Distribution of income and wealth.
The distributive story is one in which global inequality is a result of two forces, between and
within country inequality, and the interplay between them. For almost the entire twentieth
century, today’s high-income countries pulled ahead of other countries; this changed only from
1990 to 2000, when middle-income countries began to catch up. At the same time, inequality
between countries began to fall; and inequality within countries, which had been falling since the
end of the Second World War, turned around and began to increase from the 1980s onwards.
Until the end of the twentieth century, rising inequality within countries dominated and global
inequality rose, but sometime after the turn of the century convergence between countries
became so strong as to bring global inequality down. Recently, the data have shown a fall in
inequality within countries, which nevertheless remains high. The various aspects of this process
will be detailed in this section.
Labour productivity, defined at the country level as GDP per employed person, is a key determinant of income levels. As explained earlier, from the end of the First World War up to a period
between the late 1980s and early 1990s, the world witnessed a growing divergence between
high-income countries and the rest of the world. Before the turn of the millennium, however, growth in low- and middle-income countries, particularly those in Eastern and
Southern Asia, began to pick up. Figures 3.2 and 3.3 show this latter process of convergence
between countries.
Figure 3.2 shows that considerable differences
in labour productivity persist among different
country income groups. The average productivity per worker in high-income countries is
today about 20 times higher than in low-income
countries. The figure also appears to show that
middle-income countries have seen higher
labour productivity growth than high- and
low-income countries, but the differences in
productivity levels and the fact that the baselines are so different make this hard to discern.
To better understand the relative evolution of productivity, figure 3.3 uses an index in which the
productivity of each country group is divided by its 1995 value (it does not show absolute productivity levels but rather its relative change since 1995). Figure 3.3 shows that labour productivity has grown much faster in upper- and lower-middle-income countries than in either high- or low-income countries, with the low-income country group seemingly stuck on a lowproductivity trajectory since 2010 that shows no signs of inflection.
In addition, the productivity
gains have tapered off recently. Growth rates fell in the decade up to 2023 for all country groups.
If the annual growth rates from 1995 to 2013 are compared to those for the decade after 2013,
a deceleration in annual growth rates can be seen in all country groups: from 1.5 to 0.8 for
high-income countries; from 4.5 to 3.5 for upper-middle-income ones; from 3.5 to 2.6 in lowermiddle-income ones; and from 2.2 to 0.6 in low-income ones.
One result of this productivity increase in middle-income countries is a reduction of inequality
in labour productivity between countries (that is, the global inequality that would be observed
if everyone within each country had the average labour productivity of that country) as measured
by the Theil index. This inequality began to fall sometime in the late 1990s or early 2000s, even
if this appears to be tapering off in recent years (see figure 3.4). Using GDP per capita data
instead of GDP per worker (that is, labour productivity) yields almost the same results.
To form a complete picture of inequality, it is necessary to look also at inequality within countries. Inequality within countries had been falling in most countries since after the Second World War until the beginning of the 1980s – however, during that decade it began to increase again. The rise of inequality in high-income countries has been well-documented. After 1988, the formerly socialist countries similarly experienced an increase. The debt crises of the 1980s and 1990s increased inequality in low- and middle-income countries during those decades. In some middle-income countries, such as China and India, growing income inequality within countries coincided with overall economic growth +. Labour income inequality within countries will be looked at in greater detail in section 3.2, after an examining the extremes of the income distribution of the very rich and the extreme poor.
This section will look at the share of income and wealth of the top percentiles. This is a partial view of the income distribution, since it is insensitive to what goes on in the bottom 90 and 99 per cent of the income and wealth distributions. Nevertheless, given the extremely skewed nature of income and wealth distributions today, top incomes are important in understanding inequality. Figure 3.5 shows the following global trends as measured by top incomes: (i) income inequality increases until the late 1990s or early 2000s depending on which measure is used and then begins to fall; and (ii) wealth concentration simply falls (although it is measured only from 1995 onwards).
In the early 1980s, the top 10 per cent commanded about 52 per cent of the total income; this increased to about 56 per cent in the 1990s and then fell to roughly 53 per cent in the most recent year. It is important to recognize that worldwide aggregates shown in figure 3.5 may obscure what is going on within countries and within country groups. Inequality by region may help explain why the fall in inequality as measured by top incomes seems at odds with prevailing perceptions of ever-increasing income and wealth inequalities. Figure 3.6 shows the income shares of the top 10 per cent according to seven world regions. The income shares of the top 1 per cent are not shown, but they follow almost the same trends.
