This chapter assesses current well-being outcomes for different parts of the population and whether existing gaps narrowed or widened over the past decade. Well-being gaps by gender, age and educational attainment remain substantial. Compared to around 2010, outcomes have generally improved across the board, and gaps between population groups have slightly narrowed for aspects of work and job quality, safety, and in some cases trust in government. However, narrowing gaps in subjective well-being and the quality of social connections have been driven mainly by outcomes for younger people, and, in the latter case, also by outcomes for men, declining to a comparatively larger extent than for other population groups.
How's Life? 2024
Well-being and Resilience in Times of Crisis
How's Life?
Abstract
It is essential to understand not only whether life has been getting better or worse as a whole, but also, for whom. In many cases, inequalities in well-being between population groups run deep (OECD, 2020[1]; 2017[2]). This chapter considers gaps in well-being outcomes that can be compared across countries: by gender, age and educational background (Box 3.1).
Well-being gaps for different parts of the population
This part of the report presents how men and women, people of different ages, and those with different educational achievements are faring when it comes to current well-being.1 For each group, gaps vis-à-vis the population average are presented for the latest available year across the thematic areas of material conditions, quality of life and community relationships (as well as some aspects of social capital). This chapter focuses on results for the OECD average; more details about well-being inequalities for each OECD country are available in the online country profiles accompanying this report.
How to interpret population breakdowns
Age ranges differ according to each indicator and are only broadly comparable – they generally refer to 15-24/29 years for younger people, 25/30 to 45/54 years for the middle-aged, and 50 years or over for older people. Full details for each indicator can be found in the How’s Life? Well-being Database (OECD, n.d.[3]) metadata information available at https://www.oecd.org/wise/oecd-well-being-database-definitions.pdf.
People with only a primary level of education tend to make up only a small share of the total population in most OECD countries, meaning that statistically robust data (with adequate sample sizes for this educational category) are available only for a limited number of indicators. This chapter hence focuses on people who have completed either upper secondary or tertiary degrees as their highest educational level. Those with only primary education are nevertheless included in the calculation of the total population average.
Change over time
The indicators discussed here are restricted to those for which time series are long enough to show trends over time (for a discussion on well-being inequalities for which no time series exist, such as considerable gender gaps in unpaid work, see (OECD, 2020[1])). For some indicators, estimates were pooled across years to boost the sample size (particularly for outcomes for younger people). In order to standardise the approach across the dashboard, trends over time with a few exceptions hence refer to changes from around 2010 until the latest available year, rather than to those that occurred during the COVID-19 pandemic or cost-of-living crisis.
As it is important to understand why gaps might have narrowed or widened (i.e. a reduction in inequality might have been the result of outcomes worsening for one group, which is not a positive development for well-being overall), summary figures throughout also show trends from 2010 up to the latest available year specific to each population subgroup.
Intersectionality
Due to internationally comparable data availability and sample size constraints, the indicators discussed in this chapter refer to outcomes for specific single population groups, whereas an intersectional approach to data would identify well-being inequalities within and between groups based on how multiple factors of a person's identity come together (e.g. gender, age, socioeconomic background, racial and ethnic background, immigration status). Nevertheless, several national well-being monitoring initiatives in OECD countries are increasingly highlighting the importance of intersectional analyses and are strengthening data collection in this area (Statistics Canada, 2024[4]).
Note:
1. Gender-diverse data is not included in the How’s Life? well-being database, due to gaps in internationally comparable information, although several countries are increasingly collecting metrics.
Across OECD countries, men and women each face specific well-being challenges (Figure 3.1). On the one hand, compared to the population as a whole, women are on average less likely to be employed and tend to fare worse on a range of quality-of-life aspects (e.g. they perceive their health as worse, are more likely to report more negative than positive emotions, including physical pain, and feel more lonely and less safe in their neighbourhoods at night). On the other hand, men live shorter lives on average, are more likely to report very low satisfaction with personal relationships and are considerably more likely to work longer hours in paid work or to die as a victim of homicide or from a death of despair (due to suicide, drug overdose or acute alcohol abuse) (Figure 3.2).
Looking at changes in the gender gap in the last decade, across the 22 well-being indicators illustrated in this chapter, gender gaps have narrowed in more than half, widened in just two and remained stable for just over one-quarter (Figure 3.1). As it is important to understand why gaps might have narrowed or widened, both the direction of the gap and the direction of trends for all groups need to be considered jointly.
