Effect of Gender Differences and Other Factors on Remuneration of Employees in EU Countries

The level of income or salary of employee households determines their income situation, which is one of the basic factors determining their living standard. Gender income differences in the context of a position of women in societies are discussed around the world, including EU countries.

The aim of the article is to verify discussed gender differences in income from employment, i.e. in wages and salaries among men and women in all EU countries. Above all, the aim is to identify factors which cause gender differences in employment remuneration, including the size of the effect of individual factors. The assessed factors are demographic and socio-economic characteristics of the individual, and factors applying to employment – job sector, type of employment contract, number of months of performed work and others. Data available from the EU-SILC database makes it possible to focus on the effect of the factors in more detail. Special attention is paid to the “Job sector” factor. This is a factor where historically in society there is a display of ongoing awareness of typically male and female industries. The research question is whether different preferences of different genders appear in the choice of job sector and whether this is going in a positive or negative direction in terms of the income situation.

Gender differences in remuneration are recorded in all EU countries. On average women have a 15% lower wage than men (Boll, Lagemann, 2019). A disposable income of households significantly reflects the socio-economic situation of the population of the state concerned (Keeley, 2015). The household or individual disposable income should be taken into account in a measurement of living standards (Burlacu, 2016). Due to the persisting gender pay gap and on average lower remuneration of women, they do not have equal starting conditions to achieve the same living standard as men and this is seen above all in the households of single women and single female parents. As a consequence of this, these households are more prone to the risk of a life of poverty (Kramer et al., 2016). Castellano and Rocca (2014) confirm the occurrence of a pay gap in all European countries, while seeing the friendliest conditions for women on the labour market in terms of equality in remuneration in Scandinavian countries, whereas the situation is the opposite in eastern and southern Europe. Landmesser (2019) verified the occurrence of the gender pay gap in all EU countries and states that gender differences in remuneration mostly occur in countries such as Poland, the Czech Republic, Slovakia, Bulgaria, Hungary, Croatia and others, while indicating that income conditions are more balanced on the labour market in western Europe.

The differences in men’s and women’s salaries have an effect on income equality, which is the basic characteristic of a healthy society and a necessary consequence of a functioning economic system (Moller et al., 2009). On the other hand, income inequality also brings negative consequences for society in the form of criminality, social unrest, etc. (Kujala et al., 2019). Greig et al. (2007) state that at a time of economic growth income inequality can increase, which has a negative effect on low-income households that often include single female parents. Income differences have a greater impact on low-income households (Landmesser, 2019). Income inequality also has psychological consequences, such as low or no happiness, poor morale or low social cohesion (Buttrick and Oishi, 2017). Roy and Chaudhuri (2008) draw attention to the position of women in society which is affected by income inequality. According to their study, women evaluate their health conditions worse and use medical services less because of their position.

Among the factors significantly affecting income inequality are demographic changes and changes in the structure of society (Chaupain-Guillot and Guillot, 2015). Demographic and socio-economic factors of the individual in particular determine the level of his or her income (Corsi et al., 2016; Aisa et al., 2019; Gradín et al., 2010). Corsi et al. (2016) in connection with remuneration, depending on gender, also talks of another factor, which is preference of job positions or employment industries for certain genders on which the wage level is based, i.e. various levels of remuneration in the technology sector in comparison with services for example. She adds that different wages of men and women do not mean gender discrimination, but other preferences of differently evaluated job positions. The choice of job sector and profession, i.e. various preferences of men and women, are the most important factor explaining the gender pay gap (Karamessini, Ioakimoglou, 2007).

In gender-based discrimination this concerns refusal of access to equal opportunity among the labour force. In gender discrimination this usually means that women have a lower income for work than men (Loseke, 2017). There are many causes of income influenced by gender. Navarro and Salverda (2019) examined differences in household income based on satisfaction with work activity. They show that men are more satisfied if they bring higher earnings into the household so they demand a job position with higher remuneration. This does not necessarily have to be about discrimination, but about prioritising different job positions at work. In contrast, women require more free time for family and the wage level for them is not the main thing. Duvivier and Narcy (2015) draw attention to the motherhood wage penalty due to child-related career interruptions. Women with children have worse access to management positions and adjusted working conditions.