Overall, the high-income regions – Europe, North America and Oceania (including Australia
and New Zealand) – indeed show increasing income concentration of the top 10 per cent. Latin
America shows no clear upward or downward trend, while Africa, Asia and the Middle East show
decreasing income concentration, meaning in the direction of fairer distribution. None of the
regional trends resembles the global trend – which shows how aggregate numbers tell only a
partial story.
At the global level, the story becomes clearer:
falling inequality in productivity and income
between countries is the main driver of global
trends, offsetting inequality within countries
and leading to an overall decline in global
inequality. These numbers also reveal, however,
an unacceptable level of inequality even at the
aggregate level. Clearly, the world has work to
do on these measures of social justice.
One result of unfair distribution of the fruits of progress is the persistence of poverty and hunger.
As the ILO Declaration of Philadelphia recognizes:
the war against want requires to be carried on with unrelenting vigour within each nation,
and by continuous and concerted international effort in which the representatives of workers and employers, enjoying equal status with those of governments, join with them in free
discussion and democratic decision with a view to the promotion of the common welfare.13
This idea undergirds SDG 1 to “end poverty in all its forms everywhere".
Nevertheless, there are still roughly 800 million people below the extreme poverty line, which is used as a proxy for those unable to afford the minimum caloric intake to
remain healthy. Despite significant progress, in the quest to eliminate poverty,
humanity has a long way to go.
Poverty can take different forms. Most relevant to the world of work are income poverty and
working poverty. While income poverty rates measure the share of people living below the
extreme poverty line, the working poverty rate reveals the proportion of the employed population living under this same poverty line despite being employed. The world has made
outstanding progress in reducing both income and working poverty, but progress has slowed
down in recent years. Perhaps the direct consequence of extreme poverty is hunger. While
extreme poverty lines are calculated to reflect the income necessary to satisfy caloric needs, hunger can also be measured directly through its consequences on the human
body (particularly children).
In 1995, 39 per cent of the world’s population lived in extreme income poverty (on less than
US$3.00 per day in constant 2021 PPP-adjusted US dollars). By 2023, the last year for which
worldwide estimates are available, it had fallen by almost three quarters to 10 per cent. It should
be noted that extreme poverty lines cover only food, meaning someone at the extreme poverty
line has no income left for shelter, clothing, transport, healthcare and education, much less any
leisure activities. This is why figure 3.7 also reports the proportion of people living below the
thresholds of US$4.20 and US$8.30 per day. These lines are similarly calculated by the World
Bank and correspond to poverty lines that take into consideration non-food expenditures and
similarly fell from 71 to 47 per cent and from 54 to 19 per cent, respectively. Progress, however, has slowed in recent years: between 2017 and 2022, the share of people
living in extreme poverty fell by only 1 percentage point. While some of this can be attributed to
the COVID-19 pandemic, the slowdown was already visible in the data before lockdowns began.
The trends are similar for higher poverty lines of US$4.20 and US$8.30 per day. As shown by
figure 3.7, progress almost slowed from the mid-2010s across all three measures and even
reversed with the onset of the pandemic. In 2023, 1.5 billion people were living on less than
US$4.20 per day, and 3.8 billion people were living on less than US$8.30 per day.
Extreme income poverty affects most countries around the world. Although low-income countries have extreme poverty rates over 40 per cent, more than half of those living in extreme
poverty actually live in middle-income countries.
Poverty is largely a rural issue, with more than three quarters of the global extreme poor living
in rural areas. The global extreme poverty rate among rural residents is close
to 16 per cent, while among urban residents it is close to 5 per cent. Poverty
rates are almost 20 per cent for agricultural households and less than 5 per cent for non-agricultural households.
Children are also overrepresented among the poor. While extreme poverty rates
among children aged 17 or younger are more than 17 per cent (higher than rural
poverty), for adults over 25, the poverty rate is 7 per cent.
Almost half the world’s poor are children under 15. Extreme
poverty among households without children is close to 2 per cent, but rises to
over 11 per cent for households with one or two children, and to over 22 per cent
for households with three or more children.
The working poor are employed (either as wage or non-wage workers) but live in
households whose income falls below the poverty line. This means their employment and other income is insufficient to lift them and their household out of poverty.
Figure 3.8 illustrates long-term trends in working poverty. It declined rapidly from 2000 until
about 2015, but since then has been stagnant. This faltering progress is clearest in the case of
low- and upper-middle-income countries.
The global working poverty rate is consistently lower than the global poverty rate, meaning
that poverty is less prevalent among the employed than those not employed. Even though
the difference is modest (less than 2 percentage points in 2022), the impact of employment in
lifting people from poverty has increased (the gap was less than 0.4 percentage points in 1991).
