The Gender Wage Gap in Croatia – Estimating the Impact of Differing Rewards by Means of Counterfactual Distributions

Danijel Nestić The Gender Wage Gap in Croatia – Estimating the Impact of Differing Rewards by Means of Counterfactual Distributions Croatian Economic Survey : Vol. 12 : No. 1 : April 2010 : pp. 83-119 The Gender Wage Gap in Croatia – Estimating the Impact of Differing Rewards by Means of Counterfactual Distributions Danijel Nestić The Institute of Economics, Zagreb, Croatia CroEconSur Vol. 12 No. 1 April 2010 pp. 83-119 Received: December 12, 2009 Accepted: March 23, 2010 Original Scientific Paper Abstract The aim of this paper is to estimate the size of, changes in, and main factors contributing to gender-based wage differentials in Croatia. It utilizes microdata from the Labor Force Surveys of 1998 and 2008 and applies both OLS and quantile regression techniques to assess the gender wage gap across the wage distribution. The average unadjusted gender wage gap is found to be relatively low and declining. This paper argues that employed women in Croatia possess higher-quality labor market characteristics than men, especially in terms of education, but receive much lower rewards for these characteristics. The Machado-Mata decomposition technique is used to estimate the gender wage gap as the sole effect of differing rewards. The results suggest that due to differing rewards the gap exceeds 20 percent on average - twice the size of the unadjusted gap - and that it increased somewhat between 83 Danijel Nestić The Gender Wage Gap in Croatia – Estimating the Impact of Differing Rewards by Means of Counterfactual Distributions Croatian Economic Survey : Vol. 12 : No. 1 : April 2010 : pp. 83-119 1998 and 2008. The gap is found to be the highest at the lower-to-middle part of the wage distribution. Keywords: gender wage gap, quantile regression, Machado-Mata decomposition, Croatia JEL classification: J16, J31, J70 1 Introduction1 One of the most challenging fields for attaining the equal treatment of women and men is the labor market. In practically every country, women earn less than men. To explain this disparity, labor economists have usually looked at differences in human capital characteristics, such as education and experience. Since women in many countries have poorer education and less labor market experience than men, productivity gains due to these attributes could explain part of the gender wage gap. The role of job- and firm-specific factors has also been studied. However, the wage gap remains even after accounting for differences in all observed characteristics. In former socialist countries, gender equality was a highly proclaimed policy goal during the communist regime and evidence shows that the difference in wages between women and men was rather low at that time (Brainerd, 2000). An egalitarian wage structure was a feature not only of centrally planned systems, but also of the self-management system in the former Yugoslavia (Orazem and Vodopivec, 1995). During the transition period, huge changes in the structure of the overall economy induced changes in the wage structure. Wage setting mechanisms were liberalized, which mostly produced higher wage inequality. However, it seems that these changes did not contribute to the widening of the gender wage gap in Central and Eastern European (CEE) countries in the first phase of transition (Brainerd, 2000; 1 84 Th is paper is a result of the research supported by a grant from the CERGE-EI Foundation under a program of the Global Development Network. All opinions expressed are those of the author and have not been endorsed by CERGE-EI or the GDN. The author would like to thank two anonymous referees for useful comments and suggestions. The author also thanks Hiau Joo Kee for providing his Stata code for the Machado-Mata wage decomposition. The usual disclaimer applies. Danijel Nestić The Gender Wage Gap in Croatia – Estimating the Impact of Differing Rewards by Means of Counterfactual Distributions Croatian Economic Survey : Vol. 12 : No. 1 : April 2010 : pp. 83-119 Newell and Reilly, 2001). Instead, there was a significant decline in the female participation rate. In countries of the former Soviet Union, however, female relative wages were reduced strongly, while female participation was kept high due to the continued labor hoarding practice (Pastore and Verashchagina, 2007). In the later stages of transition, the gender wage gap increased, although it remained rather modest by international standards (Rutkowski, 2001). The literature suggests that the relatively small gender wage gap in most transition economies is connected with the higher human capital endowments of women compared to men, such as education and experience (Paci and Reilly, 2004). Croatia, as a post-socialist country, shares many labor market characteristics with other CEE countries. Therefore, it is not surprising that the unadjusted wage gap is relatively low. The average monthly gross wage of women was around 11 percent below that of men in 2008, making it one of the lowest unadjusted gaps in Europe - even one of the lowest among the former socialist countries.2 This gap, measured by gender differences in the average wage, masks an even higher underlining wage disadvantage for women. Bisogno (2000) took into account gender differences in education, experience, and workplace-related factors in a regression analysis framework and found the adjusted gender wage gap to be much higher, 20 percent in 1998. In a similar setting, Nestić (2005) reported a mean gap of around 15 percent in 2003. This documented difference between the unadjusted and the adjusted gap is interesting for further exploration aimed to shed more light on its sources in the context of the transition in Croatia. This study, therefore, considers changes in the gender wage gap in Croatia between 1998 and 2008, first in its unadjusted form and then by adjusting the gap in order to take into account differences in the labor market characteristics of men and women. Compared to previous studies for Croatia, the current study presents two novelties. First, it goes further in exploring the gender wage gap by using the quantile regression technique and estimating the wage gap at various points of 2 For 2007, Eurostat reports an unadjusted gender wage gap in Poland and Slovenia of 8 percent, Bulgaria 12 percent, Romania 13 percent, and Hungary 16 percent. The average for the “old” EU members (EU-15) was 18 percent; while Baltic States, the Czech Republic, and Slovakia had relatively high gender pay gaps. 85 Danijel Nestić The Gender Wage Gap in Croatia – Estimating the Impact of Differing Rewards by Means of Counterfactual Distributions Croatian Economic Survey : Vol. 12 : No. 1 : April 2010 : pp. 83-119 the distribution, not only at the mean. The gender wage gap often varies across the distribution; for example, it is highest among the high-paid workers. Second, this study focuses on the part of the gap that is caused by differences in rewards for the same observable labor market character (...truncated)