Further possibilities of why income differences occur according to Navarro and Salverda (2019) is a situation when workers of different gender dispose of the same capital (experience, education, etc.), are assigned different work. Women are usually assigned work on a lower job grade, such as administrative work, whereas men with the same education can perform corporate management activities. Gradín et al. (2010) also draw attention to discrimination against women which is a determining factor for poverty, and add that the risk-of-poverty rate rises for individuals living in a household which depends primarily on women’s earnings. Fox et al. (2015) stress that at the present time single-person households predominate among households living with the risk of poverty, including households consisting of a single female parent and children. This does not concern mainly old-age pensioners, as was the case in the past. Likewise, Aisa et al. (2019) talks of the role of gender when measuring the poverty rate, but also sees other determinants of poverty in higher attained education and the work status of the persons in a household. Di Meglio et al. (2018) states that the occurrence of poverty is derived especially from a combination of specific categories of gender, age, economic activity and higher attained education. The effect of gender on the wage level is augmented by motherhood and child care.

Besides the effect of demographic factors on the level of income, there is the influence of factors which apply to actual employment. Xiao (2001) states that the growth of remuneration at work is also related to change of jobs during a career, adult education, technical level of employees and characteristics of job positions that an employer is able to offer.

The primary data source for this study is the EU-SILC (European Union – Statistics on Income and Living Conditions) survey, specifically of 2019 data. The EU-SILC survey provides detailed data on the income situation of households and individuals, making it possible to identify the demographic and socio-economic factors of households and the individual living in them. This extensive sample of microdata also contains data on the employment of individuals according to the industry of employer’s activity (job sector) and its size, on the type of employment contract, the job description of the individual, etc. Besides data on the income situation, the EU-SILC data offers information about the living conditions of households and a subjective evaluation of various spheres of life.

In 2019 the EU-SILC survey was conducted on a total of 260,876 European households and 518,415 individuals took part in it altogether. The EU-SILC survey is compulsory in all European Union states and takes place according to uniform methodology published by Eurostat (Eurostat, 2019). Eurostat also issues uniform methodology for work with these data. Economically active individuals were filtered out of the total number of collected data (521,391 individuals), who stated their economic status in the survey as being an “employee”. So altogether this involves working with the data taken from 178,878 respondents from all EU countries.

The basic variable, which is based on further calculations and analyses, is the employment income variable which includes the total annual gross employment income of the individual (i.e. annual wage or salary) expressed in Euros for all EU countries. EU-SILC data contain a conversion coefficient which is used as a weight in the conversion of data from a selected sample to the basic sample (i.e. the entire population of the country concerned). The conversion coefficient is taken into account in all conducted analyses.

The ISCO (International Standard Classification of Occupations), which is respected within the framework of the EU-SILC database, is used in order to determine the job sector and job position of individuals.

The dependence of two variables is assessed using contingency tables and tested by the chi-square test with the materiality level of α = 0.05. The chi-square tested null hypothesis means the independence of variables. If the resulting p-value is lower than the materiality level, the null hypothesis of independence is rejected (Hebák et al., 2015).

The factors affecting employment income and factors affecting gender differences in income are determined by applying the multiple linear regression analysis in which data are entered from respondents of the entire EU. The regression analysis is conducted using the Stepwise method, i.e. by the step-by-step addition of variables, which increase the quality of the model. The quality of the model is assessed by the determination coefficient. The one with the highest determination coefficient is selected as the resulting model. The materiality of the model is tested using the F test with a null hypothesis representing the model’s insignificance and materiality level of α = 0.05.

The multiple linear regression analysis is based on the following general model:

where x_i are independent explanatory variables, y is a dependent explained variable, β_i are unknown regression parameters, m represents their number and ε represents the model’s random component (Greene, 2018).

To analyse the gender differences in income, the data were complemented by the hypothetical variable y. The constructed dependent variable y “Differences in Income” is calculated as follows:

where x m ¯<!-- ¯ --> $\overline{x_{m}}$ is the weighted arithmetic average of men’s employment income and x_i is women’s employment income for i = 1,2,...,n.

Abbreviations of EU countries are used as follows: Austria (AT), Belgium (BE), Bulgaria (BG), Cyprus (CY), Czechia (CZ), Germany (DE), Denmark (DK), Estonia (EE), Greece (EL), Spain (ES), Finland (FI), France (FR), Croatia (HR), Hungary (HU), Ireland (IE), Italy (IT), Lithuania (LT), Luxembourg (LU), Latvia (LV), Malta (MT), Netherlands (NL), Poland (PL), Portugal (PT), Romania (RO), Sweden (SE), Slovenia (SI), Slovakia (SK), United Kingdom (UK). IBM SPSS Statistics 27 is used in this study for processing EU-SILC data.