Several well-being outcomes have improved for both genders over the past decade, but more so for the comparatively more disadvantaged group in each respective instance (Figure 3.1). This has led to narrowing gender gaps for some aspects of work and job quality and safety, where women are catching up with men, such as the employment rate, the share of young people not in education, employment or training (NEET) and feelings of safety when walking alone at night. Meanwhile, men are catching up with women when it comes to fewer people working very long hours in paid work. Both men and women have also seen improvements since 2010 in the long-term unemployment rate, deaths due to assault as well as life expectancy, though this has not led to a narrowing of gender gaps overall.
However, in other instances progress has been more mixed (Figure 3.2). For example, gender inequalities in deaths from suicide, drug abuse or acute alcohol abuse have decreased as a result of fewer male fatalities, while reductions in these deaths of despair have essentially stalled for women over the past decade (and increased in the short-term, as discussed later in this section). Similarly, narrowing gender gaps in the cases of feelings of worry, pain and loneliness were due to these outcomes worsening in particular for men.1 OECD average student skills in mathematics worsened for both boys and girls over the past decade, with no clear change in the gender gap. These developments are described in more detail in the remainder of this section.
Current performance compared to the population average, progress for each group and change in gender gaps since 2010 until the 2023 or the latest available year, OECD average
Note: The first row (“group doing comparatively better”) details for each well-being outcome which population group (in this case, men or women) is doing comparatively better relative to the population average for the latest year available. It is classified as “no clear difference” when all groups are within ±0.03 points of distance to parity with the population average. The second and third rows show the direction of progress made since 2010 until the latest available year for each population group: ↑ signifies improvement, ↓ signifies deterioration and ↔ signifies no clear change with respect to indicator-specific thresholds for meaningful change (see the Reader’s Guide for more details). A shaded cell means the respective change over time for one group was comparatively larger than for the other group. The last row “gender gap change” shows the development of the ratio between population groups (in this case, men and women) since 2010 until the latest available year. It is classified as either widening or narrowing if a change of at least ±0.01 points in the ratio between groups has occurred for a specific indicator. Throughout, the compared periods refer to 2008-10 and 2022-23 for feeling safe at night, all affect indicators, perceived social support and volunteering. Indicators marked with * denote that no longer term data were available and change over time refers to 2018 until the latest available year.
Source: OECD calculations based on the OECD How’s Life? Well-being Database (n.d.[3]), http://data-explorer.oecd.org/s/fu.
Distance from parity with the population average, by gender, 2023 or latest available year, OECD average
Note: Values refer to the ratio of outcomes for a specific population group relative to the population average. Indicators are scored relative to their direction for well-being: for example, compared to the population average, women are 3.42 times less likely to die due to assault, while men are 1.1 times more likely to feel safe when walking alone at night in their neighbourhoods. Ratios that fall within ±0.03 points of parity are classified as no clear difference from the population average (indicated by the grey shaded area). In order to show the gap size for all indicators, ratios that fall outside of the figure’s scale are set to 0.5 or 1.5 and are marked by << or >> with the actual ratio value indicated in the bubble.
Source: OECD calculations based on the OECD How’s Life? Well-being Database (n.d.[3]), http://data-explorer.oecd.org/s/fu.
Material conditions by gender
Gender gaps in material conditions have generally narrowed over the past decade.
Men across OECD countries clearly fare better when it comes to the majority of labour market outcomes - relative to the population average, they are more likely to be employed as well as less likely to be long-term unemployed, and younger men are also less likely not to be in education, employment or training (NEET) (Figure 3.2). However, over the past decade, women started to catch up, partly driven by a stronger recovery in labour market outcomes over the course of the COVID-19 pandemic (OECD, 2023[5]). For instance, the OECD average employment rate for women increased by 7.8 percentage points between 2010 and 2022 (from 64.2% to 72%), compared to an increase of 4.7 percentage points for men (from 79.4% to 84.1%). Similarly, gender gaps have narrowed for the average OECD NEET rate, which stood at 12% for women and 11% for men in 2022, compared to 15% and 13% in 2010, respectively (OECD, n.d.[3]). The average gender wage gap across OECD countries (not shown in Figure 3.1), while still at 11.6% in 2022, also clearly dropped (from 14.5% in 2010) (OECD, n.d.[3]). Only the long-term unemployment rate, which has decreased from an OECD average of 3.5% to 1.6% for men and from 3.2% to 1.9% for women has seen slightly better trends for men over this period, leading to a flip in the gender gap for this indicator (Figure 3.1).
Close to 10% of male employees across OECD countries continued to spend at least 50 hours a week in paid work in 2022, versus only 4% of female employees (OECD, n.d.[3]) (Figure 3.2). This gap has been narrowing over the past decade, driven by reductions in long working hours for both genders but to a larger degree for men (who recorded a drop of 3.4 percentage points since 2010, compared to 1.2 percentage points for women) (OECD, n.d.[3]). Nevertheless, once both paid and unpaid work are taken into account, women work more than men – on average almost 25 minutes a day – in almost every OECD country (OECD, 2020[1]).
