Gender-Based Occupational Choices and Family Responsibilities: The Gender Wage Gap in Italy

ABSTRACT

This paper explores the relationship between the gender division of labor, occupational choices, and the gender wage gap in Italy. In Italy, cultural factors and low availability of formal childcare services define gender roles that are generally based on the male breadwinner model, in which childcare is almost completely entrusted to women. The analysis is carried out through an extension of the Oaxaca–Blinder decomposition and is based on data from the 2007 Italian National Institute for Workers’ Professional Development (ISFOL). The results are consistent with gender discrimination on wages and suggest that women’s occupational paths are often an outcome of limited choices, and that women’s unpaid domestic work negatively interferes with the energy women can put into paid work. These findings support the need to ensure gender equality in and out of the labor market, especially through deep changes in Italian social norms and through the development of formal childcare.

KEYWORDS:

• Gender-based occupational choices
• family responsibilities
• gender wage gap

JEL Codes:

• J31
• J71

ACKNOWLEDGMENTS

We are very grateful to the Associate Editor and to three anonymous reviewers who carefully revised the paper, giving us some constructive criticisms and several useful suggestions. This article is partly based on a paper to which the authors contributed (Centra and Cutillo 2009). The differences are extensively discussed in the paper.

Notes

1 The figures in this paper are calculated through normalized weights, obtained by dividing the weight of each unit by the weights’ average. Accordingly, sample representativeness is respected both in descriptive statistics and in the models, without effects on the statistical significance of the estimates that would derive by using sample weights.

2 We identified these jobs through the Labor Force Survey of ISTAT. Sorensen (1989) utilizes the three-digit ISCO to identify typically female jobs. Since the ISFOL database has only thirteen occupational items, we blend this information with the economic sector (twenty-seven items). Indeed, industrial crowding as well as occupational crowding may contribute to earnings disparity.

3 Since female-dominated jobs are defined through the interaction between sector of activity and occupation, failing to control for differences in the distributions of these covariates could lead to misleading results in assessing the net impact of the female-dominated decision.

4 Standard errors are made robust to heteroskedasticity through the White correction.

5 Women in female-dominated jobs can probably better adapt their work schedule through maternity leave, with negative effects on wages.

6 We also calculate the “intra-gender discrimination” through comparing women engaged in different types of jobs, and men engaged in different types of jobs. If our interpretation holds, we should expect discrimination against women in female-dominated jobs and against men in other jobs. The coefficients effect is −19.5 percent among women (at a disadvantage of women engaged in female-dominated jobs) and +15.7 percent among men (at a disadvantage of men engaged in other jobs), confirming our interpretation.

Additional information

Notes on contributors

Andrea Cutillo

Andrea Cutillo is Researcher at the Italian National Institute of Statistics (ISTAT) and responsible for the Household Budget Survey, and Adjunct Professor of Social Statistics at the European University of Rome. His main research interests are in economic inequality, labor market, discrimination, and the economics of education.

Marco Centra

Marco Centra is Senior Researcher at the Italian National Institute for Workers’ Professional Development (ISFOL) and was in charge of the Labor Market Service and of the Statistical Office. His main research interests are in labor market, gender inequality, gender discrimination, and statistical methodologies.