Do we live in an equal world? An analysis of the gender pay gap in the EU

December 9, 2019

In the past decades there has been a major progress towards equal gender conditions all around the world. New frontiers of equality have been reached and it is not as rare as it used to be to see a woman in charge. However, all these big improvements shadow subtle inequalities that still take place in the job market, hidden behind the veil of common belief that living in a modern world implies that we live in an equal world.

 

 

Pay inequality in the EU countries

Among all the possible differences in treatment between the two genders, what stands out the most is the pay gap. Gender pay gap (GPG) is a phenomenon that occurred in the vast majority of occupations in the first half of the 20th century but has seen a progressive decrease in the following decades. 

 

The evolution of (and decrease in) the GPG has brought the EU countries to the situation shown in the following graph:

 

What the graph shows is the unadjusted (or raw) gender pay gap. The unadjusted GPG 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 and it is calculated for enterprises with 10 or more employees. In 2017, women were paid 16% less than men on average according to Eurostat. There are considerable variations across the Union, ranging from 25.6% in Estonia to the low 3.5% in Romania.

 

Behind the unadjusted gender pay gap

As the name suggests, the unadjusted GPG can be potentially deceiving in identifying discrimination due to the fact that it does not take into account some major factors that may affect the difference between men's and women's earnings. Since it only points out the pure difference in hourly wages between the genders, it could potentially suggest that women earn less due to discrimination whereas the causes of the gap are not of discriminatory nature. In fact, by looking at the graph above you may be surprised to see that countries such as Italy or Romania have very low pay gaps (5% and 3.5%, respectively) while others are unexpectedly lagging behind in income parity - for example, the United Kingdom (more than 20%).
 

Men and women differ in many characteristics that may explain their differential earnings, such as education, experience and occupation. As a matter of fact, education has become increasingly less important in explaining the GPG considering that men’s and women’s average years of education and achieved qualifications are now very similar. Therefore, the type of occupation is now one of the most relevant factors that affect these numbers.
 

Occupational differences and their effect on the GPG

There are three main channels through which occupational differences can contribute to the GPG: sectoral gender segregation, occupational gender segregation and rate of female employment.

Sectoral gender segregation refers to the concentration of one gender in certain economic activities. It may explain part of the difference in earnings between men and women - if one sex tends to be concentrated in low-paying economic sectors and the other sex tends to be concentrated in high-paying sectors. Usually, women tend to work in lower-paying sectors whereas men tend to work in higher-paying sectors.

 

Similarly, occupational gender segregation defines the concentration of one sex in certain occupations, when one sex tends to be concentrated in low-paying occupations and the other sex tends to be concentrated in high-paying occupations. This could be due to the fact that men are promoted to higher positions more frequently than women, a phenomenon known as the ‘glass-ceiling’, a metaphorical barrier that prevents women from climbing the career ladder. 

 

Both these phenomena are analysed in addition to other ones (occupation, employment contract, working time, principal economic activity, enterprise size and enterprise control) in the following table that aims to quantify the extent to which differences in individual characteristics of men and women help to explain the unadjusted GPG. The overall explained part indicates the total share of unadjusted GPG that can be explained by accounting for the effect on the GPG by variables in the table. The positive (green) cells show the characteristics in which men score on average better than women - which may reflect why their incomes are on average higher. The negative (yellow) cells show characteristics in which women do better and for which they would therefore be expected to receive higher compensation than their male counterparts.

 

The differences between men and women (favoring men) related to the types of positions they occupy and types of the enterprises that employ them point to gender segregation. For example, a positive value for economic activity means that men worked, on average, in better paid economic sectors than women, while a negative value means that women worked, on average, in better paid economic sectors than men.

 

At the EU level, the overall explained part − calculated as the weighted average of the overall explained parts in EU member states − is 31%. This means that 31% of the difference between log hourly earnings of men and women can be attributed to the difference in average characteristics between male and female workers that are in favour of men. The EU explained part is mostly driven by two explanatory factors, which are economic activity and working time. They attribute 32% and 13% respectively to the difference between log hourly earnings of men and women. Job experience explains only 1% of the difference, whereas age and enterprise control have no explanatory effect. The positive explanatory factors are partially cancelled out by the negative explanatory factors for education (-7%), enterprise size (-5%), occupation (-3%) and employment contract (-1%). At the EU level, the effect of occupation is rather smaller than expected due to its uneven effect across EU Member States.

 

Rate of female employment and the GPG

As for the rate of female employment, a lower proportion of working-age women than men tend to participate in the labour market. This may severely affect the final measure of the GPG because it means that fewer women are counted - relative to men. The reason why lower rates of female participation may significantly bias the GPG is that, according to Eurostat, lower-educated or lower-skilled women tend not to engage in the labour market in some countries. This phenomenon is known as 'self-selection', and it makes the GPG seem smaller than it actually is simply due to a partial absence of lower-paid women from the labor market.

 

According to a 2018 study published by Olivetti and Petrongolo in the Journal of Labor Economics, self-selection especially occurs in middle-income countries, where only women with certain characteristics (for example, women who are unmarried or childless)  decide to enter the job market. The two researchers explain it as follows: “If women who are employed tend to have relatively high‐wage characteristics, low female employment rates may become consistent with low gender wage gaps simply because low‐wage women would not feature in the observed wage distribution”.

 

Therefore, we should observe a positive correlation between the rate of female employment and the GPG, as the following graph confirms:

 

By looking at these data it is clear that simply comparing levels of the GPG over time and across countries is not sufficient. But this is possibly the reason why there are still many people who believe that gender inequality has been overcome. In fact, according to a special Eurobarometer survey commissioned in 2017 by the European Commission and carried out in all EU countries, almost seven in ten (69%) respondents think women are paid less than men per hour of work. Almost one quarter (23%) think women are paid the same, while 2% think they are paid more. Women are more likely than men to think women are paid less (74% vs. 65%).

The survey highlights that it is not that uncommon to believe that pay equality has definitely been reached. The reason for this I believe is that we are seldom given a thorough analysis, only raw numbers that do not allow us to take into account all of the elements of discrimination to their full extent. 

 

Our society has certainly made progress since the past, but there is still plenty of room for improvement. Hopefully in the next few years more time will be dedicated to this issue by all the countries and the EU, and we will see more women leading businesses and working in male-dominated sectors.

 

All graphs used in the article come from the 2018 Eurostat study 'A decomposition of the unadjusted gender pay gap using Structure of Earnings Survey data' and the European Commission's report to the Special Eurobarometer 465 on gender equality.