repositorio.uca.edu.ar/bitstream/123456789/10218/1/acceso-servicios-salud-efectos.pdf. [234] Rojas, M. (2020), Well-Being in Latin America, Springer International Publishing, Cham, http://dx.doi.org/10.1007/978-3-030-33498-7. [199] Rojas, M. (2019), “High subjective well-being in Latin America: Person-based interpersonal relations and public-policy lessons”, Paper presented during Parallel Session 2A of the International Conference on the Policy Uses of Well-being and Sustainable Development Indicators in Latin America and the Caribbean, 23-24 October 2019, Universidad del Rosario, Bogotá, Colombia., https://www.oecd.org/statistics/lac-well-being-metrics.htm. [24] Sattar, N., I. McInnes and J. McMurray (2020), “Obesity Is a risk factor for severe COVID-19 infection”, Circulation, Vol. 142/1, pp. 4-6, http://dx.doi.org/10.1161/circulationaha.120.047659. [201] Sauter, S., L. Kim and K. Jacobsen (2019), “Loneliness and friendlessness among adolescents in 25 countries in Latin America and the Caribbean”, Child and Adolescent Mental Health, Vol. 25/1, pp. 21-27, http://dx.doi.org/10.1111/camh.12358. [99] Semple, K. and A. Ahmed (2020), Murder Rates Were Staggering. The Virus Has Brought Some Quiet, for Now., https://www.nytimes.com/2020/04/11/world/americas/coronavirus-murder-latin-america-crime.html. [168] Sen, A. (1999), Development as Freedom, Alfred Knopf. [166] Shaddick, G. et al. (2018), “Data integration for the assessment of population exposure to ambient air pollution for global burden of disease assessment”, Environmental Science & Technology, Vol. 52/16, pp. 9069-9078, http://dx.doi.org/10.1021/acs.est.8b02864. [153] Shen, G. et al. (2017), “A laboratory comparison of emission factors, number size distributions, and morphology of ultrafine particles from 11 different household cookstove-fuel systems”, Environmental Science & Technology, Vol. 51/11, pp. 6522-6532, http://dx.doi.org/10.1021/acs.est.6b05928. [109] Silverio-Murillo, A. and J. Balmori de la Miyar (2020), Families under Confinement: COVID-19, Domestic Violence, and Alcohol Consumption, p. 28, https://www.adansilveriomurillo.com/uploads/6/9/2/9/69299595/ipv_covid_19.pdf. [116] Solbrig, O. (1998), “The environmental agenda in Latin America - The issue of the 21st century”, Harvard Review of Latin America, https://revista.drclas.harvard.edu/book/environmental-agenda-latin-america#:~:text=Yet%20Latin%20America%20also%20has,alarming%3B%20soil%20erosion%20is%20severely. [37] Srinivasan, R. and J. Clifton (2020), Gallup Keeps Listening to the World Amid the Pandemic, Gallup Worl Poll, https://news.gallup.com/opinion/gallup/316016/gallup-keeps-listening-world-amid-pandemic.aspx. [95] Statista (2020), “Growth of domestic violence and sexual abuse reports during the COVID-19 lockdown in selected Latin American countries as of April 2020”, https://www.statista.com/statistics/1113975/gender-violence-growth-coronavirus-latin-america/. [169] Stutzer, A. and B. Frey (2006), “Political participation and procedural utility: An empirical study”, European Journal of Policital Research, Vol. 45/3, https://doi.org/10.1111/j.1475-6765.2006.00303.x. [240] The Economist (2021), The pandemic has changed the shape of global happiness, https://www.economist.com/international/2021/03/20/the-pandemic-has-changed-the-shape-of-global-happiness. [53] UIS (2021), Youth/adult educational attainment rates by age group and level of education, http://data.uis.unesco.org/OECDStat_Metadata/ShowMetadata.ashx?Dataset=SDG_DS&Coords=%5bSDG_IND%5d.