As noted earlier, poverty lines are calculated with the intent of revealing households whose
income is insufficient to provide the necessary caloric consumption. However, the relationship
between hunger and poverty is not linear. A more direct way to measure
hunger is its impact on the human body, and particularly on children, which can show the
scars of malnutrition in measures called stunting and wasting. Stunting is when children are
too short for their age, showing a history of hunger. Wasting is when children weigh too little
for their height, showing hunger at the time the measure was taken. Both measures refer to
children under five because their bodies reflect more accurately recent nutritional circumstances.
However, stunting as an indicator better complements the data on poverty, since it is an indicator
of chronic malnutrition, thus reflecting long-term rather than immediate poverty.
The prevalence of stunting fell by more than
13 percentage points from 1995 to 2024, but
still remains at 23 per cent (see figure 3.9). The
stalling that is visible in poverty is also visible in
stunting numbers. The extreme poverty rate is
close to 10 per cent (see figure 3.7), but a full
quarter of the world’s children are stunted, which
means that close to a quarter of the world’s population has endured and is therefore at risk of
extreme poverty that leads to inadequate food
consumption, even if they are not under the poverty line at the time their incomes are measured.
Not surprisingly, malnutrition in infancy is also directly related to labour capacity as an adult.
Physical stunting is also related to cognitive stunting. Longitudinal studies on adults who suffered malnutrition find them to be less productive and the effect is both significant and substantive. Access to food and nutrition remains an issue beyond childhood both for individual human well-being and for healthy and productive workforces.
Poorly fed workers are also unproductive workers, since dietary
deficiencies themselves reduce people’s ability to work. In order to address worker nutritional
needs, the ILO developed the Welfare Facilities Recommendation, 1956 (No. 102), which provides
guidance on the establishment of various forms of food facility services. Extreme hunger and
malnutrition limit sustainable development and can trap people in a cycle of poverty, making
them less productive, more prone to disease and unable to improve their economic conditions
through work. The poverty and hunger story, seen through different lenses, is nevertheless a consistent tale.
It shows strong improvements from 1995 until sometime after the global financial crisis, when
progress begins to taper off. Considerable deficits are still found in low- and lower-middle-income
countries. In 2024, under Brazil’s presidency, the G20 established the Global Alliance againstHunger and Poverty to support and accelerate efforts to eradicate hunger and poverty. The
Alliance builds partnerships and mobilizes financial and knowledge resources to implement policy
to this end. The ILO has joined the Alliance as part of its efforts to advance social justice globally.
The income of most households is overwhelmingly composed of labour earnings. Given the central role that wages play in labour income, especially in countries where wage employment predominates, their distribution is especially important in determining people’s perception of fairness. Accordingly, the wage or labour share of national income and the distribution of wages and labour income are crucial determinants of a fair income distribution. The income aggregate used here is market income, or income before taxes and transfers, rather than disposable income, which is income after taxes and transfers and will be further investigated. The main finding of this section is that the share of labour income in overall national income has been declining on a worldwide aggregate basis, but that labour incomes have recently become more equally distributed among workers.
Income inequality is linked with the labour
income share in total income.
Capital ownership is typically concentrated
among those at the top of the income distribution; hence, a higher share of capital income in
total income leads to more income inequality.
The global wage share of national income as
measured by gross income has been falling
since the early 1990s, losing around 6 percentage points (see figure 3.10). There was a
large fall from approximately 1990 to 2005 and
then a smaller decline after the COVID-19 pandemic. A number of reasons have
been suggested for the fall in the labour share,
including financialization, technological change, global economic
integration, domestic deregulation, the fall in
union density and collective bargaining, the composition
of taxation and the rising cost of housing. Regardless of the nature and
relative weight of these and other factors,
labour today commands a smaller slice of the
pie than before.
However, neither the trend nor the level are uniform across countries. As illustrated by figure 3.11,
the labour income share tends to be higher in high-income countries, meaning that a shift in the
GDP share from high-income to middle-income countries has the effect of reducing the global
labour share, based on world averages.
This shift was responsible for about 35 per cent of the
reduction in the global labour share.
During the last 20 years, the fall in the labour share has been strongest in high-income countries,
where it fell by 2 percentage points. Low- and middle-income countries show increases in the
labour share. There is considerable variation among countries with similar levels of GDP per
capita, whether because of differences in the structural composition of economic activity or
labour market policies. Thus, even as low- and middle-income countries are moving slowly, if not
steadily, towards greater social justice by this measure, high-income countries have lost ground.
