Statistics Explained

Gender pay gap statistics




Data from November 2021 (part "Possible causes of the unadjusted gender pay gap") and February 2021 (rest of the article).

Planned article update: 8 March 2022.

Highlights

In 2019, women's gross hourly earnings were on average 14.1 % below those of men in the EU.

In 2019, the highest gender pay gap in the EU was recorded in Estonia (21.7 %) and the lowest in Luxembourg (1.3 %).

Source: Eurostat (sdg_05_20)

This article provides a brief overview of gender pay gap (GPG) statistics, including the unadjusted gender pay gap used to monitor imbalances in earnings between men and women. The unadjusted gender pay gap is defined as the difference between the average gross hourly earnings of men and women expressed as a percentage of the average gross hourly earnings of men. It is calculated for enterprises with 10 or more employees.

Full article

Gender pay gap levels vary significantly across EU

For the economy as a whole[1], in 2019, women's gross hourly earnings were on average 14.1 % below those of men in the European Union (EU-27) and 14.9 % in the euro area (EA-19). Across Member States, the gender pay gap varied by 20.4 percentage points, ranging from 1.3 % in Luxembourg to 21.7 % in Estonia (Figure 1).

Figure 1: The unadjusted gender pay gap, 2019 (difference between average gross hourly earnings of male and female employees as % of male gross earnings)
Source: Eurostat (sdg_05_20)

Part-time versus full-time employment

Pay gaps can also be analysed from the perspective of part-time or full-time employment. Information at this level of detail is not available, however, for all EU Member States (Figure 2). In 2019, the gender pay gap for part-time workers varied from -5.1 % in Italy to 22.0 % in Croatia. A negative gender pay gap means that, on average, women's gross hourly earnings are higher than those of men. This is often due to a selection bias, especially when the employment rate is lower for women than for men: women engaging in the labour market may have comparatively higher skills and education levels that men. For full-time workers, pay gaps varied also widely in the EU Member States, ranging from -1.6 % in Italy to 24.1 % in Latvia.

Figure 2: The unadjusted gender pay gap by working time (%), 2019
Source: Eurostat (earn_gr_gpgr2wt); see Country codes

Gender pay gap much lower for young employees

The gender pay gap is generally much lower for new labour market entrants and tends to widen with age. However, those differences over age groups can have different patterns across the countries (Table 1). The gender pay gap might increase with age as a result of the career interruptions women may experience during their working life.

Table 1: The unadjusted gender pay gap by age (%), 2019
Source: Eurostat (earn_gr_gpgr2ag)

Highest gender pay gap in financial and insurance activities

The gender pay gap in financial and insurance activities is higher than in the business economy as a whole

A breakdown for the different sectors of the economy also reveals interesting patterns (Table 2). In all EU Member States, except Belgium and Spain, the gender pay gap in the financial and insurance activities (NACE Rev. 2 section K) is higher than in the business economy as a whole (NACE Rev. 2 aggregate B to N). In 2019, the gender pay gap in financial and insurance activities varied from 6.6 % in Belgium to 38.3 % in Czechia. Within the business economy as a whole, the highest gender pay gap was recorded in Estonia (23.2 %) and the lowest in Sweden (8.7 %).

Table 2: The unadjusted gender pay gap by economic activity (%), 2019
Source: Eurostat (earn_gr_gpgr2)

It is also interesting to note the economic sectors for which a significant number of Member States recorded negative gender gaps. Nine Member States registered negative gender pay gaps in the water supply, sewerage, waste management and remediation activities (NACE Rev. 2 section E) and thirteen in the construction industry (NACE Rev. 2 section F).

Gender pay gap higher in the private sector

In 2019, the majority of the EU countries (for which data are available) recorded a higher gender pay gap (in absolute terms) in the private sector than in the public sector. This might be due to the fact that, in most countries, pay is determined by transparent wage grids that apply equally to men and women. The gender pay gap varied in the private sector from 8.9 % in Belgium to 22.9 % in Germany, and in the public sector from -0.4 % in Romania to 19.7 % in Latvia.

Table 3: The unadjusted gender pay gap by economic control (%), 2019
Source: Eurostat (earn_gr_gpgr2ct)

Possible causes of the unadjusted gender pay gap

As an unadjusted indicator, the gender pay gap gives an overall picture of the differences between men and women in terms of earnings and measures – a concept which is broader than discrimination in the sense of "equal pay for work of equal value". Indeed, parts of the difference in earnings of men and women can be explained by (1) differences in the average characteristics of male and female employees and (2) differences in the financial returns for the same characteristics. In the methodological study ‘Gender Pay Gaps in the European Union – a statistical analysis’, both drivers of the unadjusted gender pay gap are analyzed based on the latest (2018) edition of the four-yearly Structure of Earnings Survey (SES).

SES microdata cover the earnings and observed characteristics of individual employees. The latter include: (1) the personal characteristics of individual employees such as age, education and job experience, (2) the types of job done by individual employees and (3) the types of enterprises in which individual employees are working.


1. Introduction and Methodology

To disentangle the different drivers of the unadjusted GPG, Eurostat applied the Oaxaca-Blinder methodology, whose different variants are commonly used by National Statistical Offices.

First, we apply a linear regression model to the log hourly earnings of both men and women separately. The model relates, in a linear way, the log hourly earnings to a selection of SES variables.

Secondly, we calculate the difference between the log of hourly earnings of men and women. This difference gives us the Oaxaca Blinder decomposition of the unadjusted gender pay gap in which we can identify three main parts:

  • The part caused by different characteristics of men and women in the labour market
  • The part caused by different financial returns for the same characteristics
  • A residual element

In a final step, the former results of the Oaxaca-Blinder decomposition are used to adjust the Gender Pay Gap. The unexplained part of the Gender Pay Gap is calculated by substracting the explained part due to different characteristics to the unadjusted Gender Pay Gap.