A sample of 178,878 economically active individuals who have “Full-time Employee” economic status was filtered from the EU-SILC survey conducted in 2019 in all EU countries. The analyses performed of this sample showed that differences between men’s and women’s gross income exist in all EU countries, which corresponds to the conclusions of Landmesser (2019), but the generalisation was not confirmed that the highest differences apply above all to countries of the former “Eastern European bloc” countries. The assertion was not confirmed of Castellano and Rocca (2014) that the most equal incomes are in Scandinavian countries. The current EU-SILC data show that the lowest employment income differences between men and women are in Romania (8%) and Slovenia (10%). Apart from the low differences in Sweden (12%) there are low differences in Belgium (13%). In contrast, Finland is one of the countries with the highest gender differences in income. The highest values, up to 27%, are attained in the Czech Republic, 22% in Austria and Lithuania, in most countries the differences range between 10 to 20%.

A conclusive negative dependence was confirmed in all EU countries, which means that women are remunerated with a lower wage than men. Apart from gender, the factors that most affect the level of men’s and women’s income are “Age”, “Education”, “Employment sector” and “Management Position”. When analysing the gender differences in income, 73% is explained by verified independent variables. The identification of the strength of the effect of independent variables, including their direction, allowed the use of the Stepwise regression analysis. The strongest effect on the differences in income was found in the “Employment Absolute Income” when the rule is that with growing income the size of the differences increases. Another significant factor according to the strength of the effect is whether a woman changed jobs in the last year. If so, the differences become more profound. The statistically conclusive “Type of Employment Contract” variable shows that women with an employment contract for a limited period are the most likely to have lower remuneration for work and according to the “Number of Months in Unemployment” variable the time spent in unemployment increases the likelihood of a lower wage. The “Education” and “Age” factors have a strong but negative effect. The lower the women’s education is the greater the difference as opposed to men’s income. With increasing age the differences decrease and this can be caused by women’s parental leave after their career start. After returning to their profession women have to catch up with men in work experience as well as in income. The “Age” and “Education” factors and their interaction are evident with the distribution of income in individual deciles. Women predominate in the lowest income decile, but there is clearly a greater representation of men from the third decile. Special attention is paid to the “Job sector” factor which has one of the strongest effects on the actual employment income. A conclusive dependence of income from employment and job sector appears. Adam (2020) says the same of the influence of an individual’s wage depending on whether he or she is employed in the private sector or in the public sector. Likewise current EU-SILC data also showed that in the public sectors which include “Training and Education” for example, there are no great gender income differences as there are in “Services and Retail” which is part of the private sector. Incomes are lower in the “Services and Retail” sector, which is dominated by women, according to data from the entire EU. 39% of employees in services and retail fall into the first income quintile. Likewise, in this sector there is a great gender difference in income (45%).

We can state that there are gender differences from the results of the analyses and confirmed effect of the verified factors. It is also apparent from the results how many aspects influence the income earned by an employee and that the individual and his or her preferences occupy a significant place in determining the income situation and differences. The actual choice of field by the subject in which he wants to work, what is his or her attitude to life, feeling of satisfaction, luck, what values he or she prioritises and decisions about his or her profession. Such decisions have a continuation and are reflected in the length of education, field of education and those indicated employment industries for concluding employment relationships. This can already be observed in the results of the analyses which conform with the views of Navarro and Salverda (2019) regarding employment preferences in connection with life fulfilment. But the decisions of the subject are made under certain set conditions dependent above all on the parameters of social policy laid down above all by the Labour Code and legislative measures.

Social policy should create such conditions which also respect the different needs and interests of men and women. It should allow women who work and want to work during child care without being discriminated for child care and penalised for going on maternity which is what is also discussed by Duvivier and Narcy (2015). State social policy and the conditions it creates should allow women when fulfilling their maternal duties, not to have to interrupt their professional career but allow them to fulfil their private family objectives. The consequences of setting up parameters need to be considered, and the example used can be maternity and parental leave. The Czech Republic with 3–4 years of parental leave is one of the EU countries with the longest maternity leave and also has the greatest difference in income (27%). The three or four-year parental leave and time spent outside their profession means that women partly lose their knowledge and expertise whereas men can achieve professional growth and this is connected with income growth. Added to this the measures which are convenient for a certain type of individual, but the lack of the subsequent social policy measures such as the possibility of working part-time during child care, distribution of working hours in professions which allow work from home, creation of sufficient child-care facilities, benefit of the effect of the number of children on the calculation of pensions and so on. The presented results show that the social policy of the state concerned has a decisive effect on the position of women in society, i.e. on their remuneration as employees, as well as the verified effect of the analysed factors. In order to set up their parameters it would be expedient to continue with research at a national level.