The distribution of national income has been tilting away from labour in many countries, which has contributed to the increase of inequality within countries in some parts of the world. However, the distribution of labour income itself has grown more equal. In other words, while in some cases the labour income pie is smaller, it is divided more equally. Moreover, aggregate numbers again obscure important parts of the story, as the reduction in labour income inequality is basically a middle-income country phenomenon. In high-income countries, there is increasing polarization between the highest earners and the lowest; while in middle-income countries, the growth in earnings in the middle of the distribution has lessened polarization, explaining the worldwide numbers. Figure 3.12 shows the ILO modelled estimates series (including measured, imputed and modelled values) as well as a series based only on directly measured observations.
While the levels are different (since they reflect different sets of countries), the trends are roughly the same: aggregate numbers show that labour income inequality within countries, as measured by the Gini coefficient, has been slowly and steadily decreasing since 2004, although with an increase during the COVID-19 pandemic. The Global Wage Report 2024–25 examines these trends and finds comparable results. The average global decline in labour income inequality was not observed in all countries. Figure 3.13 shows the change in the Gini coefficient for this measure from 2004 to 2020 on the horizontal axis (countries in which labour income inequality fell are negative; those in which it rose are positive numbers) as a function of GDP per capita as an imperfect proxy for total income levels. The area of each bubble is proportional to the labour force of each country.
The figure shows that most countries (72 per cent) and most of the world’s workers (79 per cent)
lie to the left of zero, indicating a fall in inequality. It also shows that, while there are rich countries where inequality fell and poor ones in which it rose, the overall trend was towards rising
or stable inequality in higher-income countries and falling inequality in lower-income countries.
Moving from “vertical” inequality between households or workers (as previously analysed), this
section will examine “horizontal” inequality, such as that between men and women or between
different groups in society. Wage differentials driven by factors unrelated to job characteristics,
skill levels, work experience or performance are a well-documented cause of unequal distribution.
The ILO Constitution (1919) commits Member States to the “recognition of the principle of equal
remuneration for work of equal value”. This principle was confirmed in 1944 by the ILO Declaration of Philadelphia, through its fundamental commitment to non-discrimination in all areas of life,
and then again by Article 7 of the Universal Declaration of Human Rights, which stipulates that
“All are equal before the law and are entitled without any discrimination to equal protection of
the law”, while Article 23(2) further specifies that “Everyone, without any discrimination, has the
right to equal pay for equal work”. The Discrimination (Employment and Occupation) Convention,1958 (No. 111), commits countries to fight against “distinction, exclusion or preference made on
the basis of race, colour, sex, religion, political opinion, national extraction or social origin”. The
Equal Remuneration Convention, 1951 (No. 100), commits countries to equal pay for work of equal
value. As discussed before, both instruments are considered fundamental Conventions.Convention No. 111 has been ratified by 175 Member States, and Convention No. 100 by 174.
Although wage gaps have differing historical antecedents, they remain prevalent. This section
summarizes the state of three important types of wage gap: the earnings gap between men and women; the wage gap experienced by migrants; and the wage gap experienced by workers
with disabilities.
Earnings gaps between men and women have proven difficult to close. While calculating the ratio between what employed men and women take home at the end of the day is relatively simple for many countries, others lack data requiring imputations and interpolations. Figure 3.14 shows the evolution of the annual raw earnings ratios by sex for different country income groups. In 2025, the earnings ratio stands at 78 per cent, which means a woman earns 78 cents for every dollar earned by a man. The vertical axis is shown up to 100 per cent, since that represents complete earnings equality between employed men and women.
Focusing on employees alone, pay gaps between men and women based on hourly wages are
considerably narrower than gaps based on monthly wages in all country income groups. This
suggests that hourly pay inequality is often compounded by women’s shorter working hours.
The earnings ratios in figure 3.14 are useful for capturing the overall extent of inequality between
men and women in labour income across the entire employed population (wage employees and
self-employed) but provide limited insight because they do not consider that employed women
tend to have different characteristics compared to men; for instance, in terms of age, educational
and occupational profiles, and different locations on the wage distribution. The Global WageReport 2018/19 calculates a complementary measure – factor-weighted pay gaps – that is, a
pay gap which provides an estimate of the pay gap between men and women employees with
similar educational or occupational characteristics (ILO 2018a). The results are shown in table 3.1.
The monthly global factor-weighted pay gap changes little relative to the raw pay gap. This highlights the limits of classic measures of wage inequality between men and women such as the raw pay gap, especially in countries where women’s labour force participation is low and where women cluster in particular sectors and occupations. Overall, the evidence points to the fact that women suffer from lower pay globally, with the pay gaps resulting from factors such as occupational segregation, imbalances in educational attainments and discrimination in every region. Though the world is generally moving the right direction towards equal pay for work of equal value, the remaining gap is a deficit in social justice that deprives women and women-led families of the opportunity to benefit equally from their participation in the world of work.