Quality of life by gender
Unlike gender gaps in material conditions, which have all narrowed over the past decade, gender gaps in quality of life have shown less consistent improvement over time.
Gender differences in the dimension of safety continue to be stark, but since 2010 outcomes have generally improved across the board and the gender gap in perceptions of safety has narrowed: in 2022-23, 68% of women and 82% of men felt safe walking alone at night, a significant improvement relative to 2008-10 (57% and 77%, respectively), with women making more rapid gains than men over time (OECD, n.d.[3]). In parallel, deaths due to assault reduced in absolute terms more dramatically for men (by 1.9 deaths per 100 000) than for women (by 0.4 deaths per 100 000) between 2010 and 2021 (OECD, n.d.[3]). However, men remain at much higher risk of homicide overall (at 6.3 vs 1.1 deaths per 100 000, and the gender gap has actually widened over time, since the relative change since 2010 was slightly larger for women) (OECD, n.d.[3]) (Figure 3.2).
When it comes to health outcomes, in 2022 women continued to have a higher life expectancy than men (at 83.3 years vs 78.1 years, with gender gaps slightly narrowing due to comparatively larger improvements for men since 2010) (OECD, n.d.[3]).2 Women also remained less likely to rate their health as good (64% of women vs 69% of men), with gender gaps over the past decade largely unchanged. (OECD, n.d.[3]). Deaths of despair represent the second-largest level of gender inequality among all current well-being outcomes – in 2021, men in the typical OECD country were almost three times more likely than women to die from either a suicide, drug overdose or acute alcohol abuse (Figure 3.2). Gender differences in deaths of despair have shrunk, but this was almost entirely driven by falling deaths rates for men from 40.5 deaths per 100 000 in 2010 to 35 in 2021. Over the same time period, deaths of despair for women remained relatively stable (at 13.3 deaths per 100 000 population in 2010 and 12.6 in 2021) (OECD, n.d.[3]). Of particular note, deaths of despair for women have started to rise in recent years (by 0.3 deaths per 100 000 between 2019 and 2021), while male fatalities continued to fall (by 0.5 deaths) over the same period (OECD, n.d.[3]).
Women across OECD countries are slightly more likely than men to experience a lot of worry, physical pain or more negative than positive feelings the previous day (Figure 3.2). Gender gaps in feelings of worry and pain have narrowed relative to 2008-10, as outcomes have worsened across both genders, but particularly so for men. For instance, 34% of men in the typical OECD country reported feeling a lot of worry in 2022-23, compared to 31% around a decade ago. Worry also increased for women over this period, but by a smaller amount (from 37% to 39%).3 The prevalence of pain across OECD countries also increased for both genders, with men up from 22% in 2008-10 to 25% in 2022-23, and for women up from 27% to 31% (OECD, n.d.[3]).
Lastly, there is no clear gender difference between male and female students vis-à-vis the population average for maths and science skills at age 15 (as assessed by the OECD 2022 PISA study). Since levels of performance have worsened since 2018 for both genders, this gap has remained consistent over time (Figure 3.1).Trends for test scores in reading and science have followed similar patterns (OECD, 2023[6]).
Community relationships and social capital by gender
Gender differences in community relationships and social capital also show diverging patterns depending on the indicator considered.
Across OECD countries, there is little difference in how satisfied men and women are with how they spend their time, and this has remained stable over the past decade (Figure 3.1, Figure 3.2). Interestingly, outcomes for both genders have moved in tandem since 2013: on a scale ranging from 0 (no satisfaction) to 10 (complete satisfaction), mean values for both men and women first deteriorated by more than 0.2 points between 2012 and 2018, and then improved by around 0.4 points over the course of the COVID-19 pandemic and associated changes in work patterns (OECD, n.d.[3]). In 2022, the mean score of satisfaction with time use across OECD countries was 7.3 for both genders (OECD, n.d.[3]).
