Gender pay gap wider for better educated women

The new Isfol survey ‘Plus – Participation, labour, unemployment survey’ is carried out on an annual basis in order to follow developments in the Italian labour market. According to the 2005 report, the gender pay gap is wider for better educated women and increases with length of service. Personal characteristics explain just one third of the pay gap, with the remainder being attributed to discrimination.

The 2005 report Plus – Participation, labour, unemployment survey (in Italian) was published by the National training agency (Istituto per la Formazione dei Lavoratori, Isfol). The report summarises the main results of the first comprehensive survey aimed at monitoring the most significant changes in the labour market, and thus at evaluating the impact of developments such as the 2003 labour market reform.

Study method

The survey is carried out every year and is based on a multi-stage sample of over 40,000 interviews, using computer-assisted telephone interviewing (CATI) methodology (see below for further information on the methodology). The survey was conducted for the first time in 2005.

The present analysis is restricted to a sub-sample of about 8,000 employees working under a full-time permanent contract. This sample was chosen because part-time work is strongly gender biased: only 3% of men work part time, compared with 25% of women. Furthermore, the proportion of non-response among self-employed people was very high, at over 60%.

Study findings

Influence of education

According to the 2005 survey, women’s wages are in general notably lower than those of men. Moreover, the gender pay gap widens as educational attainment increases: women earn 81% of men’s wages at compulsory educational level, that is lower secondary school, while women’s pay decreases to 74% of men’s when they both hold a degree; the gender pay gap declines even further, to 65%, at post-graduate level (Table 1).

These ratios vary slightly in the sample of respondents without children. The pay gap is narrower at compulsory educational level among people without children than for the total sample, at 83% and 81% respectively. Overall, women with an upper secondary school educational level are paid 82% of men’s salary; among women without children, this gap is slightly higher, as they reach just 81% of men’s pay. Among those with a third level degree, women in general earn 74% of men’s wages; while those without children earn less, at 71% of men’s wages. However, for post-graduate educational attainment, women without children earn 72% of men’s wages, compared with just 65% for women with children.

The findings show that women have a higher level of educational attainment than men: in general, the proportion of men reaching compulsory educational level only is almost twice as high as the proportion among women, at 22.5% and 11.6% respectively (Table 1). Moreover, the proportion of women having a degree is significantly higher, at 31.3%, compared with 19.4% for men. Educational attainment is higher among workers of both sexes who do not have children, but the gender pay gap among those holding a degree is relatively stable.

These figures suggest that men have fewer difficulties in finding a job, while women have to invest more in education in order to increase their chances of entering the labour market. However, women face lower career prospects than men do, especially those who are better educated.

Table 1: Gender pay gap, by education and children
Gender pay gap, by education and children
  Entire sample Sample without children
  Wage ratio Men (%) Women (%) Wage ratio Men (%) Women (%)
Primary school 0.719 2.2 1.9 0.748 3.6 2.6
Lower secondary school 0.808 22.5 11.6 0.829 18.2 11.4
Upper secondary school 0.818 54.2 53.3 0.809 50.0 46.5
Degree 0.739 19.4 31.3 0.708 25.4 37.1
Post-degree 0.654 1.7 1.8 0.723 2.8 2.4

Source: Isfol Plus, 2005

Length of service

Length of service is an important factor in investigating the gender wage gap, since it gives an insight into companies’ personnel policies. The gap is stable for workers with a low educational level, at a wage ratio of about 0.80. The pay gap is relatively negligible when the tenure is low, but rapidly increases both with length of service and education, especially among the higher educated workforce – widening in this group from a gap of just two percentage points to as much as 28 percentage points (Table 2). The glass ceiling effect is thus evident for women with a high education level and a substantial length of service: companies offer good career opportunities to men, while such opportunities are less available to women.

Table 2: Gender wage ratio, by education and tenure
Gender wage ratio, by education and tenure
  Low education level Medium education level High education level
Less than 4 years’ service 0.80 0.97 0.98
4–10 years’ service 0.82 0.86 0.81
More than 10 years’ service 0.79 0.80 0.72

Note: Calculation on the average value.