Migrant workers also face pay gaps compared to other workers. Unlike pay gaps between men and women, which always disadvantage women, migrant pay gaps are highly dependent on the national context. Highly skilled migrants in countries with a limited skilled workforce may command wage premiums, while others may be part of a marginalized population unable to command the same wages as nationals. Raw migrant pay gaps depend largely on the income level of the country. In high-income countries, migrant pay gaps favour nationals. In middle-income countries, however, they can favour migrants due to differences in occupations and relative skill levels. Of the 48 countries with available data on migrant status and wages, 33 are high-income and 15 are low- and middle-income. In only 7 of the 33 high-income countries do migrants earn more than other workers, but in the 15 low- and middle-income countries, migrant pay is higher in 12. Table 3.2 shows the pay gaps by country income group
In high-income countries, both raw and factor-weighted pay gaps tend to favour host country nationals. Mean and median, monthly and hourly wage levels all favour nationals, and there can be little doubt that migrants are paid less. While migrants often suffer significant occupational downgrading (Brücker et al. 2021), it is clear that in some circumstances, they are also being paid less for similar work. In low- and middle-income countries, the story is more subtle. While raw pay gaps strongly favour migrants, if factors such as education, age and occupation are taken into account, the monthly pay gap flips to the disadvantage of migrants. Overall, however, migrant workers tend to suffer from discrimination in terms of occupational downgrading or are paid less for doing the same work as other workers, especially in highincome countries and in some low- and middle-income countries. Thus, this measure of social justice reveals that more progress towards equal pay is necessary
There is also a significant and persistent pay gap for workers with disabilities. While comparability is imperfect given that different definitions of disability are used in different countries,
figure 3.15 shows that these pay gaps are large
and persistent. What is more, disadvantages
may be cumulative, leading to much lower
wages for women with disabilities.
Furthermore, recent analysis of data in
30 countries suggests that three quarters of
the disability pay gap is unrelated to education, experience or occupation. This means that the gap
is most likely driven by other factors, such as
possible limitations on workplace accommodations for workers with disabilities and wage discrimination.
Despite the challenges in addressing inequalities in income and wages, history shows a number
of proven policies that work if they are applied fairly and consistently. A renewed focus on these
tools can help support reductions in inequality to advance social justice.
Statutory or negotiated minimum wages
serve as key tools to address low pay and
wage inequality, thus promoting social justice
by advancing a fair distribution of the fruits
of labour. Setting adequate levels of minimum wages is challenging and requires an
evidenced-based approach. The Minimum WageFixing Convention, 1970 (No. 131), specifies that
minimum wage levels should take into account both the “needs of workers and their families”
and “economic factors, including the requirements of economic development [and] levels of
productivity”. New Zealand established the first nationwide minimum wage in 1894, and today the vast
majority of workers live in countries in which a minimum wage exists. In 2021, 90 per cent of the 170 countries for which data exist had some kind of minimum wage system, either statutory or negotiated in collective bargaining agreements.
Minimum wage systems differ widely across countries and range from simple to complex:
globally, roughly half the countries that have a statutory minimum wage have a single national
minimum wage rate, while the other half have more complex systems with multiple rates, determined, for example, by sector of activity, occupation, age of the employee or geographical region.
The real value of minimum wages, averaged across countries, increased strongly from 1995 to
2022 (see figure 3.16). In high-income countries, minimum wages almost doubled; in upper-middle-income countries, they more than doubled; and in lower-middle-income countries, they
increased by 88 per cent. However, in low-income countries, they fell by 44 per cent in real terms
on average.
Certain factors can undermine the effectiveness of minimum wages in reducing inequality. The first are legal exclusions: as of 2020, close to 18 per cent of countries (29 countries) with statutory minimum wages legally excluded agricultural or domestic workers. These two categories include many workers concentrated at the bottom of the wage distribution. A second issue is informality: roughly 40 per cent of wage workers are in the informal economy; they constitute the majority of those earning less than the minimum. Finally, minimum wages that are too low and only sporadically adjusted are ineffective in reducing inequality. Nevertheless, most of the literature on minimum wages is clear: minimum
wages are an important tool for reducing wage
inequality. The sustained progress in
real minimum wage levels across many parts of
the world is encouraging. However, the impact
of minimum wage policies remains constrained
by legal exclusions, high levels of informality
and challenges in enforcement. Notably, while
an estimated 57 million wage earners reside in
countries without a statutory minimum wage,
an additional 266 million are covered by minimum wage legislation yet earn less than the applicable minimum.