Overall, women spend more time than men interacting with friends and family – around 20 minutes more per week (and men around 20 minutes less), while the OECD average was six hours per week around 2018 (OECD, 2020[1]).4 However, with regard to the quality of social connections, women are more likely than men to feel lonely (Figure 3.2). Since 2018, this gender gap has narrowed – largely due to feelings of loneliness rising more for men than for women during COVID-19 (up from 4.8% to 5.4% in 2023 for men, and from 6.5% to 6.7% for women) (Figure 3.1) (OECD, n.d.[3]). During the pandemic, the share of men being very dissatisfied with their personal relationships also rose at a higher rate than it did for women, leading to a widening gender gap over time, and one in which men fare worse than women (Figure 3.1, Figure 3.2) (OECD, n.d.[3]). 5
In terms of social capital, trust in national government and volunteering rates are very similar for men and women across OECD countries on average, and changes over the past decade have not been large enough to be classified as meaningful for either gender (Figure 3.1, Figure 3.2). In 2022-23, 49% of women and 48% of men expressed trust in their government, and 22% of women and 23% of men declared they volunteered in the past month (OECD, n.d.[3]).6 Meanwhile, there has been encouraging progress towards gender parity in politics: the share of women members of parliament (not shown in Figure 3.1) increased from around one-fourth of seats in 2010 to just over one-third of seats in 2023, and two countries (New Zealand and Mexico, the latter with the support of quotas) have achieved parity in political decision-making (OECD, n.d.[3]).
People of different ages have different comparative well-being advantages – generally speaking, younger people across OECD countries tend to do relatively better when it comes to health, subjective well-being (apart from feeling angry) and some aspects of social connectedness, whereas middle-aged adults are more likely to be employed and feel safer, and older people trust their government more and are less likely to experience anger (Figure 3.3, Figure 3.4). As some of these differences can partly be a function of age itself, such as differences in health, the nature of career trajectories, or as a result of having experienced major life events, it is especially important to examine whether age gaps are narrowing or widening over time, rather than only the absolute level of inequalities per se.
Since around 2010, labour market conditions, perceptions of safety and trust in government have improved across all age groups. In the majority of cases, older people experienced greater improvements than younger people, with the exception of long-term unemployment and very long working hours (Figure 3.3). Perceived health represents one of the areas with diverging trends across age groups, with outcomes improving since 2010 for older people but worsening for the middle-aged (Figure 3.3).
Going forward, it will be important to keep an eye on how younger people are faring with regard to quality of life and community relationships. Gaps between younger and older people have generally narrowed in aspects of subjective well-being and social connectedness over the past decade. However, this was because younger people experienced the largest comparative declines in these aspects of their lives (although changes over time were in most cases not yet large enough to be considered as a clear improvement or deterioration for any age group) (Figure 3.3). These developments are described in more detail in the remainder of this section.
Current performance compared to the population average, progress for each group and change in age gaps since 2010 until 2023 or the latest available year, OECD average
Note: The first row (“group doing comparatively better”) details for each well-being outcome which population group (in this case, younger, middle-aged and older people) is doing comparatively better relative to the population average for the latest year available. It is classified as “no clear difference” when all groups are within ±0.03 points of distance to parity with the population average. The second, third and fourth rows show the direction of progress made since 2010 until the latest available year for each population group: ↑ signifies improvement, ↓ signifies deterioration and ↔ signifies no clear change with respect to indicator-specific thresholds for meaningful change (see the Reader’s Guide for more details). A shaded cell means the respective change over time for one group was comparatively larger than for the other group. The last two rows show the development of the ratio between population groups (in this case, between younger and middle-aged and between younger and older people) since 2010 until the latest available year. It is classified as either widening or narrowing if a change of at least ±0.01 points in the ratio between groups has occurred for a specific indicator. Throughout, compared periods refer to 2010-16 and 2017-23 for feeling safe at night, all affect indicators, perceived social support and volunteering. Indicators marked with * denote that no longer-term data were available and change over time refers to 2018 until the latest available year. Age ranges differ according to each indicator and are only broadly comparable – they generally refer to 15-24/29 years for younger people, 25/30 to 45/54 years for the middle-aged, and 50 years or over for older people. Full details can be found in the How’s Life? Well-being Database metadata information.
Source: OECD calculations based on the OECD How’s Life? Well-being Database (n.d.[3]), http://data-explorer.oecd.org/s/fu.
Distance from parity with the population average, by age, 2023 or latest available year, OECD average
Note: Values refer to the ratio of outcomes for a specific population group relative to the population average. Indicators are scored relative to their direction for well-being: for example, compared to the population average, younger people are 1.44 times less likely to report experiencing a lot of physical pain, while middle-aged people are 1.05 times more likely to be employed. Values that fall within ±0.03 points of parity are classified as no clear difference from the population average (indicated by the grey shaded area). Age ranges differ according to each indicator and are only broadly comparable – they generally refer to 15-24/29 years for younger people, 25/30 to 45/54 years for the middle-aged and 50 years or over for older people. Full details can be found in the How’s Life? Well-being Database metadata information. In order to show the gap size for all indicators, ratios that fall outside of the figure’s scale are set to 0.5 or 1.5 and are marked by >> with the actual ratio value indicated in the bubble.
