Living conditions in Europe - income distribution and income inequality


Data extracted in January 2019
Planned article update: March 2021

Highlights


In 2017, median disposable income averaged PPS 16 748 in the EU, ranging from PPS 5 239 in Romania to PPS 28 820 in Luxembourg.
In 2017, the 20 % of EU population with the lowest incomes accounted for less than 8 % of total disposable income.
In 2017, the EU median disposable income was almost 70 % higher for people with a high education level compared with those with a low education level.
Median equivalised net income, 2017
(in PPS)
Source: Eurostat (ilc_di03)

This article is part of a set of statistical articles that formed Eurostat’s flagship publication, Living conditions in Europe - 2018 edition. Each article helps to provide a comprehensive and up-to-date summary of living conditions in Europe, presenting some key results from the European Union’s (EU’s) statistics on income and living conditions (EU-SILC) , which is conducted across EU Member States, EFTA and candidate countries.

Full article

Policy context

Gross domestic product (GDP) is a measure of the total output of an economy; from the perspective of living conditions, GDP may also be calculated as the sum of primary incomes that are distributed by resident producer units (in the form of wages, rents, interest and profits). When a country’s population is taken into account, GDP per capita provides both a convenient measure of average incomes and of the living standards enjoyed by the inhabitants living in a specific economy, as well as (when adjusted to take account of price differences between countries — through the use of purchasing power parities (PPPs) — a measure for comparisons of living standards across countries.

Nevertheless, GDP per capita is a relatively simple, aggregate measure and in order to have a more detailed picture of living conditions, it is pertinent to analyse the distribution (rather than average levels) of household income. A number of different statistical measures are available for this purpose, including household disposable income, in other words, the total income that households have at their disposal for spending or saving. While the aggregated level of household disposable income is available from national accounts and might be used for a general analysis of the household sector, this indicator lacks any distributional dimension. It is therefore preferable to base any analysis of income distributions on micro data sources, in other words, statistical surveys for a representative sample of actual households, rather than aggregate macroeconomic measures. Such surveys allow an analysis of median income levels or the distribution of income across socio-economic strata of the population.

In order to take into account differences in household size and composition and thus enable comparisons of income levels, the concept of equivalised disposable income may be used. It is based on expressing total (net) household income in relation to the number of ’equivalent adults’, using a standard (equivalence) scale. Eurostat uses a ’modified OECD scale’ which gives a weight to each member of a household (and then adds these up to arrive at an equivalised household size), taking into account the number of persons in each household and the age of its members (more details are provided in the glossary). Total disposable household income, derived as the sum of the income received by every member of the household and by the household as a whole, is divided by the equivalised household size to determine the equivalised disposable income attributed to each household member.

The median of the equivalised disposable income distribution is typically used in the EU as a key measure for analysing standards of living within each economy. It is simply the income level that divides the population into two groups of equal size: one encompassing half the population with a level of disposable income above the median, and the other half with a level of disposable income below the median. The use of the median (in contrast to the arithmetic mean) avoids any potential distortion that may be caused by the existence of extreme values, such as a few extremely rich households that may raise the arithmetic mean.

Key findings

Median equivalised net income

In 2017, median equivalised net income varied considerably across the EU Member States, ranging from PPS 5 239 in Romania to PPS 28 820 in Luxembourg. The EU-28 average was PPS 16 748. Note, these figures have been converted into purchasing power standards (PPS) — a unit that takes account of price-level differences between countries.

Median equivalised net income fell, in real terms, in 2 out of the 28 EU Member States in 2017 — they were Sweden and Belgium. A decrease was also noted in Switzerland, Norway and Turkey (for these three latest available 2016 data were compared against 2015 data).

Trends for top and bottom 20 % of incomes diverge

Across all 28 EU Member States, the top 20 % of the population with the highest national net disposable incomes (the top quintile) accounted for at least one third of total income, a share that rose highest to 46.0 % in Bulgaria in 2017. By contrast, the bottom 20 % of the population with the lowest incomes together accounted for less than one tenth of all income, except in Czechia (10.3 %) and Finland (10.0 %). Luxembourg recorded the biggest fall in income share (-1.4 %).