In recent years, partly due to the lack of adequate minimum wage systems in many countries,
attention has focused on the issue of “living wages”, with a view to ensure that workers earn
sufficient income to afford a decent standard of living. Recalling that wage policies have been a
central subject of the ILO since its creation in 1919 – as reflected in its Constitution, several of its
declarations and international labour standards – the ILO held a tripartite meeting of experts on
wage policies, including living wages. The experts’ conclusions, endorsed by the ILO Governing
Body in March 2024, set out the concept of the living wage as (ILO 2024c):
▶the wage level that is necessary to afford a decent standard of living for workers and
their families, taking into account the country’s circumstances and calculated for the work
performed during the normal hours of work;
▶calculated in accordance with the ILO’s principles on estimating the living wage;
▶to be achieved through the wage-setting process in line with the ILO’s principles on
wage-setting.
Although the estimation and operationalization of living wages come with a number of challenges, the ILO Governing Body agreed that (ILO 2024c, para. 3):
Decent wages are central to economic and social development and essential in reducing
poverty and inequality, as well as in ensuring a decent and dignified life and in advancing
social justice.
Overall, minimum wage and living wage policies, when they are designed and implemented
in a way that considers both the needs of workers and their families and economic factors, as
well as the other ILO principles of wage-setting, can make a real impact on the fair distribution
of the fruits of progress, with consequent benefits for societies and economies.
In addition to being enabling human rights, freedom of association and the effective recognition of the right to collective bargaining have important roles to play in fair distribution. Chapter 1 examined the status of freedom of association and collective bargaining as fundamental principles and rights at work and revealed a deterioration in respect and protection of these principles and rights (see SDG indicator 8.8.2), particularly since the COVID-19 pandemic. This section will examine developments in the institutions that the effective recognition of these fundamental principles and rights enable – that is, trade unions and collective bargaining. It will focus on the consequences for fair distribution. Research on the impacts of unionization has consistently found that unions reduce inequality among their members and, at least in high-income countries, higher unionization rates reduce overall earnings inequality. The decline in unionization had a large impact on wage inequality in the United States, responsible for about 14 per cent of the increase in wage inequality from 1979 to 1988, and in Mexico, where it was responsible for roughly 6 per cent of the decline in inequality from 1984 to 1996. However, there is considerable heterogeneity in the effect of unionization on income inequality across countries. The consensus among researchers is that collective bargaining leads to more egalitarian wage distributions. First, collective agreements raise wage floors, where most trade union members tend to cluster, thus compressing the wage structure. Second, collective agreements standardize wage rates, thus reducing the effects of individual characteristics on the variation of wages. Among high-income countries, those countries with coordinated bargaining systems where a high proportion of workers are covered by collective agreements have lower wage inequality compared to those with fully decentralized bargaining systems. The collective bargaining coverage rate – that is, the share of employees whose terms and conditions of employment are determined by a collective agreement – varies from more than 99 per cent to under 1 per cent, depending on the country. Globally, only one third of employed workers have their terms and conditions of employment determined by a collective agreement. As a key instrument for fair distribution, collective bargaining is underutilized. There are ways, however, “to encourage and promote the full development and utilisation of machinery for voluntary negotiation between employers or employers’ organisations and workers’ organisations, with a view to the regulation of terms and conditions of employment by means of collective agreements.” Various factors determine the proportion of workers covered by collective agreements as well as the extent to which those agreements help reduce wage disparities. The first and most important factor is of course the unionization rate, which is strongly linked to the share of workers covered by collective agreements (correlation coefficient of 0.6). Unfortunately, average unionization rates for wage and salary earners fell from 20.0 per cent in 2008 to 16.8 per cent in 2019. There are several reasons for this, including new forms of working arrangements, the shift from manufacturing to services, non-compliance with the right to organize and the decentralization of collective bargaining.
The decline in unionization impedes the distributional
effects that unions might otherwise have on the wage distribution.
A second factor that determines the share of workers that are covered by a
collective agreement, and thus its potential effects on the wage distribution,
is the institutional setting for collective bargaining. Collective bargaining may
be carried out on a multi-employer basis or by a single employer at the firm
or establishment level. In some countries, the institutional setting may depend
upon the sector, meaning that the country has both multi-employer and single
employer bargaining – a mixed system. Not surprisingly, the most inclusive and
coordinated form of bargaining is multi-employer bargaining, typically carried
out at the sectoral or inter-professional level. These institutions
have the highest rates of coverage by collective agreements at 69.4 per cent on
average. They also allow for coordination between employers who are members
of employers’ organizations. The coverage levels in countries with only enterpriselevel and mixed-level bargaining are much lower at 15.4 and 31.9 per cent on
average, respectively.