%5bEA_3T8_AG25T99%5d&ShowOnWeb=true&Lang=en. [268] UIS (2021), Youth/adult literacy rate, http://uis.unesco.org/en/glossary-term/youthadult-literacy-rate (accessed on 23 June 2021). [187] UN (2020), Handbook on Governance Statistics, https://unstats.un.org/unsd/statcom/51st-session/documents/Handbook_on_GovernanceStatistics-Draft_for_global_consultation-E.pdf. [21] UN (2020), Policy Brief: The Impact of COVID-19 on Latin America and the Caribbean, https://www.un.org/sites/un2.un.org/files/sg_policy_brief_covid_lac.pdf. [191] UN (2020), SDG Goal 16: Promote peaceful and inclusive societies for sustainable development, provide access to justice for all and build effective, accountable and inclusive institutions at all levels, https://sdgs.un.org/goals/goal16. [144] UN (2018), SDG Indicator 11.5.1: Number of deaths, missing persons and directly affected persons attributed to disasters per 100 000 population, metadata, https://unstats.un.org/sdgs/metadata/files/Metadata-11-05-01.pdf. [15] UN DESA (2019), Indicator 3.8.2: Proportion of population with large household expenditure on health as a share of total, metadata file, https://unstats.un.org/sdgs/metadata/files/Metadata-03-08-02.pdf. [29] UNDESA (2017), Living Arrangements of Older Persons: A Report on an Expanded International Dataset, United Nations, Department of Economic and Social Affairs, Population Division, https://www.un.org/en/development/desa/population/publications/pdf/ageing/LivingArrangements.pdf. [110] UNDP (2017), Comparing policy interventions on domestic violence in Latin America: criminalization, female empowerment and male engagement, http://www.svri.org/sites/default/files/attachments/2016-08-17/FUNDAR_Domestic%20Violence_final.pdf. [159] UNEP (2020), Waste management as an essential service in Latin America and the Caribbean, https://wedocs.unep.org/bitstream/handle/20.500.11822/32615/COVID19_WASTE_LAC.pdf?sequence=1&isAllowed=y. [120] UNEP (2018), Waste Management Outlook for Latin America and the Caribbean, https://www.unep.org/resources/report/waste-management-outlook-latin-america-and-caribbean. [64] UNESCO (2021), Education: From disruption to recovery, https://en.unesco.org/covid19/educationresponse#durationschoolclosures. [27] UNFPA (2020), Provision of Sexual and Reproductive Health care and Family Planning during the COVID-19 pandemic health emergency, https://lac.unfpa.org/sites/default/files/pub-pdf/3-Covid-SSRyPF_ENG%20%281%29.pdf. [143] UNGA (2016), Report of the open-ended intergovernmental expert working group on indicators and terminology relating to disaster risk reduction (OEIWG), https://www.preventionweb.net/files/50683_oiewgreportenglish.pdf. [31] UNICEF (2020), The impact of COVID-19 on the mental health of adolescents and youth, https://www.unicef.org/lac/en/impact-covid-19-mental-health-adolescents-and-youth. [40] United Nations (2005), Budapest Initiative Task Force on Measurement of Health Status: Survey module for measuring health state. [90] UNODC (2020), Research brief: Effect of the COVID-19 pandemic and related restrictions on homicide and property crime, https://www.unodc.org/documents/data-and-analysis/covid/Property_Crime_Brief_2020.pdf. [86] UNODC (2020), UNODC-INEGI Centro de Excelencia para Información Estadística de Gobierno, Seguridad Pública, Victimización y Justicia, http://www.cdeunodc.inegi.org.mx/index.php/viclab11/. [80] UNODC (2016), International Classification of Crime for Statistical Purposes, Version 1.0, https://www.unodc.org/documents/data-and-analysis/statistics/crime/ICCS/ICCS_English_2016_web.pdf. [196] Uthoff, A. and L. Beccaria (2007), Panorama social de América Latina 2007. [198] Vega Angarita and González Escobar (2009), “Social support key element in confronting chronic illness”, Enfermería Global, http://scielo.isciii.es/scielo.php?script=sci_arttext&pid=S1695-61412009000200021. [250] Voßemer, J. et al. (2017), “The effects of unemployment and insecure jobs on well-being and health: The moderating role of labor market policies”, Social Indicators Research, Vol. 138/3, pp. 1229-1257, http://dx.doi.org/10.1007/s11205-017-1697-y. [126] Wang, H. et al. (2020), “Global age-sex-specific fertility, mortality, healthy