2. Impact of differences in characteristics

At EU level, the explained part represents 3.0 percentage points. This means that women are expected to earn 3.0 % less than men, according to their average characteristics on the labour market (which are less remunerative than those of males). This rather limited adjustment is because countries record positive or negative adjustments that partly cancel out at EU level.

As a result, the adjusted (or unexplained) GPG is 11.4 % against 14.4 % for the unadjusted GPG, that is, women still earn 11.4 % less than men, on average, after correcting for their different average characteristics (see Figure 3, where countries are ranked in increasing order of the unadjusted GPG).

Figure 3: Explained and unexplained parts of the unadjusted gender pay gap, 2018
(%)
Source: Eurostat - Structure of earnings survey


If we rank countries according to their unexplained gender pay gap, the picture changes (see Figure 4) with some countries moving by more than 10 positions after correcting for the different characteristics of men and women in the labour market.

Figure 4: Unexplained gender pay gap, 2018
(%)
Source: Eurostat - Structure of earnings survey


For further information on the contributions of each explanatory factor to the part of GPG which is explained by the different characteristics of men and women in the labour market, please refer to the above mentioned study ‘Gender Pay Gaps in the European Union – a statistical analysis’.


3. Gender differences in financial returns

Turning to the second part of the decomposition, we observe in most countries that financial returns are significantly higher for men than for women as regards age, and in several cases, for experience. This could point to the impact of career breaks, which are more frequent among women than among men.

Also, many countries record lower financial returns for men than for women as regards age2 or job experience2. This is expected, as women on average receive higher returns at the end of their careers (and lower returns at the beginning) compared with men (see the publication on Wage determinants in the EU). It was also found that working part-time is generally more penalising, in terms of earnings, for men than for women.  


4. Other segregation effects

The decomposition of the unadjusted GPG does not capture all segregation effects between men and women in the labour market. In particular, women work on average fewer hours per month than men do. Moreover, a lower proportion of women than men participate in the labour market. The next figure shows all the possible factors that may influence the expected earnings of women and men of working age (15–64 years old).

Figure 5: Gender segregation effects


To give a complete picture of the gender earnings gap, Eurostat has developed a synthetic indicator, the ‘gender overall earnings gap’, which measures the impact of three combined factors on the difference in the average earnings of all women of working age – whether employed or not employed – compared to men. Those factors are: the average hourly earnings, the monthly average of the number of hours paid and the employment rate for men and women. The results are displayed in Figure 6.

Figure 6: Gender overall earnings gap, 2018
(%)
Source: Eurostat (teqges01)

For more information, see Gender Statistics


5. Conclusions

The unadjusted GPG, owing to its simple definition, is widely used to monitor gender equality policies. It is sometimes viewed as a tool to measure possible gender discrimination on the labour market. However, ‘unequal pay for equal work, among male and female workers’ is just one of the possible causes of the unadjusted GPG.

The other possible drivers of the GPG are the different characteristics of male and female employees and possible gender gaps in financial returns for the same characteristics. Analyzing the different determinants of the unadjusted GPG is therefore important to better target gender equality policies.

The explained part of the decomposition should not be understood in a normative way or as a ‘natural’ GPG. It measures the difference in the hourly earnings caused by the different characteristics of male and female employees, be they caused by gender stereotypes, gender-biased preferences, fields of education or other reasons. This component may also include some discrimination elements if there are barriers to entering specific economic activities or occupations.

It is also tempting to interpret the unexplained component as a measurement of a possible discrimination through ‘unequal pay for equal work’ reflected in different financial returns for the same characteristics. This is not recommended, though, as important variables, such as total work experience, are not collected in the SES. Including such additional variables in the regression analysis may change the results.

Finally, it is reminded that the decomposition of the unadjusted GPG does not account for all segregation effects between men and women in the labour market. In particular, women tend to work on average fewer hours per month than men, which is not captured by the unadjusted GPG calculated on an hourly basis. Moreover, in most countries, the employment rate is lower for women than for men. To embrace all segregation effects, it is recommended to use synthetic indicators calculated on the whole population of working age such as the gender overall earnings gap (GOEG).

Data sources

From reference year 2006 onwards, the unadjusted gender pay gap is based on the methodology of the Structure of earnings survey (SES) according to Regulation (EC) No 530/1999. The SES is carried out with a four-yearly periodicity. The most recent reference years available for the SES are 2014 and 2018. Eurostat computed the gender pay gap for these years on this basis. For the intermediate years countries provide to Eurostat gender pay gap estimates benchmarked on the SES results.

Source data for tables and graphs

Context

Reducing the gender pay gap is one of the key priorities of gender policies at both EU and national levels. At EU level, the European Commission prioritised "reducing the gender pay, earnings and pension gaps and thus fighting poverty among women" as one of the key areas in the framework of the A Union of Equality: Gender Equality Strategy 2020-2025. The unadjusted gender pay gap indicator is used to monitor imbalances in earnings between men and women.

Direct access to

Other articles
Tables
Database
Dedicated section
Publications
Methodology
Visualisations




Gender pay gap in unadjusted form (sdg_05_20)


Earnings
Gender pay gap in unadjusted form (earn_grgpg)
Gender pay gap in unadjusted form - Nace rev.2 (earn_grgpg2)
Gender pay gap in unadjusted form - Nace rev.1.1 (earn_grgpg1)


Earnings
Gender equality


Notes

  1. Here defined as industry, construction and services except public administration and defence and compulsory social security: NACE Rev. 2 Sections B to S with the exception of Section O.