The S80/S20 income quintile share ratio is based on a comparison of the income received by the top quintile and that received by the bottom quintile of the population. In contrast, the Gini coefficient measures the extent to which the distribution of income differs between a perfectly equal distribution (where each member of the population has exactly the same income) and full inequality (where a single person receives all of the income).

Distribution of income measured by Gini coefficient

Based on the Gini coefficient, Bulgaria and Lithuania (40.2 % and 37.6 %) experienced the highest levels of inequality in equivalised disposable income in the EU in 2017; note that high coefficients were also recorded in Turkey and Serbia (42.6 % and 37.8 %). The lowest levels of income inequality in EU Member States, using this measure, were recorded in Slovakia, Slovenia, and Czechia (less than 25 %).

Income inequality across different age groups

The EU-28 income quintile share ratio for elderly people (defined here as those aged 65 and over) was lower — at 4.1 in 2017 — than the average ratio for the whole population (5.1). This pattern was repeated across all of the EU Member States, except for Cyprus. Income distribution among the elderly was also relatively unequal (compared with the average for the whole population) in Iceland and Switzerland (2016 data).

Social transfers, the main instrument for the realisation of welfare policies, can play a major role by helping to reduce income inequalities

In 2017, social transfers reduced income inequality among the EU-28 population: the Gini coefficient for income (including pensions) was 51.7 % before social transfers and fell to 30.7 % after taking account of these transfers.

Income distribution

Map 1: Median equivalised net income, 2017
(in PPS)
Source: Eurostat (ilc_di03)


Median disposable income in the EU-28 is still rising

In 2017, median equivalised net income (hereafter referred to as median disposable income) averaged PPS 16 748 in the EU-28. Across the EU Member States, it ranged from PPS 28 820 in Luxembourg to PPS 5 239 in Romania.

Map 1 reveals the highest levels of median disposable (more than PPS 22 000) income recorded in central countries of the EU, in particular in Austria (PPS 23 334) and Luxembourg (PPS 28 820), but also in Norway (PPS 28 875), Switzerland (PPS 27 602) and Iceland (PPS 22 193). By contrast, median disposable incomes were often lower in peripheral countries, levels of income of less than PPS 9 000 were recorded in Hungary (PPS 8 364), Bulgaria (PPS 7 517) and Romania (PPS 5 239). Low values were also recorded in Turkey, Serbia and North Macedonia.

Figure 1: Median equivalised net income by educational level, 2017
(in PPS)
Source: Eurostat (ilc_di03) and (ilc_di08)


Among the population aged 18-64 years, those persons with a tertiary education degree (ISCED levels 5-8) consistently recorded higher levels of median disposable income than those persons who had completed either a low (ISCED levels 0-2) or medium (ISCED levels 3-4) level of educational attainment (see Figure 1).

In 2017, EU-28 median disposable income was almost 70 % higher for people with a high level of educational attainment (PPS 22 626) when compared with the level of income for people with a low level of educational attainment (PPS 13 460). The largest absolute income gaps between persons with low and high levels of educational attainment were recorded in Luxembourg, Malta, Belgium and Germany (more than PPS 10 000); this was also the case in Switzerland. By contrast, the gap in income levels between those people with high and low levels of educational attainment was considerably less in Romania, Hungary, Greece and Slovakia (less than PPS 6 000); this was also the case in Iceland (2016 data), Serbia and North Macedonia (2016 data).


Table 1: Median equivalised net income in national currency before and after adjusting for inflation
Source: Eurostat (ilc_di03), (prc_hicp_aind) and (cpc_ecprice)


Median disposable income - changes over time

The EU-28’s median disposable income in nominal terms (in other words, without adjusting for inflation) rose by 2.3 % between 2016 and 2017 (see Table 1). In all EU Member States the income increased, most in Romania (13.1 %) and Bulgaria (13.9 %). Growth rates were lowest in Italy (1.8%), Finland (1.4%) and Greece (1.3%).

Figure 2: Real change in median equivalised net income in national currency terms, 2016-2017
(Share of total population)
Source: Eurostat (ilc_di03), (prc_hicp_aind) and (cpc_ecprice)


Median disposable income fell in real terms in two EU Member States.

