Abstract:

In the last decades, the increase in female labour force participation has been joined by a reduction in gender wage gap almost in all European countries. Despite noticeable progress, the female employment rate is still consistently lower than the male one and gender inequalities in labour market, and especially in wages, remain significant. This gap is maintained by many forces and mechanisms, which include the choice of the field of study and the education, cultural and social attitudes and customs, economic determinants and discrimination aspects. These behaviors affect also the mechanisms connected with skill and proficiency acquisition, i.e. the way in which competences are acquired and spent in labour market and their degree of remuneration. This paper explores the determinants of gender gap in five European countries: France, Spain, Italy, Germany and United Kingdom, focusing on the role played by educational and skill mismatches. Data used in our analysis come from the 2012 edition of OECD Programme for the International Assessment of Adult Competencies (PIAAC) which provides a large set of variables about job characteristics and skills got by a employees, allowing to control for a greater number of factors related to wage. The analysis investigates whether and how educational and skill mismatches affect wages and gender differentials through over and under education (ORU) model and skill-mismatch measures.

The first step of our analytical approach is to estimate different specifications of mincerian wage equation, controlling for measures of educational and skills mismatch and several personal and job characteristics. Subsequently, a quantile regression model is applied as a complementary technique which allows to analyze the contribution of each variable to wages at different quantiles, in order to provide a more accurate analysis of mechanisms concerning wages and gender gap.

Lastly, with the aim to identify the extent to which the gap can be ascribed to the different personal characteristics of employees and how much of it depends by the different remuneration reserved to individual characteristics, the Oaxaca and Blinder and the Machado and Mata decomposition are applied. The main results underline different levels of gender wage gap across countries in terms of mean values, but also an increase of gap at higher percentiles of wage distribution, especially for the United Kingdom and Germany. The decomposition analysis highlights for Italy and Spain the lowest levels of gender wage gap, but the highest unexplained parts. Controlling for mismatch in education allows to significantly reduce gender wage gap in France, Germany and Spain, denoting a stronger incidence of over-education for female employees in these countries while mismatch in skills concerns especially men and is significant only in Spain and in the United Kingdom. Everywhere, but especially in Spain, the expansion of higher education occurred in the last decades generated a substantially growth of high-skilled workers who is associated a more slowly demand advance, contributing to increase inequalities in income and occupational mismatches.

Keywords:

Gender gap, quantile regression, OECD PIAAC.