life expectancy (HALE), and population estimates in 204 countries and territories, 1950–2019: A comprehensive demographic analysis for the Global Burden of Disease Study 2019”, The Lancet, Vol. 396/10258, pp. 1160-1203, http://dx.doi.org/10.1016/s0140-6736(20)30977-6. [136] Warn, E. and S. Adamo (2014), The Impact of Climate Change: Migration and Cities in South America, https://public.wmo.int/en/resources/bulletin/impact-of-climate-change-migration-and-cities-south-america. [41] Washington Group on Disability Statistics (2016), The Washington Group Short Set on Functioning (WG-SS). [5] WDI (2021), World Bank Development Indicators, https://databank.worldbank.org/source/world-development-indicators. [70] WFP (2020), World Food Programme gears up to support children left without meals due to COVID-19 school closures, https://www.wfp.org/news/world-food-programme-gears-support-children-left-without-meals-due-covid-19-school-closures. [7] WHO (2021), Adult mortality rate (probability of dying between 15 and 60 years per 1000 population), metadata, https://www.who.int/data/gho/indicator-metadata-registry/imr-details/64. [264] WHO (2021), Ambient (outdoor) air pollution, https://www.who.int/news-room/fact-sheets/detail/ambient-(outdoor)-air-quality-and-health (accessed on 16 March 2021). [43] WHO (2021), Mental health and substance use data, https://www.who.int/data/gho/data/themes/mental-health. [32] WHO (2020), Mental health and COVID-19, https://www.euro.who.int/en/health-topics/health-emergencies/coronavirus-covid-19/technical-guidance/mental-health-and-covid-19. [23] WHO (2020), Rapid assessment of service delivery for noncommunicable diseases (NCDs) during the COVID-19 pandemic, https://www.who.int/publications/m/item/rapid-assessment-of-service-delivery-for-ncds-during-the-covid-19-pandemic. [98] WHO (2020), Road traffic injuries, https://www.who.int/news-room/fact-sheets/detail/road-traffic-injuries. [11] WHO (2019), Suicide, https://www.who.int/news-room/fact-sheets/detail/suicide. [25] WHO (2018), Global Health Estimates 2016: Deaths by Cause, Age, Sex, by Country and by Region, 2000-2016, http://www.who.int/healthinfo/global_burden_disease/estimates/en/. [122] WHO (2016), Preventing disease through healthy environments: a global assessment of the burden of disease from environmental risks. [88] WHO (2013), Global and regional estimates of violence against women: prevalence and health effects of intimate partner violence and non-partner sexual violence, https://apps.who.int/iris/handle/10665/85239. [8] WHO (2013), WHO-AIMS Report on Mental Health Systems in Latin America and the Caribbean, https://www.paho.org/hq/dmdocuments/2013/ENG-WHOAIMSREG-(For-Web-Apr-2013).pdf. [125] WHO (2006), WHO Air quality guidelines for particulate matter, ozone, nitrogen dioxide and sulfur dioxide - Global update 2005 - Summary of risk assessment., http://whqlibdoc.who.int/hq/2006/WHOSDEPHEOEH06.02_eng.pdf. [1] WHO (1948), Constitution, https://www.who.int/about/who-we-are/constitution. [10] WHO and Calouste Gulbekian Foundation (2014), Social determinants of mental health, https://www.who.int/mental_health/publications/gulbenkian_paper_social_determinants_of_mental_health/en/. [113] WHO Regional Office for Europe (2016), Urban green space and health: a review of evidence, http://www.euro.who.int/__data/assets/pdf_file/0005/321971/Urban-green-spaces-and-health-review-evidence.pdf?ua=1. [227] Wong, K., A. Chan and S. Ngan (2019), “The effect of long working hours and overtime on occupational health: A meta-analysis of evidence from 1998 to 2018”, International Journal of Environmental Research and Public Health, Vol. 16/12, p. 2102, http://dx.doi.org/10.3390/ijerph16122102. [235] World Bank (2021), GNI per capita, PPP (constant 2017 international$), https://data.worldbank.org/indicator/NY.GNP.PCAP.PP.KD.