After adjusting for inflation (using the harmonised indices of consumer prices (HICP)), the development of median disposable incomes between 2016 and 2017 was relatively similar (which may reflect the still low levels of inflation recorded during the period under consideration). In fact, median disposable incomes rose in the EU-28 by 2.1 % in real terms (compared with a 2.3 % increase in nominal terms).

Median disposable incomes fell, in real terms, in 2 of the 28 EU Member States (see Figure 2). A reduction occurred in Belgium (-1.8 %), while the decline observed in Sweden was relatively small (less than 1.0 %). Disposable incomes also fell in Norway (-3.6 %), Switzerland (-1.8 %) and in Turkey (- 4.3 %). For these three countries, the latest available 2016 data were compared against 2015 data.

Between 2016 and 2017, by far the highest increases in real disposable incomes (more than 5 %) were recorded in Romania (13.1 %) and Bulgaria (14.1 %), followed by Croatia (6.6 %).

Table 2: Share of national net disposable income for the first and fifth income quintiles, 2011 and 2017
(%)
Source: Eurostat (ilc_di01)


The top 20 % of earners shared almost two fifths of the total disposable incomes

A more detailed analysis that focuses on income distribution is presented in Table 2. It is based on ordering the disposable incomes of individuals and then dividing these into quintiles (fifths), in other words, the top 20 % of the population with the highest incomes (referred to as the top or fifth income quintile) down to the 20 % of the population with the lowest incomes (referred to as the bottom or first income quintile).

In 2017, some 38.6 % of the total disposable income in the EU-28 could be attributed to people in the top 20 % of the income distribution, while people in the bottom quintile of the income distribution received a 7.8 % share of total disposable income.

In 2017, the top 20 % of the highest earners in Bulgaria, Lithuania, Portugal, Latvia, the United Kingdom, Spain and Greece received more than 40.0 % of the total disposable income within their respective economies in 2017. A similar situation was recorded also in Serbia and Turkey (2016 data). In most of the EU Member States, the share of the top 20 % of highest earners was within the range of 35.0 % - 40.0 %, although this fell to 34.9 % in Belgium, 33.5 % in Slovenia and 32.8 % in Slovakia.

At the other end of the income scale, people in the bottom quintile of the income distribution received less than 7.8 % (which was the EU-28 average) of total disposable income in 12 EU Member States, out of which in Lithuania, Romania and Bulgaria (2016 break in time series) had 6% or less. Only in Finland and Czechia was the share at least ten percent (10.0 % and 10.3 %).

Figure 3: Share of equivalised income by income quintile, 2017
(Share of total population)
Source: Eurostat (ilc_di01)


Decline in the share of total disposable income attributed to the bottom and top income quintiles

Between 2011 and 2017, the share of EU-28 disposable income that was accounted for by the bottom and top income quintiles was lower: The share of total income for those with the lowest incomes fell slightly from 7.9 % to 7.8%, while the share for top earners fell from 38.9 % to 38.6 %.

There were 15 EU Member States that reported a falling share of total disposable income being attributed to the lowest income quintile over the period 2011-2017; note that there was a break in time series in Estonia (2014), Bulgaria and the Netherlands (2016). By contrast, there were 9 Member States where the share of the lowest income quintile rose and four where it remained unchanged. The share of the bottom income quintile increased most in Croatia (0.5 pp), followed by Finland and Poland (0.4 pp each). Large increases were recorded also in North Macedonia (1.8 pp) and Turkey (0.5 pp).

At the other end of the spectrum, in 15 EU Member States the share of disposable income attributed to the top income quintile fell between 2011 and 2017, while the share of disposable income accounted for by the top income quintile rose in 13 Member States. The share of disposable income accounted for by the top income quintile rose by more than 2.0 pp in Luxembourg and Lithuania, with a peak of 4.4 pp recorded in Bulgaria. By contrast, the share of the top income quintile fell by more than 1.0 pp in Croatia, Portugal, Poland, Romania and France, with the biggest decline recorded in Slovakia (- 2.3 pp).

Table 3: Transitions of income within three years by decile
(Share of total population)
Source: Eurostat (ilc_di30c)


Around 15 % of the EU-28 population moved more than one income decile on the income ladder

This next section analyses the share of the population who experience fluctuations in their economic wellbeing from one year to the next, by studying the proportion of people that move up/down the income ladder receiving a higher or lower level of income.

