Three essays on gender wage differentials.
Date of Issue2011
School of Humanities and Social Sciences
Gender wage gaps in favor of men have been widely documented across a large number of countries. Such gaps can be partially explained by differences in productivity related characteristics, such as education and experience between men and women; the part not explained by differences in characteristics is often referred to as potential discrimination in the labor market. The issue has been studied extensively in the literature, especially at the mean; however, distributional patterns may be of more interest as they relate to policy-making, when targeted intervention is deemed appropriate. This thesis consists of three self-contained essays which study the gender wage differentials and potential gender discrimination across the earnings distribution. Within the growing literature on gender wage differentials over the wage distribution, two terms - glass ceilings and sticky floors - are used. Glass ceilings refer to a phenomenon where the counterfactual gender wage gap is wider at the top than at the bottom, while sticky floors refer to a wider gap observed at the bottom relative to the top. The objectives of the first chapter of the thesis are to expand the body of evidence, providing an international update from more than 50 countries in 5 continents and to suggest potential explanations to the stylized facts documented across countries. We find that glass ceilings are present in 22 European countries, while no country in Asia but Kazakhstan and Uzbekistan show evidence of glass ceilings. In 6 Latin American countries, the evidence is mixed. In Africa the estimates from the black populations of South Africa and Lesotho point to a clear sticky floor effect, while glass ceilings are found in Egypt and Morocco. Possible explanations could be differences in labor market institutions such as regulation, presence of minimum wage legislation, availability of childcare and job segregation, as well as societal, cultural practices. To understand the evolution of gender wage gaps at different points of the distribution and the contribution of individual determinants to the gap, Chapter 2 adopts unconditional quantile regression combined with Oaxaca-Blinder decomposition to study the gender wage inequality in Thailand from 1991 to 2007.