Wages and labour costs developments in the EU and its Member States

The lasting social effects of the high inflation period and fairness issues

The sizeable drops in real wages from the end of 2021 onwards have had a large negative impact on low- and middle-income households.

This decline in real wages reflected to a large extent high inflation rates primarily driven by soaring energy and food prices, which both account for a significant proportion of these households' expenditures. Despite support measures and higher wage growth, their purchasing power eroded over 2022-2023. This section delves into the social consequences of this decline in purchasing power, examining the interrelated challenges of declining purchasing power, material deprivation, and financial distress among workers (Section 2.3.1). It also investigates the disproportionate impact on low-wage earners and middle-income groups (Section 2.3.2), as well as the implications for labour shares and concerns about fairness (Section 2.3.3).

Scars of the 2022-2023 high inflation period are still visible

Households’ purchasing power losses have been mitigated by recent wage developments, as well as by transfers and tax reductions, though the role of the latter has diminished recently.

The growth of real gross disposable household income (GDHI) started to decline at the beginning of 2022 and turned negative for most of the year, reflecting a drop in the real compensation of employees and real net property incomes. GDHI growth turned positive again at the beginning of 2023, mainly driven by a recovery in wage growth and an increase in net social benefits (see Graph 2.5). Direct transfers and tax policies (notably reductions and non-indexation of tax thresholds) were a major driver of real GDHI growth until the second quarter of 2023, following the introduction of support measures during the energy crisis. But since the second half of 2023, their contribution declined and turned negative, partly due to the withdrawal of these measures.

Graph 2.5: Real GDHI growth and its main components, EU

Real GDHI growth and its main components, EU

Note:

The nominal GDHI is converted into real GDHI by deflating values using the deflator (price index) of household final consumption expenditure. Net transfers notably include net social benefits and taxes on income and wealth (negative contributions). GDHI = gross disposable household income; Q1 = first quarter; Q2 = second quarter; Q3 = third quarter; Q4 = fourth quarter.

Sources:

European Commission calculations based on Eurostat, National Accounts [nasq_10_nf_tr and namq_10_gdp].

At the same time, the material and social deprivation of employees increased in both 2022 and 2023 despite the rebound in households’ purchasing power .

The share of individuals of material and social deprivation for employed people in the EU increased from 7.2 % in 2021 to 8.2 % in 2022 and 9.1 % in 2023. One reason is that lower-income households have been more affected by the increases in prices in energy and food, which represent a relatively higher share of their consumption basket. Most Member States experienced increases between 2022 and 2023, from 0.1 (Latvia) to 3.0 (Lithuania) pps but the rate continued to fall in Romania (-4.4 pps), Ireland (-1.8 pps), Greece (-1.4 pps), Cyprus (-1.3 pps), Croatia (-1.2 pps) and Estonia (-0.6 pps). In the latter group of Member States the rebound in real wages helped cushion the impact of high inflation on low-income households.

High financial distress also persists among employees, with large differences across Member States.

The proportion of workers reporting financial distress increased significantly, peaking at 15.5 % in February 2023 for the EU, 5 pps higher than in January 2022 and more than 3 pps higher than in the pre-pandemic period. It then stabilised at high levels, reaching 13.7 % in August 2024, even though real wages started to grow again in the second half of 2023. Between 2021 and August 2024, the reported financial distress for workers increased in all but six Member States (Graph 2.6). The largest increases were recorded in Estonia, Romania, Malta, Hungary and Greece (above 7 pps), followed by France and Denmark (4 to 7 pps). Five of these Member States experienced deep losses in real wages in 2022 and a weak or no rebound in 2023.

Graph 2.6: Financial distress of workers

Financial distress of workers

Note:

Information on the financial distress of workers in 2023 and 2024 is given in percentages; information on change between 2021 and 2024 is given in percentage points.

Source:

European Commission, Business and consumer surveys

In-work poverty continued to decline, but in some countries this reflected deeper losses for middle-income households than for lower-income ones .

At the EU level, in-work poverty continued its long-term downward trend since 2015, standing at 8.3 % in 2023 (based on 2022 incomes) and is forecast to remain at 8.4 % in 2024 (based on 2023 incomes). At the level of Member States, the rate increased in 14 countries in 2023 (based on 2022 incomes), notably in Bulgaria, Luxembourg and Slovakia, pointing to a rise in the share of vulnerable workers (Graph 2.7). By contrast, in 12 countries in-work poverty diminished. This means that for these countries real median incomes tended to decline proportionally more than the incomes of the poorest, since in-work poverty is a relative measure that compares low incomes to median incomes. Indeed, most of these countries have experienced very high inflation and large drops in real wages. Therefore, the share of vulnerable workers may not have decreased as suggested by the fact that material and social deprivation and financial distress tended to increase in those countries (Graph 2.6). Overall, the analysis in this subsection shows that the impact of the high inflation on workers varied along the wage distribution, with important differences across Member States. Section 2.3.2 delves deeper into income dynamics for different income groups.