[60] World Bank (2020), Literacy rate, adult total (% of people ages 15 and above), https://data.worldbank.org/indicator/SE.ADT.LITR.ZS.

[65] World Bank (2020), Simulating the potential impacts of COVID-19 school closures on schooling and learning outcomes: a set of global estimates, http://pubdocs.worldbank.org/en/798061592482682799/covid-and-education-June17-r6.pdf.

[26] World Bank (2006), Maternal Mortality at a Glance, http://hdl.handle.net/10986/9617.

[217] World Economic Forum (2020), A pandemic of solidarity? This is how people are supporting one another as coronavirus spreads, https://www.weforum.org/agenda/2020/03/covid-19-coronavirus-solidarity-help-pandemic/ (accessed on 20 October 2020).

[17] Worldometer (2021), Latin America and the Caribbean Population, https://www.worldometers.info/world-population/latin-america-and-the-caribbean-population/.

[249] Wulfgramm, M. (2014), “Life satisfaction effects of unemployment in Europe: The moderating influence of labour market policy”, Journal of European Social Policy, Vol. 24/3, pp. 258-272, http://dx.doi.org/10.1177/0958928714525817.

[146] Wu, X. et al. (2020), Exposure to air pollution and COVID-19 mortality in the United States: A nationwide cross-sectional study, Cold Spring Harbor Laboratory, http://dx.doi.org/10.1101/2020.04.05.20054502.

[243] WVS Database (2020), World Values Survey, https://www.worldvaluessurvey.org/WVSContents.jsp.

[30] YouGov (2020), How COVID-19 is affecting mental health across the globe, https://today.yougov.com/topics/health/articles-reports/2020/12/10/covid-19-mental-health-global.

[157] Zhang, Y. et al. (2011), “Can commonly-used fan-driven air cleaning technologies improve indoor air quality? A literature review”, Atmospheric Environment, Vol. 45/26, pp. 4329-4343, http://dx.doi.org/10.1016/j.atmosenv.2011.05.041.

[155] Zhao, Y. et al. (2019), “Characterization of particulate matter from heating and cooling several edible oils”, Building and Environment, Vol. 152, pp. 204-213, http://dx.doi.org/10.1016/j.buildenv.2019.02.007.

## Notes

← 1. Throughout this report, the eleven focal countries refer to Argentina, Brazil, Chile, Colombia, Costa Rica, Dominican Republic, Ecuador, Mexico, Paraguay, Peru and Uruguay.

← 2. Health status is consistently ranked as one of the most valued aspects in people’s lives in public consultations that have informed the construction of national well-being frameworks in OECD countries (e.g. in Italy, Germany, Israel and Scotland) and by the users of the OECD Better Life Index (Balestra, Boarini and Tosetto, 2018[261]).

← 3. The “epidemiological transition” (from communicable to non-communicable diseases) observed in many OECD countries is also affecting Latin America and the Caribbean, where the burden of non-communicable diseases among adults is increasing over time (OECD/The World Bank, 2020[4]).

← 4. According to the WHO Global Health Estimates, mental disorders and neurological conditions include: depressive disorders, bipolar disorders, schizophrenia, anxiety disorders, eating disorders, autism and Asperger’s syndrome, idiopathic intellectual disability, Alzheimer’s disease and other dementias, Parkinson’s disease, epilepsy, multiple sclerosis, migraine and non-migraine headache (WHO, 2018[25]).