The analysis is based on how people’s disposable income moves over a three-year period in relation to a set of income deciles — these are similar to income quintiles but instead of ranking the disposable incomes of individuals and then dividing these into fifths, the ranking is divided into tenths; as such, the highest decile refers to the top 10 % of the population with the highest incomes. It is important to note that upward or downward income transitions may occur as a result of direct changes in an individual’s financial situation (more or less income), but may also result from aggregate changes across the whole economy. For example, an individual may see their income frozen, while there is a more general increase in incomes across the remainder of the population and as a result that individual may move to a lower income decile (even if their income remains unchanged). It is also important to consider that these measures of income mobility reflect not only changes in income but also other dynamic aspects of labour markets (such as the demand for labour, unemployment levels, flexible working patterns, job (in)security, etc.). The measures also reflect the changes in family composition — given the indicator is based on equivalised disposable income attributed to each household member.

These remarks notwithstanding, around 15 to 16 % of the EU-28 population moved either upwards or downwards on the income ladder by more than one income decile during the three-year period prior to 2017 (see Table 3).

Among the EU Member States, more than one fifth (20.5 %) of the population in Greece made an upward transition of more than one income decile in the three years prior to 2017, while an even higher share was recorded in Hungary (22.6 %). By contrast, in the three years prior to 2017, at least one fifth of the population in Bulgaria (20.6 %, 2016 break in time series), Poland (20.5 %) and Hungary (20.4 %) experienced a downward transition of more than one income decile.

Income mobility appeared to slow

When considering developments over time and comparing results for 2011 with those for 2017, it was commonplace to find that both upward and downward income mobility was reduced. Upward income mobility affected 17.8 % of the EU-28 population in 2011, a share that had fallen to 15.8 % by 2017. In a similar vein, the share of the EU-28 affected by downward income transitions was 17.2 % in 2011, a share that had fallen to 15.1 % by 2017. Upward income mobility was reduced at a particularly fast pace in Lithuania (-10.1 pp over 2011 to 2016), Ireland (-6.3 pp over 2011 to 2015) and Latvia (-5.0 pp), while downward income mobility was reduced at a relatively fast pace in Romania (-5.0 pp), Ireland (-6.6 pp over 2011 to 2015), and Lithuania (-10.9 pp over 2011 to 2016).

There were relatively few examples of increases in income mobility in 2017 (compared with 2011). This may reflect, at least to some degree, the impact of the global financial and economic crisis on the earlier reference period, with a higher share of the population exposed to fluctuating income levels during the crisis. That said, the share of the population who moved upward by more than one income decile (during the three years prior to the survey) rose by at least 1.0 pp in Finland (1.0 pp), Slovakia (1.6 pp; 2011-2016) and Hungary (3.4 pp). There were also several examples where a growing share of the population was exposed to the risk of falling incomes between 2011 and 2017. Some 19.3 % of the population in Croatia reported a downward transition of more than one income decile during the three years prior to the survey in 2017; this was 1.3 pp higher than the corresponding share from 2013 (18.0 %). Similar results were recorded for Austria — where the share of the population experiencing a downward transition grew by 1.2 pp — and Slovakia (up 0.9 pp; 2011-2016).

Figure 4: Contribution of social transfers to median equivalised net income, 2017
(in PPS)
Source: Eurostat (ilc_di03), (ilc_di13) and (ilc_di14)


Impact of social transfers on income

This next section compares the situation for disposable income before and after social transfers to assess the impact and redistributive effects of welfare policies. These transfers cover assistance that is given by central, state or local institutional units and include, among others, pensions, unemployment benefits, sickness and invalidity benefits, housing allowances, social assistance and tax rebates.

Social transfers led to median disposable income rising by PPS 4 777 in the EU-28

Figure 4 shows the overall impact of social transfers; this information is split between transfers for pensions and other transfers, for example, social security benefits and social assistance that have the aim of alleviating or reducing the risk of poverty.

In 2017, median disposable income in the EU-28 was PPS 4 777 higher as a result of social transfers (pensions included), and was PPS 1 268 higher if pensions are excluded from the analysis.