Graph 2.7: In-work at-risk-of-poverty rate for employees

In-work at-risk-of-poverty rate for employees

Note:

Information on the financial distress of workers in 2023 and 2024 is given in percentages; information on change between 2021 and 2024 is given in percentage points.

Source:

European Commission, Business and consumer surveys

Vulnerabilities increased for low-wage earners and the middle-income groups

Wide parts of society have been affected by the decline in purchasing power, leading also to marked changes in the income distribution.

While on average income inequality in the EU has been overall stable , income variations by deciles over 2022-2023 (Table 2.1A in the annex) reveal a mixed picture. In 2022, lower and medium-income deciles were disproportionately affected in a majority of Member States, while in 2023 higher income deciles were affected to a greater extent (see Table 2.2A in the annex). These developments led to differentiated changes in the income distribution at the Member States level. Between 2021 and 2023 the lowest income deciles lost 0.3 pps or more in their share of national equivalised income in 15 Member States suggesting a rise in income inequality (see Graph 2.8) . Simultaneously, the share of national equivalised income increased for the lowest income deciles in 10 Member States – with the highest increases observed in Greece, Spain and Romania – hinting at a decrease in income inequality.

Support measures and minimum wage increases helped protect low-wage earners, though to different degrees across Member States.

Exceptional support measures put in place to protect households from the effects of high inflation tended to benefit the lower income deciles to a greater extent. In Member States in which less than 2 % of GDP was allocated to exceptional support measures, the lower income deciles lost more than other income deciles (Graph 2.8) . In other Member States, including Bulgaria, Germany and Poland, relatively larger support measures and significant increases in minimum wages supported low-wage earners’ real incomes relative to higher income deciles. Losses in real income for all income groups may not be fully recovered in the short term. It is therefore important to address the remaining social consequences of the decline in purchasing power, with particular attention to vulnerable low-income and lower middle-income households.

Graph 2.8: Average change in the nationally equivalized income for the lower, middle and upper income deciles for 2021-2023 across Member States

Source:

EU -Statistics on Income and Living Conditions and European Community Household Panel Surveys, Eurostat [ilc_di01]

Statutory minimum wage increases broadly offset losses in purchasing power for minimum wage earners in most Member States.

Between January 2022 and July 2024 statutory minimum wages grew in nominal terms by more than 10 % in all Member States where such wages are in place . In real terms, they increased by more than 10 % in eight countries, and by 4 to 10 % in another six countries, while they declined in France by 1.3 % and by more than 3 % in Czechia and Slovakia (Graph 2.9). These minimum wage developments, associated with high labour shortages, probably contributed to higher increases in hourly wages in sectors where low-wage earners are more represented. During the high-inflation period in 2022-2023, lower-paying sectors such as wholesale and retail trade, transport, accommodation and food services or construction, have seen higher growth rates than other, on average higher paying sectors .

Although difficult to assess, non-compliance with minimum wage protection, either statutory or based on collective bargaining, may be a challenge in some Member States (see Box 2.2).

It is estimated that countries where minimum wages are closer to the average or median wage tend to have higher non-compliance . Also, young people, unskilled workers, women, workers in smaller firms, part-time workers, and those working in the services sectors may be particularly affected by non-compliance with existing minimum wage protection rules. Improving compliance would allow more workers to receive a higher remuneration in practice, in line with national minimum wage regulations.

Graph 2.9: Minimum wage developments in Member States with statutory minimum wages

Note:

Cyprus was excluded as the statutory minimum wage was introduced in 2023.

Sources:

Eurofound and Eurostat [earn_mw_cur, prc_hicp_midx].

Box 2.2: Estimating non-compliance with minimum wages

Many workers in the EU, who are in theory covered by minimum wage protection, receive lower remuneration in practice (European Commission, 2020). The estimated average rate of non-compliance across the EU based on data from EU-Statistics on Income and Living Conditions (EU-SILC) is 8.1 %, while that calculated on the basis of data from the Structure of Earnings Survey (SES) is 1.43 % (both figures are based on the 2018 waves of surveys). Estimations of non-compliance per Member State also widely differ, with much higher rates obtained when using data from EU-SILC than when using the SES. This divergence could be due to differences in the target groups of the surveys, since the SES only covers larger firms (more than 10 employees) and data is reported by employers (who are less likely to report that they are paying wages below the minimum), while for EU-SILC the respondents are households .

