EurWORK European Observatory of Working Life

Gender pay gap most acute among highly educated women

About

Country: 
Greece
Author: 
Sofia Lampousaki
Sector: 
Education
Institution: 
Labour Institute of Greek General Confederation of Labour (INE/GSEE)

In 2007, the Bank of Greece published a study examining the link between educational level and wage differentials between men and women in the labour market. The research calls attention to the fact that highly educated women who are top earners in Greece receive much lower pay than men with similar levels of education. Taking account of a variety of labour market characteristics, the study finds that a sizeable part of the gender pay gap remains unexplained.

About the study

In February 2007, the Bank of Greece (Τράπεζα της Ελλάδος) published a study on Education, labour market and wage differentials in Greece (in Greek, 2.7Mb PDF). The report examines the gender pay gap in Greece according to level of education and analyses whether the differences result from worker productivity or from factors that cannot be explained on the basis of productivity. Overall, the study tends to confirm the ‘glass ceiling’ theory (see Albrecht et al, 2003), which argues that wage differentials between men and women increase as workers move into the upper end of the wage distribution.

Methodology

The data used for the statistical analysis come from the 2004 Survey of Income and Living Conditions (EU-SILC) of the National Statistical Service of Greece (Εθνική Στατιστική Υπηρεσία Ελλάδας, ESYE), which refer to incomes in the year 2003.

A variant of the Oaxaca (1973) and Blinder (1973) decomposition technique was used in order to investigate whether gender wage differentials are due to gender disparities in labour market characteristics such as marital status, education, working hours, experience, economic sector, type of job, position of responsibility, occupation or company size. However, besides these variables explaining the gender pay gap, there is also an ‘unexplained part of the gender differential’ which many economists consider as the ‘discrimination factor’ or simply ‘discrimination’.

The category ‘workers with low level of education’ includes workers with preschool, primary, lower secondary and upper secondary education. The category ‘workers with high level of education’ includes workers with a level of education beyond secondary.

Wage differential

In examining the gender wage differentials across the whole pay distribution and on an average level (Figure 1), the study findings reveal that:

  • women’s earnings averaged about 84% of those of their male counterparts;
  • among the group of low-paid workers, the wage differential between men and women is slightly smaller than the average wage differential. For average incomes, the wage differential is much smaller than the average differential, while the gender earnings disparities are much larger than the average differential in the higher income brackets.

Gender wage differentials across whole wage distribution and average level, 2004 (%)

Gender wage differentials across whole wage distribution and average level, 2004 (%)

Source: Processing of EU-SILC data for 2004

Gender wage differentials across whole wage distribution and average level, 2004 (%)

Wage differentials and educational level

In examining gender wage differentials by educational level (Figure 2), the following results emerge:

  • up to the 60th percentile of the wage distribution, pay differences between men and women are greater among workers with a low level of education, compared with workers with a high level of education;
  • among higher educated workers, the gender wage differential increases towards the upper end of the wage distribution: women who are in the 90th percentile of the wage distribution are paid 67% of the pay of their male counterparts.

Gender wage differentials, by educational level, 2004 (%)

Gender wage differentials, by educational level, 2004 (%)

Source: Processing of EU-SILC data for 2004

Gender wage differentials, by educational level, 2004 (%)

Unexplained part of gender wage differentials

The study then examined whether the existing wage differentials result from worker productivity or from factors that cannot be explained on the basis of productivity – in other words, the unexplained part of the wage differential. The findings of the analyses were as follows:

  • for workers with low levels of education, the unexplained part of the gender wage differential for the whole sample is significantly high (77.4%). The share of the unexplained wage differential is particularly high in the low deciles of the wage distribution and decreases significantly in the higher deciles (Table 1);
  • for workers with a high level of education, the greater part of the wage differential (58.3%) can be explained by differences relating to worker productivity. Regarding this group of workers, it should be noted, however, that at the lower end of the wage distribution the wage differential between men and women is mainly the result of differences in worker productivity, whereas at the high end of the wage distribution the wage differential remains unexplained (Table 2).
Table 1: Unexplained part of wage differential between women and men with low level of education
  Total 10% 25% 50% 75% 90%
Wage differential 0.234 0.193 0.205 0.272 0.186 0.189
Unexplained difference as a percentage (%) of total pay differential 77.4 87.6 91.2 93.0 34.2 18.1

Notes: Wage differential figures are in logarithmic form. 10%=1st decile, 25%=1st quartile, 50%=2nd quartile, 75%=3rd quartile, 90%=9th decile.

Source: Processing of EU-SILC data for 2004

Table 2: Unexplained part of wage differential between men and women with high level of education
  Total 10% 25% 50% 75% 90%
Wage differential 0.187 0.138 0.098 0.087 0.241 0.406
Unexplained difference as a percentage (%) of total pay differential 41.7 - 42.9 62.1 67.0 98.9

Notes: Wage differential figures are in logarithmic form. 10%=1st decile, 25%=1st quartile, 50%=2nd quartile, 75%=3rd quartile, 90%=9th decile.

Source: Processing of EU-SILC data for 2004

Commentary

In Greece, where the proportion of women who are active in the labour market is relatively low, women who participate in the labour market generally have a high level of education. As a result, the overall wage differentials between men and women are relatively small. According to Eurostat data for 2004, the ratio of women’s and men’s average gross hourly pay to that of men in the whole economy stood at 90% in Greece and 85% in the EU25. However, it seems that highly educated women have restricted access to the upper levels of the wage distribution for reasons that cannot be explained on the basis of their productivity.

Female workers with low levels of education appear to accept jobs in the first years of their careers which are characteristic for this group of population, that is, jobs with a high probability of leaving the labour market. Nonetheless, as they remain in the job, their skills levels as well as their pay increase in line with the work that they are assigned; therefore, the gender pay differentials in this group of workers remain low. Overall, however, the existing wage differences between men and women in the Greek labour market are to a large extent the result of the ‘unexplained part of the wage differential’.

References

Albrecht, J., Bjorklund, A. and Vroman, S., ‘Is there a glass ceiling in Sweden?’, Journal of Labor Economics, Vol. 21, No. 1, 2003, pp. 145–177.

Blinder, A.S., ‘Wage discrimination: Reduced form and structural estimates’, Journal of Human Resources, Vol. 8, No. 4, 1973, pp. 436–455.

Oaxaca, R.L., ‘Male-female wage differentials in urban labour markets’, International Economic Review, Vol. 14, No. 3, 1973, pp. 693–709.

Papapetrou, P., ‘Education, labour market and wage differentials in Greece’, Economic Bulletin, No. 28, February 2007, available (in Greek) at: http://www.bankofgreece.gr/publications/pdf/oikodelt200702.pdf

Sofia Lampousaki, Labour Institute of Greek General Confederation of Labour (INE/GSEE)