Source: Isfol Plus, 2005

Factors causing gender pay gap

The gender gap is further investigated by analysing the characteristics that affect wage levels and the residual which identifies the actual discrimination, according to the so-called Oaxaca-Blinder distribution decomposition procedure. Overall, only 9.2 percentage points of the 26.86% pay gap, or approximately one third, can be explained by personal characteristics: having worked part time explains about seven percentage points of the pay gap; experience explains less than three percentage points; the particular sector accounts for less than two percentage points; while education and competence levels negatively explain about two percentage points of the gender pay gap (Table 3).

Thus, the contributing factors total less than 10 percentage points overall. The remaining two thirds of the almost 27% can be attributed to pure discrimination. Such figures show that the excuse of varying human capital cannot hold.

Table 3: Factors explaining gender pay gap
Factors explaining gender pay gap
  Characteristics Total
Work experience 0.0288 0.0982
Education -0.0197 -0.0412
Competence in English -0.0009 -0.0106
Computing competencies 0.0012 0.0058
Training -0.0008 -0.0006
Years lost at school -0.0024 0.0004
Education and competencies (overall) -0.0225 -0.0462
Geographic area -0.0054 -0.0069
Company size 0.0020 0.0007
Public/private organisation 0.0015 0.0043
Having children 0.0004 0.0014
Married -0.0006 0.0036
Employment status 0.0044 -0.0017
Part-time work 0.0699 0.0884
Sector 0.0188 -0.0014
Professional status -0.0054 -0.0208
Total 0.0920 0.2686

Source: Isfol Plus, 2005


Although women’s participation in the labour market is a relatively recent circumstance (IT0502NU02), their role in the workplace reveals significant segregation factors, as reflected in the gender pay gap. All relevant studies performed in the last 10 years pinpoint this problem, but few have broken down the pay gap by personal characteristics and unexplained factors, in other words discrimination (for an overview, see the report Education and wage differentials by gender in Italy, 2006 (300Kb PDF)). Furthermore, the extent of the gender pay gap is sensitive to the type of data, different points over time and estimation techniques.

Figures from the Isfol Plus survey are consistent with recent findings from other sources, in particular those of the report Women’s numbers (in Italian; 843Kb PDF), based on microdata from the ‘Survey on household income and wealth’ (SHIW) by Banca d’Italia. The Isfol data show an increasing pay gap in comparison with data from the 1990s cited in the EIRO report Gender pay equity in Europe (TN0201101S).

Survey methodology

The questionnaire comprises three general modules: personal and family information, welfare services and training. It also includes specific sub-modules according to: employment status, including being inactive and unemployed; and personal characteristics, such as nationality, sex or age. A further module exists on working conditions for all people employed, including employees and those who are self-employed.

The general design of the questionnaire is therefore more oriented towards monitoring individual behaviour in the labour market, using as a control some key indicators on working conditions, such as working time, remuneration, job satisfaction and the recruitment process.

The sub-section on pay is the first general information source on remuneration in the labour market; other sources are either too broad, without being broken down into smaller categories – such as data from the Ministry of Economy and Finance (Ministero dell’Economia e delle Finanze) – or investigate families as a micro unit, namely, the SHIW. Recently, the National Social Security Institute (Istituto Nazionale Previdenza Sociale, INPS) prepared two longitudinal panels: Clap (Campione longitudinale degli attivi e dei pensionati), a longitudinal sample of employed and retired people, and WHIP (Work Histories Italian Panel; in Italian); however, they both include microdata on private sector employees only.

Further information

The gender pay gap: background paper explores different facets of the gender pay gap issue, based on data collected from two of the Foundation’s projects: the EIRO report Pay developments – 2005 (TN0606101U), and the fourth European Working Conditions Survey, carried out in 2005. See also other Foundation publications on the subject of gender.

Mario Giaccone, Fondazione Regionale Pietro Seveso

Useful? Interesting? Tell us what you think. Hide comments

Eurofound welcomes feedback and updates on this regulation

Add new comment