The third factor is whether is whether workers in the public sector are able to engage in
collective bargaining. Countries where more than 75 per cent of workers are covered by
collective agreements are also those that guarantee the right to collective bargaining to
public servants. Given that women are more likely to be employed in lower-paid sectors such
as healthcare and childcare services, collective bargaining in public services can also have
important implications for reducing pay gaps between men and women and advancing equal
pay for work of equal value.
Finally, how collective agreements apply to workers and employers is crucial in determining the
coverage of a collective agreement and hence its effects on compressing the wage distribution
and reducing wage disparities, for example between fixed and temporary workers. Erga omnes
provisions may exist in law, making a collective agreement applicable to all workers in a bargaining unit – irrespective of whether they are a member of the trade union that negotiated
the agreement. In addition, extension policies – according to which a public administration may
decide to extend a collective agreement to all employers within a determined jurisdiction based
on certain pre-determined criteria – as set out in the Collective Agreements Recommendation,1951 (No. 91) – can be an important means of
shoring up the coverage of collective agreements. It can also bring temporary workers
and workers in small and medium enterprises
within the scope of a collective agreement, while
ensuring that public authorities give due consideration to the potential effects of such extension
on employment.
Social security is a significant mechanism to address inequalities. Before we identified the right tosocial security as a fundamental human right that provides the foundations for the development
of social protection systems. This section will look at how social security has developed from
universal aspiration to legislative reality and gauge its gradual and continuing expansion. While
recognizing that health is also an important part of social security systems,
this section will look primarily at the importance of social security transfers in fair distribution,
including child benefits, unemployment protection and pensions.
Figure 3.17 shows how the social security legislation has evolved since 1900. It has increased
from little or no legal coverage for any social risk to a situation where over 90 per cent of
countries have legislation in place for old age, disability, sickness, maternity and work injury.
Child benefits and unemployment benefits have lagged somewhat behind but the increases
are still impressive.
Figure 3.18 shows the increase in effective (not only legal) coverage from 2015 to 2023 for various
life cycle contingencies. Whereas figure 3.17 shows countries with laws on the books guaranteeing a given benefit, figure 3.18 captures people who are actually: (i) either receiving benefits;
or (ii) contributing to systems that are reasonably sure to guarantee them these benefits.
The most important takeaway from figure 3.18 is that for the first time in history, more than
half (52 per cent) the world’s population is covered by at least one social protection scheme. This is up from 43 per cent in 2015. Yet, this also means that almost half the world
is still entirely excluded, and many more are only partially covered. Disaggregated
by type of protection, the highest rate of coverage is that of older persons, 79 per cent of whom are covered. Coverage of persons with disabilities, children, mothers with newborns
and work injury is lower, ranging between 28 per cent (for children) and 39 per cent (for
persons with severe disabilities) in 2023. Unemployment protection remains low with a global
coverage rate of 17 per cent.
Figure 3.19 breaks down this trend by level of economic development. Coverage of almost all
types of benefit is high in high-income countries. However, middle-income countries
have had noticeable increases in the past several years. The proportion of the population covered
by at least one type of benefit increased by
14 percentage points in upper-middle-income
countries and 11 points in lower-middle-income
countries, compared to 5 points in high-income
countries. Low-income countries, on the other
hand, have been largely left behind, with both
very low coverage rates and little improvement
in those rates.
A gender gap exists in social security coverage. While 55 per cent of men are covered
by at least one benefit, only 50 per cent of women are covered. Gender gaps are also apparent
in coverage by contributory pensions. While 41 per cent of working-age men contribute to social
security, only 29 per cent of working-age women do so, in large part due to lower labour force
participation. While non-contributory pensions can close coverage gaps, pension adequacy
remains a challenge.
Social security transfers will be increasingly important as the world navigates the three societal
transitions that are the focus later. The first thing that many workers will need when
losing their jobs or otherwise coping with climate, digital or demographic transitions is income
support.
Figure 3.20 shows the reduction in income inequality brought about by redistribution through
social transfers. This reduction is calculated using two household income distributions: market
incomes and disposable incomes. Market income is the income derived from the remuneration
of the factors of production – labour income and capital income (although the latter is typically
not well measured in household surveys). In the OECD and Commitment to Equity (CEQ) Institute
methodology, used in the chart, contributory pensions are considered market income as well,
categorized as deferred earnings or forced savings. Disposable income is the income after taxes
and government transfers, apart from contributory pension schemes, which are considered
market income.