Source: OECD calculations based on the OECD How’s Life? Well-being Database (n.d.[3]), http://data-explorer.oecd.org/s/fu.
Material conditions by age
Labour market participation across OECD countries is highest for middle-aged adults (at an employment rate of 82% and a long-term unemployment rate of just under 2% in 2022), followed by older people (at an employment rate of 65% and a long-term unemployment rate of 2%) and then the young (at an employment rate of 41% and a long-term unemployment rate of 2.5%) (OECD, n.d.[3]). Consequently, middle-aged employees are also most likely to work long hours: 8% worked 50 hours or more a week in 2022 (around 2 and 1 percentage points, respectively, more than the corresponding share among younger and older adults) (OECD, n.d.[3]) (Figure 3.4).
Work and job quality conditions across OECD countries generally improved for all age groups since 2010 (Figure 3.3). Younger people in the average OECD country experienced the comparatively highest improvements in several labour market aspects, which narrowed the gaps with people of other ages: at close to 3 percentage points, the reduction in their long-term unemployment rate was more than double that of older age groups. And, although younger people were already the least likely age group to work long hours, over the past decade their share working 50 hours or more a week at work also fell by almost 3 percentage points (about 1 percentage point more than for other ages) (OECD, n.d.[3]). On the other hand, older people saw an average rise of 12.6 percentage points in the share of employed since 2010, more than double that of their younger peers (OECD, n.d.[3]). Since this was partially due to increases in the effective age of labour market exit in OECD countries over the past decade, the impacts on older people’s well-being are likely to depend on individual preferences regarding their desired retirement age (OECD, 2023[7]).
Quality of life by age
Perceptions of safety is the one quality-of-life indicator that has improved for adults of all ages over the past decade. Compared to 2010-16, the share of those feeling safe rose for all ages by at least 3.8 percentage points, and most of all for older people (by 5.4 percentage points) – this led to a narrowing of the age gap between younger and older people and a slight widening of the age gap between younger people and the middle-aged (Figure 3.3). In 2017-23, around 75% of both younger and older people felt safe when walking alone at night in their neighborhoods, while 77% of the middle-aged gave the same answer (Figure 3.4) (OECD, n.d.[3]).
There are large age-gradients in health outcomes, partly because health declines as one ages (Figure 3.4). In 2022, only 42% of older adults reported to be in good or very good health, compared with 65% of the middle-aged and 89% of younger people (these most recent estimates are all close to pre-pandemic levels) (OECD, n.d.[3]). Similarly, a much higher share of older adults (34%) reported experiencing a lot of physical pain in 2017-23 relative to the middle-aged (24%) and the young (19%) (OECD, n.d.[3]). Concerningly, perceived health worsened for the middle-aged (who experienced a drop of 6.2 percentage points since 2010 in the share of people rating their health positively), while it actually improved for older people (who saw a corresponding increase of the same magnitude) (OECD, n.d.[3]). Consequently, the gap in perceived health between younger and middle-aged adults widened, but narrowed compared with older adults (Figure 3.3).
Across the board, younger people across OECD countries experience slightly better affective outcomes than other age groups, and these then generally worsen with age (Figure 3.4). For instance, just over 10% of younger people reported more negative than positive feelings, compared to around 14% for both the middle-aged and older people in 2017-23 (OECD, n.d.[3]). The one exception to this pattern is anger – older people are the least likely age group to say they felt a lot of anger the previous day (13% on average, compared to around 17% for both younger age groups) (Figure 3.4).
For almost all affect indicators, gaps between younger and older ages narrowed over the past decade, but concerningly this was due to the outcomes of younger adults worsening to a comparatively larger degree (Figure 3.3) (OECD, n.d.[3]).7 For instance, the share of younger people saying they smiled or laughed a lot the previous day dropped by 3 percentage points (from 84% in 2010-16 to 81% in 2017-23). The corresponding drop was only 1.3 percentage points for the middle-aged and 0.8 percentage points for older adults over the same period (OECD, n.d.[3]).
Community relationships and social capital by age
Middle-aged people are the least satisfied with how they spend their time (Figure 3.4). In 2022, their mean satisfaction with time use on a scale from 0 “not at all satisfied” to 10 “completely satisfied” was 6.3, compared to 6.9 for younger people and 7.3 for older people (OECD, n.d.[3]). Relative to 2013, mean time use satisfaction scores of younger and middle-aged people each decreased by around 0.1 scale points, while those of older people increased by about the same amount.8 This led to gaps in satisfaction with time use between younger and older people widening (Figure 3.3).