← 5. Cross-country comparisons of suicide data need to be handled with care: estimates largely depend on the quality of the vital registration system, which can vary from one country to another, affecting both current levels as well as trends. Moreover, suicides often go under-reported, meaning that in countries with a low coverage of deaths in general (i.e. where a high share of deaths do not end up in the vital registration system), a large share may be accounted for by suicides (Mascayano et al., 2015[251]). The social stigma around suicide and mental disorders may also impair reporting levels, thereby affecting the comparability of data among countries. In practice, this situation is also a risk factor, since it may prevent effective and timely access to health-care services when these are most needed (Feigelman, Gorman and Jordan, 2009[257]; Ferré-Grau et al., 2011[256]; Jaen-Varas et al., 2014[253]).

← 6. It is important to note that the UHC is a composite index, which combines access indicators with outcome indicators (for instance, prevalence of blood pressure above certain levels). The UHC index also includes certain indicators of available resources (availability of hospital beds and health workers). Outcome indicators are influenced not only by health policy, but also by individual preferences and behaviours. They are not direct measures of access to health-care services. Similarly, indicators of available resources are not direct measures of access to health-care services. Moreover, not all indicators included in the UHC index are equally well-suited for different contexts (for instance, malaria prevention in non-tropical countries). Finally, data from the UHC index must be interpreted with caution in this report, since the main data source for certain tracer areas are health administrative records, and not all focal countries have the same levels of coverage and quality of administrative records. An example of these caveats is the fact that in 2017, access to essential services in Chile according to the UHC (second-lowest among the focal group of countries) is somewhat counter-intuitive when considering levels of life expectancy or maternal and infant mortality.

← 7. These shares represent almost 26.7 million confirmed cases and over 846 000 deaths as of 14 April 2021 (Dong, Du and Gardner, 2020[2]).

← 8. Prevalence estimates were extracted for the following disease categories by age, sex and country: (1) cardiovascular diseases (CVD), including CVD caused by hypertension; (2) chronic kidney disease (CKD), including CKD caused by hypertension; (3) chronic respiratory disease; (4) chronic liver disease; (5) diabetes; (6) cancers with direct immunosuppression; (7) cancers without direct immunosuppression, but with possible immunosuppression caused by treatment; (8) HIV/AIDS; (9) tuberculosis; (10) chronic neurological disorders; and (11) sickle cell disorders.

← 9. For the purpose of computing summary statistics on longevity inequality, the category “No schooling” was merged with “Primary and lower secondary” to form the category “Low level of education” (Murtin et al., 2017[50]).

← 10. In all of these countries, the share of adults with tertiary education is among the lowest in OECD and partner countries (less than 25%), which may partially explain the large earnings advantage of tertiary-educated workers.

← 11. PISA measures the performance of only those 15-year-olds who are still in school. In the case of Costa Rica, lack of progress in average scores since 2009 masks the fact that a higher share of the youth cohort has been attending school (including more students from disadvantaged backgrounds) and is hence participating in PISA. However, other countries in the region, such as Peru, have succeeded in simultaneously enrolling more children and improving PISA average learning outcomes (OECD, 2017[49]).

← 12. In 2018, 600 000 students representing approximately 32 million 15-year-olds in the schools of the 79 participating countries sat the 2-hour PISA test. “Top performers” are those who achieved Level 5 or 6 in a given subject, whereas “low-achievers” are those who scored below Level 2 (OECD, 2019[55]).

← 13. At Levels 5 or above, students can comprehend lengthy texts, deal with concepts that are abstract or counterintuitive, and establish distinctions between fact and opinion, based on implicit cues pertaining to the content or source of the information (OECD, 2019[263]).

