Gender segregation and earnings differences in the Spanish labour market

ABSTRACT

This article analyses the effects of different types of gender segregation on the gender wage differential for the Spanish labour market. Matched employer–employee data from a sample of 226,535 workers are used. These workers are employed in 61 occupations within 26,492 establishments in 51 different industries. Workers belonging to the same industry, establishment or job share common factors which cannot be observed and these factors affect wages. If these unobservable variables are correlated with the explanatory variables, their estimated effects will be biased. For this reason, we estimate the effects of each type of gender segregation on the wage gap using a robust specification to these possible correlations. We obtain that industrial segregation by gender explains a lower part of the wage gap between men and women than previous researches found using standard regressions, while the contributions of establishment segregation and occupational segregation within each establishment are greater.

KEYWORDS:

• Gender wage difference
• industrial segregation
• establishment segregation
• occupational segregation within each establishment
• gender discrimination
• Spain

JEL CLASSIFICATION:

• J16
• J31
• J71

Acknowledgements

We thank Tomi Kyyrä for his useful suggestions. Any remaining errors are responsibility of the authors.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ In this study, we consider that individuals belonging to the same occupation with the same supervisory status within the same establishment hold the same job.

² Theoretical work experience is equal to the age of each worker minus his/her years of completed schooling minus 6.

³ If the mean difference between men and women for a variable or its estimated coefficient is not significantly different from zero at a 10% level, we assign zero to its relative impact.

⁴ Additionally, we repeat the estimations defining each job by all workers holding the same broad occupational category within the same establishment independently of their supervisory status. The results obtained from standard regressions show that 27.5% of the gender wage gap is explained by industrial segregation, 13.7% is explained by establishment segregation and 10.2% by occupational segregation within each establishment. Comparing the estimations from the FE model with those from the standard regressions, we observe the same pattern as that obtained including supervisory tasks in the definition of each job, except that the contribution of occupational segregation within each establishment is lower than that obtained by OLS. Additionally, the wage difference between men and women with the same observable characteristics and the same job obtained using the 9 broad occupational categories is much higher than that obtained using the 61 occupations. In particular, this wage gap is 9.6% in the OLS estimations, and 10.8% in the FE estimations for the widest occupational categories, while it is 7.8% in the OLS estimations and 8.28% in the FE estimations for the narrowest definitions of each job. Furthermore, we observe that the narrower the definition of each job, the higher the relative impact of occupational segregation within each establishment on the gender wage gap.