High rates of non-compliance with minimum wage protection are reported in eight Member States. Based on EU-SILC data, the countries where non-compliance is estimated to affect more than 8 % of the workforce are Germany, Spain, Italy, Cyprus, Lithuania, Hungary, Portugal and Sweden. This group includes two countries where minimum wage protection is provided only by collective agreements (Italy and Sweden). By contrast, the non-compliance rates estimated for Belgium, Czechia, Malta, Slovakia and Finland are relatively low (less than 2 %). While these estimates should be read with caution, they provide an indication of the relative magnitude of the problem. Estimations for Germany and Italy (Eurofound, 2023), for instance, roughly coincide with those in previous studies (DIW, 2019; Garnero, 2018; and Garnero et al., 2022).

Young people, unskilled workers, women and workers in smaller firms, as well as part-time workers, are more affected. These workers, who are affected by non-compliance, are generally younger, less educated, and more likely to be female, on a fixed-term or part-time contract and employed by smaller firms (Eurofound 2023). This seems logical as they correspond to the groups that are more likely to earn the minimum wage, as highlighted in Recitals 10 and 14 of the directive on adequate minimum wages . Regarding the sectoral composition, non-compliance is relatively higher in the agriculture, construction, and services sectors than in the manufacturing sector (8-13 % versus 6 %).

For some Member States, estimated non-compliance rates are significantly higher when using hourly wages instead of monthly wages, suggesting the existence of non-declared working hours. Higher rates of non-compliance when using hourly wages are notably observed in Bulgaria, Croatia, Cyprus, Luxembourg and Portugal and, to a lower extent, in Germany and Italy. This suggests that some employers may impose longer hours on workers than stated in their contract, so that they may seem to comply with minimum wages calculated on a monthly basis but do not comply on an hourly basis.

The directive on adequate minimum wages aims at reducing non-compliance. One of its main objectives is to establish a framework for enhancing the effective access of workers to minimum wage protection. It obliges Member States with statutory minimum wages to take measures to strengthen its enforcement, by providing for effective, proportionate and non-discriminatory controls and field inspections, as well as by developing the capability of their enforcement authorities . In addition, it requires all Member States to reinforce compliance through measures related to public procurement, information, right to redress, protection from adverse treatment and penalties .

  1. Related note aThe measurement error arises because reported working hours are those current at the time the survey was taken, while earnings are those received the year prior to the survey. The advantage of EU-SILC is that it covers the entire working-age population and asks workers to report their own incomes. At the same time, its results should also be interpreted with caution as hourly wages in EU-SILC may contain inaccuracies.
  2. Related note bDirective 2022/2041.
  3. Related note cArticle 8 of Directive 2022/2041.
  4. Related note dArticles 9, 11, 12 and 13 of Directive 2022/2041.

Labour share developments have raised fairness concerns

In the last two decades the labour share, that is the part of national income allocated to employees as compensation, slightly declined, amid cyclical fluctuations .

The labour share dropped from 63.6 % in 2000 to 62.3 % in 2017, with part of the decrease over that period explained by Ireland, and then slightly increased to 62.8 % in 2019 . It increased further during the COVID-19 pandemic to 63.3 % in 2020 because of labour hoarding by firms and job-retention schemes adopted by governments. However, the labour share started to drop again to 61.6 % in 2022 with the gradual roll-back of these retention schemes and the witnessed decline in real wages. In parallel, the business profit share in the EU (and a majority of Member States) has remained above pre-pandemic levels in the last few quarters. It reached 46.2 % on average between the first quarter of 2023 and the first quarter of 2024, compared with 45.5 % in 2019 . These developments have raised questions on how to ensure a fair sharing of the burden of the remaining social challenges between companies and workers . Going forward, the labour share is expected to bounce back to 62.5 % in 2024, thanks to the mild rebound in real wages, and to remain stable in 2025. Although these wage dynamics will help to make up for most of the losses in the labour share in recent years, it is forecast to still remain slightly below its 2019 level in most Member States. For most Member States, the changes in the labour share between 2000 and 2017 have been mainly due to within-sector changes rather than shifts of economic activity between sectors.

Over the past few decades structural factors have put downward pressure on the labour share in Member States as well as other advanced economies and are also likely to affect it in the future .