Older people have overall fewer and weaker social connections than younger age groups. In 14 OECD countries with available data, older and middle-aged adults spent on average nearly 2 hours 20 minutes per week less than younger people interacting with friends and family (OECD, 2020[1]).9 In 2017-23, around 90% of older and middle-aged people across OECD countries said they had friends or family to rely on in times of need (compared to just under 95% of younger people). The prevalence of loneliness among older people in 2023 stood at 7.4%, relative to 4.6% for the younger two age groups (OECD, n.d.[3]).
However, younger people are an emerging additional risk group: the narrowing of the age gap in social connectedness outcomes, particularly since COVID-19, has been almost entirely driven by comparatively larger declines for youth (Figure 3.3). Indeed, while changes in social connectedness over the past decade were generally small, younger people across OECD countries often recorded the largest comparative declines, while other age groups were less affected (Figure 3.3). For instance, the share of youth who were dissatisfied with their relationships rose by 0.8 percentage points (from 2.7% in 2013 to 3.5% in 2022), while the corresponding share slightly dropped (by around 0.3 percentage points) for middle-aged and older adults (Figure 3.3) (OECD, n.d.[3]). Similarly, younger people were the only age group for which loneliness in 2023 had risen (by just over 1 percentage point) compared to pre-pandemic levels in 2018 (OECD, n.d.[3]). And, between 2010-16 and 2017-23, the share of people with friends and family to rely on in times of need declined by 0.8 percentage points for youth (compared to a drop of 0.25 percentage points for the middle-aged), while it actually increased by 1.1 percentage points for older people (OECD, n.d.[3]).
Age-related inequalities in social capital, while still apparent, are smaller than for other areas of well-being (Figure 3.4). Volunteering rates in 2017-23 were very similar across all age groups (around 21%) and did not change much compared to 2010-16 (Figure 3.3) (OECD, n.d.[3]). Meanwhile, middle-aged people were the least likely age group to trust their government (at 45% in 2017-23, compared to around 50% of younger and older adults) (OECD, n.d.[3]). Since 2010-16, trust in government improved across the board for all age groups by at least 4 percentage points, and most for older people (at just over 6 percentage points) (OECD, n.d.[3]). It should be noted that the estimates on trust in government by age were pooled over multiple years due to sample size, and hence include the pandemic years. As noted in Chapter 2, figures for the population overall show that the OECD average for levels of trust in government fluctuated over the past decade, with lower scores in the early 2010s and a peak in trust at the onset of the COVID-19 crisis, and now a decline again in 2022-23. This data should be further monitored for different ages once a longer time series becomes available.
Inequalities in current well-being are perhaps nowhere as striking as when it comes to people with different levels of education. Indeed, across OECD countries, people who have completed tertiary education are systematically doing better than those who have achieved up to secondary education, across all the well-being outcomes considered here (Figure 3.5, Figure 3.6). This is the case not only for employment outcomes, for which dividends to education are well established, but also for non-material well-being aspects such as social connectedness and health.
The majority of these inequalities are not only large, but also persistent, and there have been no clear changes in the size of well-being gaps by education for the majority of indicators since 2010 (Figure 3.5). Notable positive exceptions for which outcomes improved for both groups over the past decade are the employment rate, feelings of safety when walking alone at night and trust in government (in the first two cases to a larger degree for people with secondary education).
But, concerningly, while education-related gaps in volunteering and feeling lonely or worried have narrowed since 2010, this was because outcomes stagnated or worsened across the board, and to a larger degree for people with tertiary degrees (though only in the case of the share of people volunteering was the magnitude of change greater than 3 percentage points, and therefore shown in Figure 3.5 as deteriorating). In other cases, in particular perceived health and low satisfaction with personal relationships, the outcomes worsened especially for people with only secondary education, thus widening the gap (Figure 3.5). These developments are described in more detail in the remainder of this section.
Current performance compared to the population average, progress for each group and change in gaps by educational attainment since 2010 until 2023 or the latest available year, OECD average
Note: The first row (“group doing comparatively better”) details for each well-being outcome which population group (in this case, people who have achieved tertiary education and those with only secondary education) is doing comparatively better relative to the population average for the latest year available. It is classified as “no clear difference” when all groups are within ±0.03 points of distance to parity with the population average. The second and third rows show the direction of progress made since 2010 until the latest available year for each population group: ↑ signifies improvement, ↓ signifies deterioration and ↔ signifies no clear change with respect to indicator-specific thresholds for meaningful change (see the Reader’s Guide for more details). A cell shaded in light purple means the respective change over time for one group was comparatively larger than for the other group. The last row “education gap change” shows the development of the ratio between population groups (in this case, people who have achieved tertiary education and those with only secondary education) since 2010 until the latest available year. It is classified as either widening or narrowing if a change of at least ±0.01 points in the ratio between groups has occurred for a specific indicator. Throughout, compared periods refer to 2009-13 and 2019-23 for feeling safe at night, all affect indicators, perceived social support and volunteering. Indicators marked with * denote that no longer-term data were available and change over time refers to 2018-19 until the latest available year. For sample size reasons, people with only a primary level of education are not shown as a separate group in this figure, but are included in the calculation of the total population average.