← 14. In terms of reading proficiency, at Level 2, students begin to show the capacity to use their reading skills to acquire knowledge and solve practical problems. Students who fail to attain Level 2 proficiency in reading often encounter difficulties when confronted with material that is unfamiliar to them or that is of moderate complexity and length. They often need to be prompted with instructions or cues before being able to engage with a text. In the context of the UN 2030 Agenda, Level 2 proficiency has been identified as the “minimum level of proficiency” that all children should acquire by the end of secondary education (OECD, 2019[55]). In terms of proficiency in mathematics, at Level 2 students begin to show the initiative and ability to use mathematics in simple real-life situations. Although students who score below this minimum level can be considered particularly at risk, Level 2 proficiency is not necessarily a “sufficient” level of mathematics proficiency for making well-founded decisions and judgements in personal or professional situations for which mathematical literacy is required (OECD, 2019[55]). Nonetheless, it is also the level of proficiency considered for the UN SDGs. Level 2 in science is also an important benchmark for student performance. On the PISA scale, it represents the level of achievement at which students begin to show the science competences that enable them to engage in reasoned discourse about science and technology (OECD, 2018[262]). At Level 2, the competences and attitudes required to effectively engage with science-related issues are only just emerging. Students demonstrate everyday scientific knowledge, as well as a basic comprehension of scientific enquiry, which they can mainly apply in familiar contexts (OECD, 2019[55]).

← 15. The questions asked in surveys vary between countries, and not all of them rely on the “able to read and write a simple statement” definition of literacy (UIS, 2021[268]).

← 16. A numerate adult will respond appropriately to mathematical content, ideas and information represented in different ways to solve problems and manage situations in a real-life context. Although performance on numeracy tasks is, in part, dependent on the ability to read and understand text, numeracy involves more than applying arithmetical skills to information embedded in text (OECD, 2019[62]). Akin to the literacy scale, the scale for numeracy proficiency is divided into six levels: Levels 1 to 5 and below Level 1. Tasks below Level 1 require the respondents to carry out simple processes (counting, sorting, performing basic arithmetic operations with whole numbers or money) or to recognise common spatial representations in familiar, concrete contexts where the mathematics content is explicit, i.e. with little/no distractors or text. Tasks at Level 5, on the other hand, require respondents to understand complex representations and abstract and formal mathematical and statistical ideas, possibly embedded in complex texts (OECD, 2019[62]).

← 18. Calculation: Cumulative drop-out rate to last grade of primary education = 100% - Survival rate to last grade of primary education.

The cumulative drop-out rate to the last grade of secondary education cannot be derived from any other data in the UIS database.

← 19. This indicator was included in the Gallup World Poll survey as of 2019, and therefore no time series is available.

← 20. Countries considered: Belize, Costa Rica, El Salvador, Guatemala and Honduras.

← 21. Fine particulate matter (PM2.5) is the air pollutant that poses the greatest risk to health globally, affecting more people than any other pollutant (WHO, 2021[264]).

← 22. Currently, no environmental outcomes other than air pollution can be computed at sub-national level with a harmonised international method (OECD, 2015[96]). Broadening the available environmental indicators is a priority for both Latin American and OECD countries.

← 23. Data on average annual exposure to fine particulate matter per region shown in Figure 3.24are not limited to urban areas, but instead cover all parts of the country (though they are weighted by population, such that rural areas account for a much smaller share of the average estimate in most LAC countries, due to the highly urbanised population). What this means is that average annual exposure can be considerably higher in cities, and in certain locations within cities, but if a large share of the population live in rural areas this will offset the regional average.

← 24. A disaster is a calamitous and sudden event that seriously disrupts the functioning of a community or society and causes human, material, economic and environmental loss that exceeds the capacity of the affected community or society to cope with the situation with their own resources (ECLAC, 2021[266]).

← 25. In this specific sentence, the term “affected” refers to the population that “requires immediate basic assistance, including food, water, shelter, sanitation and medical assistance in a period of emergency caused by a natural disaster. It corresponds to the sum of all injured, homeless and affected people” (ECLAC, 2021[266]).