Among the EU Member States, there were considerable variations in the contribution made by social transfers to median disposable income in 2017. The largest transfers were observed in Luxembourg, where social transfers (including pensions) raised the median disposable income by PPS 10 070. Social transfers (including pensions) were also relatively high in Austria (PPS 7 310) and France (PPS 6 608), as well as in Norway (PPS 7 195; 2016 data).

A somewhat different pattern emerges if pensions are excluded from the analysis: in 2017, social transfers (excluding pensions) of above PPS 2 000 were recorded in Luxembourg (PPS 3 125), Sweden (PPS 2 747), Denmark (PPS 2 473) and Ireland (PPS 2 428).

It is interesting to compare the level of social transfers across the EU Member States including and excluding pensions. In Denmark and Estonia, social transfers including pensions were only around 1.7 times as high as social transfers excluding pensions in 2017. However, in Greece the same ratio was much higher, as the value of social transfers including pensions was 8.7 times higher than social transfers excluding pensions. The next highest ratios were recorded in Romania and Portugal where transfers including pensions were next to 6 times as high as transfers excluding pensions.

Table 4: Median equivalised net income before and after social transfers by household type, 2017
(in PPS)
Source: Eurostat (ilc_di04) and (ilc_di13b)


Social transfers were often targeted at nuclear families

Across the EU-28, median disposable income before social transfers was higher (PPS 14 086 in 2017) for persons living in nuclear households comprising two or more adults with dependent children than for the other two types of household. This is shown in Table 4 (PPS 10 628 for those living in households with two or more adults without dependent children and was PPS 8 426 for those living in households composed of a single person with dependent children); note this analysis includes pensions.

This pattern held across each of the EU Member States, except for Slovakia, where those living in households with two or more adults without dependent children had a slightly higher level of disposable income (PPS 8 825 in 2017) than those living in households composed of two or more adults with dependent children (where median disposable income was PPS 583 lower).

The impact of social transfers was considerable, as the level of median disposable income for those living in EU-28 households composed of two or more adults without children rose to PPS 18 815 in 2017, some 77.0 % higher than before social transfers (PPS 10 628). For comparison, social transfers led to a 50.4 % increase in the median disposable income of those people living in households composed of a single person with dependent children. The impact of social transfers was considerably lower for those people living in households composed of two or more adults with dependent children (up 15.3 %).

The re-distributive impact of social transfers generally resulted in the highest levels of median disposable income (after social transfers) being recorded for those people living in households that were composed of two or more adults without dependent children. This pattern held across all but two of the EU Member States in 2017; the only exceptions were Denmark and Estonia— where the highest levels of median disposable income were recorded for people living in households composed of two or more adults with dependent children.

A comparison of median disposable income before and after social transfers reveals that most governments chose to direct the greatest share of their support — in the form of social transfers — towards households with two or more adults without dependent children. For example, the median disposable income among those people living in this type of household in Greece rose from PPS 3 312 to PPS 9 692 as a result of social transfers (an increase of 293 %). Social transfers also resulted in median disposable incomes doubling, or more, for those people living in households with two or more adults without dependent children in Luxembourg, Portugal, Belgium, France, Italy, Slovenia and Spain. A similar situation was recorded in Serbia.

In Ireland and the United Kingdom, social transfers were targeted more towards single-parent households

There were two exceptions to this pattern, where the impact of social transfers was felt more by those people living in households composed of a single person with dependent children. In particular, median disposable income in Ireland rose by 268.9 % in 2017 as a result of social transfers for single-parent households (compared with a 63.0 % increase for people living in a household composed of two or more adults without dependent children, and a 15.4 % increase for people living in households with two or more adults with dependent children). The redistributive impact of social transfers was also felt most by single-parent households in the United Kingdom (where incomes rose by 244.2 % in 2017 as a result of social transfers).

In absolute terms, the highest increases in income were recorded for people living in Luxembourg in households with two or more adults without dependent children; they saw their income rise in 2017 by PPS 22 287 as a result of social transfers. There were also considerable increases in incomes for people living in this type of household in France, Belgium, Austria, Finland and Sweden, as social transfers resulted in median disposable income rising by more than PPS 10 000 in 2017.

Median disposable incomes for people living in single-parent households rose by PPS 8 554 as a result of social transfers in the United Kingdom and by PPS 8 512 in Ireland.