Source: OECD calculations based on the OECD How’s Life? Well-being Database (n.d.[3]), http://data-explorer.oecd.org/s/fu.
Distance from parity with the population average, by educational attainment, 2023 or latest available year, OECD average
Note: Values refer to the ratio of outcomes for a specific population group relative to the population average. Indicators are scored relative to their direction for well-being: for example, compared to the population average, people with tertiary education are 1.55 times less likely to feel lonely and 1.11 times more likely to be employed. Values that fall within ±0.03 points of parity are classified as no clear difference from the population average (indicated by the grey shaded area). For sample size reasons, people with only a primary level of education are not shown as a separate group in this figure, but are included in the calculation of the total population average. In order to show the gap size for all indicators, ratios that fall outside of the figure’s scale are set to 0.5 or 1.5 and are marked by << or >> with the actual ratio value indicated in the bubble.
Source: OECD calculations based on the OECD How’s Life? Well-being Database (n.d.[3]), http://data-explorer.oecd.org/s/fu.
Material conditions by educational attainment
The labour market dividends to education continue to be clearly visible when examining labour participation rates – between 2021 and 2022, people with tertiary education across OECD countries were 1.1 times more likely to be employed and 1.5 times less likely to be long-term unemployed compared to the overall population (Figure 3.6). While long-term unemployment rates remained stable, the likelihood to be employed rose for all levels of educational attainment over the past decade. It actually did so at a slightly higher pace for people with only secondary education (from 73% in 2010 to 77% in 2022) compared to those with tertiary degrees (from 83% to 86%) (OECD, n.d.[3]). However, this was not enough to change the overall gap between the groups (Figure 3.5).
Quality of life by educational attainment
Among all quality-of-life indicators considered in this chapter, perceptions of safety is the only one in which both groups, people with secondary or tertiary degrees, experienced improvements over the past decade and in which the gap between the groups narrowed (Figure 3.5). In 2019-23, 77% of those with a tertiary degree said they felt safe walking alone at night, compared with 74% of those with only a secondary education. Relative to 2009-13, the share of people feeling safe increased by close to 5 percentage points for those with only a secondary education, around 1 percentage point more than for their tertiary-educated peers (OECD, n.d.[3]).
Trends in how people perceive their health have moved in the opposite direction, and the already substantial inequalities have further widened as the outcomes worsened for people with only secondary education (Figure 3.5, Figure 3.6). In 2022, around 68% of people with at most secondary education said they perceived their health as good or very good, more than 10 percentage points less than people with tertiary education (who on average self-reported rates for good health of close to 80%). Compared to 2010, the share of people with up to secondary education who rated their health positively fell by almost 4 percentage points, while there was virtually no change in outcomes for the more educated (OECD, n.d.[3]).
Education-related inequalities in subjective well-being have largely been persistent, with outcomes for either education group not changing much between 2009-13 and 2019-23 (Figure 3.5). Gaps are most visible when it comes to experiencing either pain or more negative than positive feelings on the previous day, with people with tertiary education around 1.3 times less likely than the overall population to report negative outcomes (Figure 3.6). People with tertiary degrees also have slightly better outcomes in the case of life satisfaction and specific feelings reported (such as anger, worry or laughter) (Figure 3.6). Feeling a lot of worry is the only indicator for which the education gap narrowed over the past decade, driven by a slight increase for people with tertiary education (OECD, n.d.[3]).10
Community relationships and social capital by educational attainment
Education-related differences in how socially connected people feel are clearly visible. In 2023, 6% of people with only secondary education felt lonely, and 3.7% felt very dissatisfied with their personal relationships, figures that were 2 and 1 percentage points, respectively, higher than their tertiary-educated peers (Figure 3.6).11 Since 2018, over the course of the COVID-19 pandemic, the prevalence of loneliness and of dissatisfaction with relationships slightly increased for all education groups (by around 0.5 percentage points) and led to small changes in the respective education gaps, which should be monitored further by policy makers (Figure 3.5) (OECD, n.d.[3]).12
Fewer people across all levels of educational attainment have volunteered in recent years compared to a decade ago. For instance, 20% people with at most secondary education volunteered at least monthly in 2019-23, compared to 23% in 2009-13 (and the respective shares were 25% versus 30% for the tertiary educated) (OECD, n.d.[3]). Given the comparatively larger drop for the more highly educated, the gap between both groups narrowed over this period (Figure 3.5). It should be noted that volunteering rates by educational attainment were pooled over multiple years due to sample size, and hence include the pandemic years – and the decrease in volunteering could have potentially been driven by restrictions that prevented in-person social engagement. However, this is unlikely to be the case: figures for the population overall show that the share of people who regularly volunteer dropped between 2008-10 and 2017-19, before COVID-19, and actually rose since then (see also (OECD, 2021[8])). Once a longer time series becomes available, this trend is also likely to be visible in the pooled educational inequality estimates. Lastly, 50% of those with a tertiary education trusted their national government in 2019-23, compared to 47% of those with only secondary education (OECD, n.d.[3]). Relative to 2009-13, trust in government improved by more than 5 percentage points for both groups, although the overall education gap remained the same (Figure 3.5).