The difference between the two captures the capacity of social transfers and tax systems to reduce income inequality, thus reinforcing fair distribution, shown in figure 3.20. Figure 3.20 shows that the median high-income country succeeds in reducing inequality by 35 per cent through social transfers and progressive taxation. The variation is also large, going from a 4 to 47 per cent reduction in income inequality. Upper- and lower-middle-income countries also show a wide range of inequality reductions; however, in the median they do much worse than high-income countries, which suggests that policies may play an important role. Low-income countries simply do not have the resources to bring about large reductions in inequality through redistribution. Social security expenditures are typically financed through taxes and social security contributions, meaning that both sources of revenue, as well as other government spending obligations such as external debt, help determine how much the social security system can reduce inequality, particularly for the most vulnerable.
As recognized by the Fourth International Conference on Financing for Development in 2025,
fiscal capacity is an important determinant of the capacity of national social protection systems
to reduce inequality. From the relationship shown in figure 3.21 and the fact that social security
transfer expenditures constitute a significant share of public expenditures in most countries,
it becomes clear that, without adequate fiscal space, social security systems cannot effectively
function. While this fiscal space may be created through economic growth, improved collection
of taxes and social security contributions, as well as debt relief (particularly for low-income
countries), it remains crucial to ensure the delivery of adequate benefits and contribute meaningfully to a fairer distribution to advance social justice.
This need not necessarily be done through
increases in tax rates, as most countries could
improve collection and reduce tax avoidance
and evasion. Almost as
important as the volume of taxes and contributions collected is their composition. Indirect
taxes, such as sales taxes or value-added taxes,
are regressive, as their impact falls disproportionately upon the poor. Property taxes and direct
taxes on the incomes of individuals and corporations tend to be more progressive, since they
fall typically upon those in the upper end of the income and wealth distributions.
Figure 3.23 shows these impacts. Figures 3.21 and 3.22 show that current levels of expenditure are insufficient to significantly reduce inequality in most of the world. This would require closing coverage and adequacy gaps in social protection, including social security. Using data on 133 low- and middle-income countries, the ILO has estimated the financing gap to ensure a social protection floor for all – including access to at least a basic level of income security. Not surprisingly, the financing gaps are larger for lower-income countries. For upper-middle-income countries, closing the financing gap for a social protection floor (excluding healthcare) would require an additional 0.7 per cent of their GDP, for lower-middle-income countries 2.3 per cent of their GDP and for low-income countries, 19.8 per cent of their GDP. Given the low tax capabilities and high debt burden of low-income countries, this gap cannot be bridged without significant international financial cooperation to support national efforts in progressively enhancing their fiscal capacities. Social transfers and the taxes and contributions that finance them must be part of the strategy to keep inequality in check, addressing both the primary distribution and disposable incomes. A more equal “market” distribution, resulting from stronger labour market institutions and other variables, reduces the need for transfers and the taxes that finance them. Nevertheless, in any society, there will always be those unable to make ends meet through their labour alone, thus requiring a comprehensive and adequate social protection framework for fair distribution.
Social justice is “concerned with fairness in distributional outcomes including a just share of
the benefits of economic growth”. This involves the fair sharing of productivity gains, or “a just share of the fruits of progress” in the language of the ILO Declaration of
Philadelphia. The question of whether there is a just share of productivity gains is in essence
an analysis of inequality – especially income inequality, between and within countries. Inequality
between countries has been falling since the 1990s or early 2000s (depending on the measure),
resulting from an increase in productivity in middle-income countries. This convergence in labour productivity between countries is well known, but its magnitude is sometimes not entirely
appreciated. With regard to labour incomes specifically, a recent reduction of inequality within
countries can be observed.
In almost three out of four countries, the fruits of progress (as
measured by labour incomes) have become more equally distributed since the 2010s, although
the COVID-19 pandemic partially arrested this development.
Nevertheless, the story is not a simple, linear tale of falling inequalities. Low-income countries
have fallen behind middle-income ones in productivity growth, showing that the convergence
in labour productivity does not extend to everyone. The reduction in labour earnings inequality
within countries has also been a largely middle-income country phenomenon, with some highincome countries seeing little change or even increases in labour income inequality. Even so, the
labour income share has been declining in many countries.
Most of all, reductions in inequality both between and within countries are incipient, and the remaining gaps remain significant. Statutory minimum wages have an important role to play in reducing labour income inequality. Collective bargaining – a proven strategy for achieving a fair distribution – has been used to negotiate collective agreements that cover around a third of employees, and its potential thus seems underutilized. Addressing gender, migrant and disability pay gaps requires enforcement of anti-discrimination tools and other policies to ensure fair pay and integration into the workplace – much of which has been discussed earlier. This chapter has also shown the importance of social protection, including social security, in significantly reducing inequalities. There is a need to invest in the adequacy, coverage and sustainability of social security systems. History has given the evidence needed to promote the fair distribution pillar of social justice. The data are clear and the tools are known – they just need to be used.
Comments
Post a Comment