Income inequality

As noted above, while median disposable income provides a measure of average living standards, devoid of the potential distortion of aggregate measures such as GDP per capita, it still fails to offer a complete picture as it does not capture the distribution of income within the population and thereby does little to reflect economic inequalities.

The Gini coefficient

The Gini coefficient is a leading indicator that is used to measure income inequality. The Gini coefficient may range from 0, corresponding to perfect equality (in other words, income is equally distributed among every individual in a given society) to 100, corresponding to perfect inequality (in other words, when all of the income is received by a single person); thus, a lower Gini coefficient reflects a more egalitarian distribution of income.

Map 2: Gini coefficient for equivalised disposable income, 2017
(%)
Source: Eurostat (ilc_di12)


In 2017, the Gini coefficient for the EU-28 was 30.7 %. The highest income disparities among the EU Member States (with a Gini coefficient of at least 35.0 % — as shown by the darkest orange shade in Map 2) were recorded in Bulgaria and Lithuania. A second group of countries, with a Gini coefficient above the EU-28 average (in the range of 31.0 % - 34.5 %) comprised Estonia, Italy, Romania, the United Kingdom, Greece, Portugal, Spain and Latvia. At the other end of the range, income was more evenly distributed in Slovakia, Slovenia, Czechia, Finland and Belgium, as well as Iceland and Norway, where the Gini coefficient was 26.0 % or less.

The S80/S20 income quintile share ratio

Income inequalities within countries may also be illustrated through the income quintile share ratio, which is calculated as the ratio between the income received by the top quintile and the income received by the bottom quintile. High values for this ratio suggest that there are considerable disparities in the distribution of income between upper and lower income groups.

Figure 5: Income quintile share ratio, 2017
(ratio)
Source: Eurostat (ilc_di11)

In 2017, the income quintile share ratio for the EU-28 was 5.1 (see Figure 5); this signifies that, on average, the income received by the top 20 % of the population with the highest incomes was more than five times as high as the income received by the 20 % of the population with the lowest incomes.

The income quintile share ratio ranged from a low of 3.4 in Slovenia and Czechia, and 3.5 in Finland and Slovakia to a value between 6.0 - 7.0 in Greece, Latvia, Romania and Spain, peaking at 7.3 in Lithuania and 8.2 in Bulgaria.

The distribution of income was more often more equitable among the older generations

On the basis of the same measure, elderly people (aged 65 and over) in the EU-28 experienced less income inequality than the whole population, as their income quintile share ratio was 4.1 in 2017. This pattern of a more equitable distribution of income among the elderly (compared with the total population) was evident in the vast majority of EU Member States, the only exceptions being Cyprus (4.7 % vs 4.6 %) and Slovenia (where the ratio was identical, 3.4 %). Income inequality was also slightly higher among the elderly in Switzerland and Iceland (both 2016 data).

Impact of social transfers on inequalities

The effect of European welfare systems, in other words, pensions and other social transfers, in addressing income inequality can be demonstrated by comparing Gini coefficients before and after social transfers, to provide a quantitative assessment of their redistributive impact.

Figure 6: Gini coefficient before and after social transfers, for equivalised disposable income, 2017
(%)
Source: Eurostat (ilc_di12), (ilc_di12b) and (ilc_di12c)

In 2017, the EU-28 Gini coefficient for median disposable income before social transfers (pensions included) was 51.7 %, which fell to 30.7 % after social transfers. The impact of pensions and other social transfers on income inequality was particularly large in Portugal, Greece and Germany — where the Gini coefficient fell around 25 points —, while the largest impact was recorded in Sweden where the coefficient was reduced by 29.6 points (see figure 6).


Data sources

The data used in this section are primarily derived from data from EU statistics on income and living conditions (EU-SILC). EU-SILC is carried out annually and is the main survey that measures income and living conditions in Europe, and is the main source of information used to link different aspects relating to the quality of life at the household and individual level.

The reference population is all private households and their current members residing in the territory of an EU Member State at the time of data collection; persons living in collective households and in institutions are generally excluded from the target population. The EU-28 aggregate is a population-weighted average of individual national figures.

Direct access to
Other articles
Tables
Database
Dedicated section
Publications
Methodology
Legislation
Visualisations
External links