← 26. In well-being measurement frameworks worldwide, concepts range from proximity to natural areas (Japan, Scotland), perception of accessibility (New Zealand, Australia, Scotland), density (Korea), and number of visits to the outdoors (Australia, Canada, Israel, Scotland, the United Kingdom) (Exton and Fleischer, forthcoming[265]).

← 27. For further information on compulsory voting information from this section, please refer to: https://www.idea.int/data-tools/data/voter-turnout/compulsory-voting.

← 28. The Dominican Republic formally abandoned compulsory voting in 2010, but it was not enforced by sanctions (IDEA, 2021[174]). In the 2020 presidential election, abstention reached 45%, according to the Junta Central Electoral (JCE, 2020[267]).

← 29. Based on the +/- 3.0 p.p. threshold for assessing change over time for this indicator, established in Annex 5.A of How’s Life? 2017 (OECD, 2017[46]).

← 30. Increases in social support were also documented in a random sample of the general Hong Kong population following the 2003 SARS outbreak (Lau et al., 2006[252]).

← 31. Since computer technology may foster a wider network with weak ties, rather than a narrower network with strong ties, its impact on social interactions is likely substantial (OECD, 2019[71]).

← 32. There remains some controversy in the literature about whether self-reported survey measures can be analysed as if they were interval-level data (e.g. (Frey and Stutzer, 2000[254]; Ferrer‐i‐Carbonell and Frijters, 2004[255]; Diener and Tov, 2012[258]; Bond and Lang, 2019[260]; Chen et al., 2019[259])). Much of this controversy centres on the analysis of questionnaire items that use short categorical response scales (e.g. “not too happy/ pretty happy/ very happy”) yielding discrete ordinal level data. This report relies on 0-10 numerical response scales that are intended to convey equal intervals to respondents from the outset, and are anchored such that zero refers to the absolute minimum value (i.e. “worst possible”). Responses to these 0-10 numerical response scales are often analysed as if they were interval data (e.g. summarised through mean averages; analysed using OLS regression). While an imperfect representation of the data, the mean is reported here for several practical reasons. First, the mean offers a simple summary of central tendency that can provide an “at a glance” picture of results across a large number of countries and over time (essential for a comparative report). Compared to the median value, the mean is both more sensitive to changes in the distribution of values on a bounded 0-10 scale, and less biased than a median value when it falls at the threshold between two response categories (OECD, 2013[228]). A series of histograms to show the full distribution of responses across 11 response categories for each country, at every time point, would be an ideal representation of the data, but is not a practical option due to space constraints. Imposing binary thresholds on the data (i.e. reporting the share of the population responding above or below a certain threshold value) can be useful for communication purposes and to assess deprivations, but these too involve making strong assumptions about how to carve the distribution into meaningful segments – and, crucially, they can overstate the importance of a difference or change when these occur close to the threshold, whilst overlooking differences or changes that occur in other parts of the distribution (see (OECD, 2013[228])). Finally, reporting the mean average has become common across much of the literature that uses 0-10 life evaluation measures, thereby facilitating comparisons between the results reported here and other studies in this field.

← 33. To put this into perspective, evidence suggests that unemployment has a detrimental effect on life satisfaction by approximately 1 scale point (controlling for individual characteristics) (Wulfgramm, 2014[249]; Voßemer et al., 2017[250]).

← 34. Self-reported measures of objective concepts, such as self-rated health, or self-reported financial difficulty, are not considered within the scope of subjective well-being. While the measurement tool for questions of this sort are self-reports, the subject matter being investigated is not inherently subjective, i.e. it can be observed by a third party. People’s satisfaction with specific domains of life, such as their satisfaction with their financial status or their social relationships, could be considered as subsets of life evaluations – although within the context of the How’s Life? indicator dashboard, they would most logically appear as subjective measures within their respective domains (income and wealth; social connections). What is specific about the concept of subjective well-being is that only the person under investigation can provide information on their evaluations, emotions and psychological functioning – it is people’s own views of their feelings that are the subject of interest (rather than their self-reports of objective phenomena) (OECD, 2020[39]).