Satisfaction with time use is the only well-being aspect considered here for which those with tertiary education are not strictly better off than those with maximum secondary education (Figure 3.6). In 2022, those with up to secondary education reported an average of 7.0 on a scale of 0 to 10 when asked about how satisfied they are with how they spend their time, compared to a mean score of 6.8 for the tertiary educated. Relative to 2013, outcomes for neither group changed (Figure 3.5).
References
[5] OECD (2023), OECD Employment Outlook 2023: Artificial Intelligence and the Labour Market, OECD Publishing, Paris, https://doi.org/10.1787/08785bba-en.
[7] OECD (2023), Pensions at a Glance 2023: OECD and G20 Indicators, OECD Publishing, Paris, https://doi.org/10.1787/678055dd-en.
[6] OECD (2023), PISA 2022 Results (Volume I): The State of Learning and Equity in Education, PISA, OECD Publishing, Paris, https://doi.org/10.1787/53f23881-en.
[8] OECD (2021), COVID-19 and Well-being: Life in the Pandemic, OECD Publishing, Paris, https://doi.org/10.1787/1e1ecb53-en.
[1] OECD (2020), How’s Life? 2020 - Measuring Well-being, OECD Publishing, Paris, https://doi.org/10.1787/9870c393-en.
[2] OECD (2017), How’s Life? 2017: Measuring Well-being, OECD Publishing, Paris, https://doi.org/10.1787/how_life-2017-en.
[3] OECD (n.d.), How’s Life? Well-being Database, http://data-explorer.oecd.org/s/fu (accessed on 3 May 2024).
[4] Statistics Canada (2024), Gender, Diversity and Inclusion Statistics Hub, https://www.statcan.gc.ca/en/topics-start/gender_diversity_and_inclusion (accessed on 13 August 2024).
← 1. In the case of loneliness, male progress is still below the threshold for what would have been counted as deterioration over time (see the Reader’s Guide for more details).
← 2. In the short-term since 2019 and during the COVID-19 pandemic, life expectancy declined at similar rates for both genders.
← 3. This increase is just below the 3-percentage point threshold considered deterioration for this indicator (see the Reader’s Guide for more details).
← 4. Information on the quantity of social connections is available only from time use surveys, which do not permit the analysis of trends due to data gaps.
← 5. However, changes in loneliness and low satisfaction with personal relationships for either gender were under the indicator-specific threshold of deterioration (of 1.5 percentage points, and 0.5 percentage points, respectively) (see the Reader’s Guide for more details).
← 6. This resulted in the gender gap in trust in government flipping compared to 2008-10.
← 7. However, changes across subjective well-being outcomes and across age groups were generally not yet large enough to be considered as deterioration (see the Reader’s Guide for more details).
← 8. This increase is however below the 0.2-scale point threshold considered deterioration for this indicator (see the Reader’s Guide for more details).
← 9. Information on the quantity of social connections is available only from time use surveys, which do not permit the analysis of trends due to data gaps.
← 10. This increase is just below the 3-percentage point threshold considered deterioration for this indicator (see the Reader’s Guide for more details).
← 11. For sample size reasons for other indicators, results for people with only primary education are generally not discussed in this chapter. However, it should be noted that in the case of social connectedness, disparities are especially large for this group: in 2023, 9.7% of those with only primary education experienced loneliness, and in 2022, 5.3% of those with only primary education were very dissatisfied with their relationships (OECD, n.d.[3]).
← 12. However, changes in loneliness and low satisfaction with personal relationships for secondary- and tertiary-educated people were under the indicator-specific threshold of deterioration (of 1.5 percentage points, and 0.5 percentage points, respectively) (see the Reader’s Guide for more details).