Purpose: The purpose of this paper is to highlight the pivotal role of overeducation in explaining the unexplained part of the gender pay gap (GPG), i.e. the component usually attributed to gender discrimination in the Oaxaca-Blinder decomposition. Design/methodology/approach: The study uses a large Italian data set (ISFOL PLUS 2005–2014) to estimate the GPG among properly educated and overeducated workers. The model simultaneously accounts for both participation bias and endogeneity bias by applying an extension of the Heckman’s two-stage procedure. Findings: Estimates show that the GPG is significantly higher among overeducated than among properly educated workers because women’s unobservable characteristics driving female employment into overeducation also drag down female wages more than men’s unobservable characteristics drag down male wages. Correcting for the participation and overeducation decisions, the unexplained portion of the GPG disappears among overeducated workers, while it remains significant among properly educated individuals. Originality/value: The authors draw the conclusion that overeducation is, first, a first-best matching for individuals (both men and women) compensating with more education for their lower productive characteristics. Second, it may be a signaling device for women spending their useless-for-the-job diploma to inform employers on their valuable though unobservable productive characteristics and fight gender wage discrimination. The results favor education as a tool of counteracting gender discrimination. Hence, as females are less overeducated than males despite their larger representation in higher education, there should not be concern that expanding higher education will disadvantage females.

Overeducation and the gender pay gap in Italy

Töpfer M.
2018-01-01

Abstract

Purpose: The purpose of this paper is to highlight the pivotal role of overeducation in explaining the unexplained part of the gender pay gap (GPG), i.e. the component usually attributed to gender discrimination in the Oaxaca-Blinder decomposition. Design/methodology/approach: The study uses a large Italian data set (ISFOL PLUS 2005–2014) to estimate the GPG among properly educated and overeducated workers. The model simultaneously accounts for both participation bias and endogeneity bias by applying an extension of the Heckman’s two-stage procedure. Findings: Estimates show that the GPG is significantly higher among overeducated than among properly educated workers because women’s unobservable characteristics driving female employment into overeducation also drag down female wages more than men’s unobservable characteristics drag down male wages. Correcting for the participation and overeducation decisions, the unexplained portion of the GPG disappears among overeducated workers, while it remains significant among properly educated individuals. Originality/value: The authors draw the conclusion that overeducation is, first, a first-best matching for individuals (both men and women) compensating with more education for their lower productive characteristics. Second, it may be a signaling device for women spending their useless-for-the-job diploma to inform employers on their valuable though unobservable productive characteristics and fight gender wage discrimination. The results favor education as a tool of counteracting gender discrimination. Hence, as females are less overeducated than males despite their larger representation in higher education, there should not be concern that expanding higher education will disadvantage females.
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11567/1071806
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 15
  • ???jsp.display-item.citation.isi??? 13
social impact