Digital technologies as well as artificial intelligence can be important engines of job creation and productivity but may also lead to sectoral disruptions, job displacement and further polarisation in the workforce between low- and high-skilled employees . In addition, the ongoing shift of economic activity towards services and the development of alternative forms of work, such as platform work or remote working, can contribute to a further decline in unionisation rates and workers’ bargaining power, which may in turn weigh on the labour share . The possible future slowdown in globalisation could mitigate the decline in labour shares through less offshoring, although this depends on whether local jobs are sustainable in light of the growing competitive pressure from abroad.

Notes

  1. Material and social deprivation refers to the inability to afford a set of specific goods, services, or social activities that are considered by most people as essential for an adequate quality of life. It is measured as the percentage of households that experience a lack of at least 5 out of the 13 deprivation items.
  2. The European Commission ‘business and consumer surveys’ define financial distress as the need to draw on savings or to run into debt to cover current expenditures.
  3. Between 2021 and August 2024 financial distress decreased in Belgium, Bulgaria, Cyprus, Luxemburg, Poland and Slovakia.
  4. Denmark, Estonia, Greece, Malta and Sweden.
  5. In-work poverty refers to the percentage of persons in the total population who are employed or self-employed and at risk of being relatively poor, that is they have less than 60 % of the national median equivalised disposable income after social transfers
  6. The flash estimates are based on microsimulations and macro-economic models. Their aim is to provide timelier social statistics. Since this is a forecast, these estimates should be interpreted with care. https://ec.europa.eu/eurostat/web/experimental-statistics/income-inequality-poverty-indicators
  7. As reflected by the slight decline in S80/S20 income quintile share ratio.
  8. Calculations are based on a proxy in relation to the cut-off values for the deciles.
  9. Except for Denmark, Estonia, Ireland, Austria, Portugal and Sweden.
  10. The highest losses were in Sweden (9.0 pps), Slovakia (4.8 pps) and France (4.5 pps).
  11. Between September 2021 and January 2023, less than 2 % of GDP was allocated by governments to shield households and firms from the energy crisis in Denmark, Finland, Cyprus, Sweden Ireland, Hungary, Belgium and Estonia. Many measures were withdrawn in 2023 with the fall in energy prices (Sgaravatti et al., 2023).
  12. All but five Member States (Denmark, Italy, Austria, Finland and Sweden) have statutory minimum wages. They notably increased by more than 40 % in Bulgaria, Latvia, Romania and Poland, all countries with relatively low minimum wage levels. However, growth was also strong, at more than 20 %, in Belgium, Germany, Estonia, Ireland, Croatia, Greece, Lithuania, Hungary and the Netherlands, some of them with already relatively high minimum wage levels.
  13. Hourly wages are proxied by the labour cost index. The financial sector, information and communication, professional, scientific, and technical activities, the real estate sector and manufacturing continue to pay comparatively high wages. However, hourly wages increased more in the wholesale and retail trade and construction sector (see Graph 2.16 in the annex).
  14. Calculations are based on EU Statistics on Income and Living Conditions (SILC) data and refer to the so-called Kaitz indexes measured by the ratio of the minimum wage to the average or median wage.
  15. Wages and employment are less easily adjustable than output, due to the rigid nature of labour markets and labour hoarding during recessions. This leads to rising labour shares during recessions and declining ones during recoveries. This was also visible during the financial, economic and debt crisis that started in 2009 and the COVID-19 recession in 2020 (see also Chapter 1). The moderate decoupling between labour productivity and wage growth between 2000 and 2019 means that real wages have increased at a lower rate than productivity. As a result, capital has claimed a growing share of productivity gains.
  16. Based on the AMECO database. The labour share is calculated as the ratio of total compensation per employee over nominal GDP per person employed. The labour share of Ireland decreased from 52.6 % in 2000 to 35.6 % in 2019, which accounted for more than half of the decline in the EU-wide labour share during that period.
  17. The business profit share is defined as gross operating surplus divided by gross value added. Due to its volatility, an average over four quarters is considered. In Greece, Ireland, Cyprus, Latvia, Lithuania, Luxembourg, Hungary, Malta, the Netherlands and Slovenia, the increase in business profit share is sizeable (by 1.5 pps more, compared with 2019). In Czechia, Germany, France, Italy, Slovakia and Sweden, the increase was also significant (by 1–1.5 pps). In Belgium, Denmark, Croatia, Poland and Romania, it remained stable (increased or decreased by less than 1 pps) and in Bulgaria, Estonia, Spain, Austria, Portugal and Finland it decreased (by 1–1.5 pps).
  18. OECD (2023c). This has also led governments to increase taxation of (super-) profits.
  19. Schneider (2011).
  20. OECD (2018c), Pissarides and Maayan (2023).
  21. Labour market deregulation during 1970-2015 has been found to have negatively affected labour shares (Ciminelli et al. 2018)..